PREFACE

I have little to add to the Preface that I wrote for Volume I of this
commentary. What I wrote then, especially by way of thanks to a
number of people who in various ways have helped me, is still true.
Some things are perhaps better arranged in this volume than in the
first; one learns from one’s mistakes. This volume contains a more or
less conventional ‘Introduction’. It will, I think, be clear why I could
not write it until I had cleared my mind on a number of questions that
arise in Acts 15-28.
Since Volume I was completed, the 27th edition of Nestle-Aland’s
Novum Testamentum Graece has appeared. After some thought I
have decided in favour of consistency, and have retained the text and
apparatus of NA26. Some day, perhaps, the commentary may be
revised and the sigla of the later edition used throughout.

C. K. BARRETT
 ADDITIONS TO BIBLIOGRAPHIES AND ABBREVIATIONS

Commentaries referred to by the Author's Name only

Johnson, L. T. The Acts of the Apostles, Sacra Pagina 5, Collegeville,
Minnesota, 1992.
Overbeck, F. C. Kurze Erklärung der Apostelgeschichte von Dr. W. M. L. de
Wette, vierte Auflage bearbeitet und stark erweitert von F. Overbeck,
Leipzig, 1870.
Taylor, J. Les Actes des deux Apôtres. N Commentaire Historique (Act.
9.1—18.22), Études Bibliques, Nouvelle Série 23, Paris, 1992; VI (Act.
18.23—28.31), Études Bibliques, Nouvelle Série 30, Paris, 1996.
Wikenhauser, A. Die Apostelgeschichte, Regensburger NT, 4th edn 1961.

Monographs on (Luke-)Acts
Boismard and Lamouille: M. É. Boismard and A. Lamouille, Les Actes des
deux Apôtres, I—III, Études Bibliques, Nouvelle Série 12, 13, 14, Paris,
1990.
Book ofActs 1,2,3: The Book of Acts in its First Century Setting, ed. B. W.
Winter.
I Ancient Literary Setting, ed. B. W. Winter and A. D. Clarke, Grand
Rapids/Carlisle, 1993.
II Graeco-Roman Setting, ed. D. W. J. Gill and C. Gempf, Grand Rapids/
Carlisle, 1994.
III Paul in Roman Custody, by B. Rapske, Grand Rapids/Carlisle, 1994.
Cadbury, History: H. J. Cadbury, The Book of Acts in History, London,
1955.
Cadbury, Style: H. J. Cadbury, The Style and Literary Method of Luke,
Cambridge, Mass., 1920.
Cassidy and Sharper, Political Issues: R. J. Cassidy and P. J. Sharper (eds.),
Political Issues in Luke-Acts, Maryknoll, 1983.
Ehrhardt, Acts: A. Ehrhardt, The Acts of the Apostles, Manchester, 1969.
Filson: F. V. Filson, Three Crucial Decades, London, 1964.
Parsons and Pervo: Μ. C. Parsons and R. I. Pervo, Rethinking the Unity of
Luke and Acts, Minneapolis, 1993.
Ramsay, Luke: W. Μ. Ramsay, Luke the Physician and other Studies in the
History of Religion, London, 1908.
Tajra: H. W. Tajra, The Trial of St. Paul. A Juridical Exegesis of the Second
Half of Acts, Tübingen, 1989.
Talbert: C. H. Talbert (ed.), Perspectives on Luke-Acts, Edinburgh, 1978.
Thornton: C.-J. Thornton, Der Zeuge des Zeugen. Lukas als Historiker der
Paulusreisen, Tübingen, 1991.
van Unnik: W. C. van Unnik, Tarsus of Jeruzalem, De Stad van Paulus'
Jeugd. Mededelingen der koninklijk Nederlandse Akademie van Weten-
schappen, Afd. Letterkunde, Nieuwe Reeks, 15.5, Amsterdam, 1952.
xi
xii COMMENTARY ON ACTS

Other works referred to by abbreviations

Bacher, Terminologie: W. Bacher, Die exegetische Terminologie der jüdi-
schen Traditionsliteratur, Leipzig, 1899,1905; reprint Hildesheim, 1965.
Bammel-Moule: E. Bammel and C. F. D. Moule (eds.), Jesus and the Politics
ofHis Day, Cambridge, 1984.
Bauer, Rechtgläubigkeit: W. Bauer, Rechtgläubigkeit und Ketzerei im ältes-
ten Christentum, BhTh 10, Tübingen, 1934.
Benoit 1,2,3,4: P. Benoit, Exegèse et Théologie, Vol. I, Paris, 1961; Vol. II,
1961; Vol. III, 1968; Vol. IV, 1982.
Bornkamm 1,2,3,4: G. Bornkamm, Gesammelte Aufsätze, Band I, München,
1966; Band II, 1970; Band ΙII, 1968; Band IV, 1971.
Bousset, RJ: W. Bousset, Die Religion des Judentums, HNT 21,3rd edn ed. H.
Gressmann, Tübingen, 1926.
Brandon, Fall: S. G. F. Brandon, The Fall of Jerusalem and the Christian
Church, London, 1951.
Calvin, Institutes: J. Calvin, Institutes of the Christian Religion, tr. H. Bever-
idge, Vol. I, Edinburgh, 1845; Vol. II, 1845; Vol. III, 1846.
von Campenhausen, Amt: H. von Campenhausen, Kirchliches Amt und geist-
liche Vollmacht in den ersten drei Jahrhunderten, Tübingen, 1953.
Cerfaux, Ree. 2: Recueil L. Cerfaux, Vol. II, Gembloux, 1954.
Clarke, NT Problems: W. K. L. Clarke, New Testament Problems, London,
1929.
Cranfìeld, Romans: C. E. B. Cranfìeld, The Epistle to the Romans, ICC,
Edinburgh, Vol. I,1975; Vol. II, 1979.
Danby: H. Danby, The Mishnah, London, 1933.
Dodd, AS: C. H. Dodd, According to the Scriptures, London, 1952.
Dunn, Romans: J. D. G. Dunn, Romans 1-8, Word Biblical Commentary
38A, Dallas, 1988; Romans 9-16, WBC 38B, 1988.
Freedom and Obligation: C. K. Barrett, Freedom and Obligation, A Study of
the Epistle to the Galatians, London, 1985.
Gerhardsson, Memory and Manuscript: B. Gerhardsson, Memory and Manu-
script, Acta Seminarii Neotestamentici Upsaliensis 22, Uppsala, 1961.
Goulder, Two Missions: Μ. Goulder, A Tale of Two Missions, London, 1994.
luster: J. luster, Les Juifs dans l'Empire romain, 2 vols., Paris, 1914.
Käsemann, EVB 1,2: E. Käsemann, Exegetische Versuche und Besinnungen,
1. Band, Göttingen 1960; 2. Band 1964.
Kümmel, Einleitung: W. G. Kümmel, Einleitung in das Neue Testament, 19th
edn, Heidelberg, 1978.
Lightfoot, Ignatius: J. B. Lightfoot, The Apostolic Fathers, Part II, S. Igna-
tius, S. Polycarp. 3 vols., London, 1885.
MPG: J. P. Migne, Patrologia Graeca.
Mommsen, Provinces: Th. Mommsen, The Provinces of the Roman Empire,
ET, 2 vols., London, 1886.
Nock, Essays: A. D. Nock, Essays on Religion and the Ancient World, ed.
Zeph Stewart, 2 vols., Oxford, 1972.
Partsch, Kephallenia: J. Partsch, Kephallenia und Ithaka, in Ergänzungsheft
98 zu Petermanns Mitteilungen, Gotha, 1890.
Pastorals: C. K. Barrett, The Pastoral Epistles, NCB, Oxford, 1963.
Pratscher, Jakobus: W. Pratscher, Der Herrenbruder Jakobus und die Jako-
bustradition, FRLANT 139, Göttingen, 1987.
 ADDITIONS TO BIBLIOGRAPHIES AND ABBREVIATIONS xüi

Safrai-Stem: S. Safrai and Μ. Stern (eds.), The Jewish People in the First
Century, Vol. I, Assen, 1974.
Sherwin-White: A. N. Sherwin-White, Roman Society and Roman Law in the
New Testament, Oxford, 1963.
Stendahl, Scrolls: K. Stendahl (ed.), The Scrolls and the New Testament,
London, 1958.
Trebilco: P. R. Trebilco, Jewish Communities in Asia Minor, SNTSMS 69,
Cambridge, 1991.
Turner, Insights: N. Turner, Grammatical Insights into the New Testament,
Edinburgh, 1965.

Festschrifts (FS) with abbreviated titles

Bammel: Templum Amicitiae, Essays on the Second Temple presented to E.
Bammel. Ed. W. Horbury. JSNTSS 48. Sheffield, 1991.
Barrett: Paul and Paulinism. Essays in honour of C. K. Barrett. Ed. Μ. D.
Hooker and S. G. Wilson. London, 1982.
Beasley-Murray: Eschatology and the New Testament. Essays in Honor of
G. R. Beasley-Murray. Ed. W. H. Gloer. Peabody, Mass., 1988.
Black (1969): Neotestamentica et Semitica. Studies in Honour of Μ. Black.
Ed. E. E. Ellis and Μ. Wilcox. Edinburgh, 1969.
Bruce (1980): Pauline Studies. Essays presented to F. F. Bruce on his 70th
birthday. Ed. D. A. Hagner and Μ. J. Harris. Exeter, 1980.
Bultmann (1964): Zeit und Geschichte. Dankesgabe an R. Bultmann zum 80.
Geburtstag. Ed. E. Dinkier. Tübingen, 1964.
Corsani: Protestantesimo 48.3 (1994). I metodi dell’indagine sul Nuevo
Testamento. Omaggio per B. Corsani. Ed. S. Rostagno. Rome, 1994.
Farmer: Jesus, the Gospels, and the Church. Ed. E. P. Sanders. Macon,
1987.
Furnish: Theology and Ethics in Paul and his Interpreters. Essays in honor of
V. P. Furnish. Ed. E. H. Lovering, Jr, and J. L. Samney. Nashville, 1996.
Goulder: Crossing the Boundaries. Essays in Biblical Interpretation in Hon-
our of Μ. D. Goulder. Ed. S. E. Porter, P. Joyce, and D. E. Orton. Leiden,
1994.
Hengel: Geschichte-Tradition-Reflexion. Festschrift fur Martin Hengel zum
70, Geburtstag. Ed. H. Cancik, H. Lichtenberger, P. Schäfer. Tubingen,
1996.
Hofius: Christus als die Mitte der Schrift. [Festschrift für] O. Hofius. Ed. H.
Lichtenberger, H.-J. Eckstein, and C. Landmesser. [Tübingen], 1997.
Jaeger: Unio Christianorum. Festschrift für Erzbischof L. Jaeger, Paderborn.
Jeremias: Judentum- Urchristentum- Kirche. Festschrift für J. Jeremias. Ed.
W. Ehester. BZNW 26. Berlin, 1960.
Jervell. Mighty Minorities? Essays in honour of J. Jervell on his 70th birthday.
Ed. D. Hellholm, H. Moxnes, and T. K. Seim. Oslo/Copenhagen/Stock-
holm/Boston, 1995.
Kee: The Social World of Formative Christianity and Judaism. Essays in
Tribute to H. C. Kee. Ed. J. Neusner, E. S. Frerichs, P. Borgen, and R.
Horsley. Philadelphia, 1989.
Kuhn: Tradition und Glaube. Das frühe Christentum in seiner Umwelt.
Festgabe für K. G. Kuhn zum 65. Geburtstag. Ed. G. Jeremias, H.-W.
Kuhn, H. Stegemann. Göttingen, 1971.
xiv COMMENTARY ON ACTS

Mitton: ExpT78.1 (1976). In honour of C. L. Mitton. Edinburgh, 1976.

Mowinckel: Interpretationes ad Vetus Testamentum pertinentes S.
Mowinckel Septuagenario missae. Norsk teologisk tidsskrift 56. Oslo,
1955.
Neirynck: The Four Gospels 1992. Festschrift F. Neirynck. 3 vols. Ed. F. Van
Segbroeck, C. Μ. Tuckett, G. Van Belle, and J. Verheyden. BEThL 100.
Leuven, 1992.
Osborn: ABR 35 (1987. Special Issue in Honour of E. Osborn. Ed. I. Breward.
Melbourne, 1987.
Ramsay: Anatolian Studies presented to Sir W. Μ. Ramsay. Manchester,
1923.
Schnackenburg: Neues Testament und Kirche. Für R. Schnackenburg. Ed. J.
Gnilka. Freiburg/Basel/Wien, 1974.
Schneider: Der Treue Gottestrauen. Ed. C. Bussmann and W. Radl. Freiburg/
Basel/Wien, 1991.
Schweizer: Die Mitte des Neuen Testaments. Einheit und Vielfalt Neutesta-
mentlicher Theologie. Festschrift für E. Schweizer zum 70. Geburtstag.
Ed. U. Luz and H. Weder. Göttingen, 1983.
Smit Sibinga: NovT 38.2 (1996). Special number in honour of J. Smit Sibinga.
Ed. P. Borgen and others. Leiden, 1996.
Zimmermann: Begegnung mit dem Wort Festschrift H. Zimmermann. Ed. J.
Zmijewski and E. Nellessen. Bonn, 1979.
de Zwaan: Studia Paulina in honorem J. de Zwaan Septuagenarii. Ed. J. N.
Sevenster and W. C. van Unnik. Haarlem, 1953.

Works of reference
Altaner: B. Altaner, A. Stuiber, Patrologie. 7th edn Freiburg/Basel/Wien,
1966.
A Dictionary of the Bible, ed. J. Hastings, Edinburgh, 1906.
Digest. Digesta lustiniani, ed. Th. Mommson, Leipzig, 1866,1870.
Hort: F. J. A. Hort, The Apocalypse of St John I-III, London, 1908.
KP: K. Ziegler and W. Sontheimer (eds.), Der kleine Pauly 5 vols., Stuttgart
1964ff.
LCL: Loeb Classical Library.
Lobeck: C. A. Lobeck, Phrynichi Eclogae, Leipzig, 1820 (repr. 1965).
LS Supp: LS, Revised Supplement, ed. P. G. W. Glare and A. H. Thompson,
Oxford, 1996.
Μ. 4: A Grammar ofNew Testament Greek, by J. H. Moulton, Vol. IV, Style,
by N. Turner, Edinburgh, 1976.
ND 6, 7: New Documents illustrating Early Christianity, Vol. VI and Vol.
VII, by S. R. Llewelyn and R. A. Kearsley, Macquarie University, 1992,
1994.
Quasten: J. Quasten, Patrology, 4 vols, I Utrecht, Brussels 1950; II Utrecht,
Antwerp 1953; III, IV, Westminster, Maryland, 1986.
RGG: Die Religion in Geschichte und Gegenwart, ed. K. Galling, 3rd edn, 6
vols., Tübingen, 1957ff.
Robertson: A. T. Robertson, A Grammar of the Greek New Testament in the
Light of Historical Research, 3rd edn, London, 1919.
Rutherford, Gram.: W. G. Rutherford, First Greek Grammar, London, 1907.
 ADDITIONS TO BIBLIOGRAPHIES AND ABBREVIATIONS XV

SOE: The Shorter Oxford English Dictionary, ed. C. T. Onions, Oxford,

31955.
Simonson: G. Simonson, A Greek Grammar: Syntax, London and New York,
1911.
Swete: H. B. Swete, The Apocalypse of St John, London, 1906.
TWAT. Theologisches Wörterbuch zum Alten Testament, ed. G. J. Botterweck
and H. Ringgren, Stuttgart, Berlin, Köln, Mainz, 1970-.
Winer-Moulton: A Treatise on the Grammar of New Testament Greek, by
G. B. Winer, tr. and ed. by W. F. Moulton, 3rd edn, Edinburgh, 1882.
Winer-Schmiedel: G. B. Winer, Grammatik des neutestamentlichen Sprachi-
dioms, 8th edn, ed. by P. W. Schmiedel, 1. Teil, Göttingen, 1894.

Periodicals and Series

AbThANT: Abhandlungen zur Theologie des Alten und Neuen Testaments
AnBib: Analecta Biblica
ANRW: Aufstieg und Niedergang der römischen Welt
AThR: Anglican Theological Review
BBB: Bonner biblische Beiträge
BHTh: Beiträge zur historischen Theologie
BiK: Bibel und Kirche
BZ: Biblische Zeitschrift
BZNW: Beihefte zur ZNW
Coniect. Neot.: Coniectanea Neotestamentica
FRLANT: Forschungen zur Literatur des Alten und Neuen Testaments
HNT: Handbuch zum Neuen Testament
JAC: Jahrbuch für Antike und Christentum
JJS: Journal ofJewish Studies
JQR: Jewish Quarterly Review
NGG: Nachrichten der Gesellschaft der Wissenschaften zu Göttingen
NCB: New Century Bible
OCT: Oxford Classical Texts
PEQ: Palestine Exploration Quarterly
REJ: Revue des Études Juives
RHR: Revue d'Histoire des Religions
RScPhTh: Revue des Sciences Philosophiques et Théologiques
RScR: Revue de Science Religieuse
RTR: Reformed Theological Review
SB: Stuttgarter Bibelstudien
SBLDiss: Society of Biblical Literature Dissertation Series
SEÅ: Svensk Exegetisk Årsbok
StANT: Studien zum Alten und Neuen Testament
ThQu: Theologische Quartalschrift
ThR: Theologische Rundschau
ThZ: Theologische Zeitschrift
WMANT: Wissenschaftliche Monographien zum Alten und Neuen Testament
WUNT: Wissenschaftliche Untersuchungen zum Neuen Testament
ZKG: Zeitschrift für Kirchengeschichte
ZKTh: Zeitschriftfür katholische Theologie
ZThK: Zeitschriftfür Theologie und Kirche
 INTRODUCTION

I. TEXT
It is unnecessary here to repeat the information, given in 1.2–20,
concerning the Greek and other manuscript sources on which the text
of Acts is based,1 or to do more than restate in outline the problem
which those sources present; for this problem see 1.20–29. It
exaggerates only slightly to say that the authorities for the text fall
into two blocks: that which in this commentary is described as the
Old Uncial text and is represented best by the MSS B and X, and that
which is known as Western—a term which, it must be repeated, is
geographically misleading but so well established that it scarcely
needs the inverted commas with which the adjective is often
provided. So far as this simple statement is exaggerated the exag-
geration is twofold: there is a third, later, conflate form of text
(possibly also a Caesarean text, but this is doubtful); and the Western
text, though for many purposes it may reasonably be treated as a
unity, manifests a good deal of variety in detail. The view taken in
this commentary has already been made clear. It is that the Old
Uncial text gives on the whole a trustworthy account of what Luke
wrote, though the Western text, which is undoubtedly old, may
occasionally point back to the original wording. This view is not
taken a priori but as the result of the discussion of many passages; if
it is adopted, even if it is right, the textual critic’s task is not ended.
What is the original form (or, what are the earliest forms) of the
Western text? And, how, when, and where did the Western text come
into being? These are the most pressing, interesting, and divisive, but
pot the only questions that confront the textual critic.
It seems right to return to the most notable attempt in recent years
to vindicate the originality and authority of the Western text, that of
Μ. E. Boismard and A. Lamouille,2 though serious discussion of it

1I should however add references to D. C. Parker, Codex Bezae: An Early Christian

Manuscript and its Text, Cambridge, 1992, and K. Aland, Text und Textwert der
griechischen Handschriften des Neuen Testaments, III Die Apostelgeschichte, Berlin
and New York, 1993, neither of which was published when Volume I was completed
and given to the press. Parker thinks that D was written about AD 400, in Berytus. See
also K. E. Panten, A history of research on Codex Bezae, with special reference to the
Acts of the Apostles, unpublished Ph.D dissertation, Murdoch University, Australia,
1995.
2M. E. Boismard and A. Lamouille, Texte Occidental des Actes des Apôtres, two
volumes, Paris, 1984. See 1.24f.
xix
XX COMMENTARY ON ACTS

would double the size of this Introduction, and perhaps of the

commentary. Boismard and Lamouille attempt to establish the
original form of TO (Texte occidental. Western text). This is by no
means a matter of simply reprinting the text of the MS D, which, they
maintain, is an edited Western text. This points at once to a basic
difficulty of which of course Boismard and Lamouille are well
aware. The other main authorities for the Western text are not in
Greek, but in Latin, Syriac, and Coptic, from which an original
Greek has to be reconstructed by retroversion. This can be done; it is
well done by Boismard and Lamouille; but inevitably a measure of
uncertainty remains. The next step is, on the basis of the recon-
structed authorities, to establish TO. TO thus established, Boismard
and Lamouille investigate its language and style. Drawing up lists of
characteristic idioms they argue that it is Lucan; it must have been
written by Luke himself, by the author who was responsible for the
TA (Texte alexandrien, the Old Uncial text). Thus both forms of the
text go back to the same author. The Western text is a partly modified
form of Luke’s original text; the Old Uncial text is his own revision.
That is, both texts are Lucan; the Western text, so far as we can
recover it, is the older.
The facts are essentially as Boismard and Lamouille have stated
them;3 their lists of Lucan vocabulary and idiom demonstrate a good
measure of Lucanism in the Western text. The question is whether
this observation can prove their conclusion. Probably it cannot and
does not. The alternative view is that the Western text is a revision
(or, the various forms of the Western text are revisions) of the earlier,
Lucan, Old Uncial text. It seems almost inevitable that a revision,
unless an extremely radical one, will continue to show characteristics
of the original text. Many will simply slip through into the new text,
and even where a new form of words is chosen the reviser, working
constantly on a specific, in this case Lucan, base, will tend to adopt
the Lucan manner – he may even do so deliberately, in order that his
revision may have the appearance though not die substance of
authenticity.4
The point to which we may return is that of the variety and the

3See however the review by J. N. Birdsall, JTS 39 (1988), 571–7.

4See Μ. Wilcox, ‘Luke and the Bezan Text of Acts' (Kremer, Actes, 447–55), who
recognizes the text of D as Lucan, and thus offers some support to my essay in FS
Black (1979), 15–27, and in a different way to Boismard and Lamouille, though he
does not refer to my paper and their book was not published when he wrote. See also
C. Martini, ‘La tradition textuelle des Actes des Apôtres et les tendences de l'Église
ancienne' (Kremer, Actes, 21–35), with the double conclusion: there is no evidence of
clear doctrinal tendencies, but there is a distance from the original context and from
Judaism, with concern for fidelity to tradition and freedom with regard to the letter of
the text.
 INTRODUCTION xxi

characteristics of the Western readings. Some examples have already

been given,5 a few more examples follow.
There are familiar and frequently discussed places where the
Western text offers a different view of the matter in hand from that of
the Old Uncials. The best example is 15.20, 29; cf. 21.25, where the
Apostolic Decree is given in a different form and with (at least) a
different emphasis. Other examples are 17.15, where, according to D,
Paul passes by Thessaly because he is forbidden to preach the word;
18.7, where the Western text suggests that Paul left the house (or the
business?) of Aquila; 18.17, where it asserts that the Greeks seized
and beat Sosthenes; 19.1,2, where D and others add that Paul wished
to go to Jerusalem but was sent by the Spirit to Asia; at 20.3 there is
a similar spiritual direction; at 21.16 the Western text gives a
different account of the last stages of Paul’s journey to Jerusalem.
Variants such as these could all be the work of a single reviser who
had his own notions of what happened and what it signified. But
another class of variants is significant because of their insignificance.
Examples are frequently pointed out in the notes; here the following
may represent many.
16.25: τὸ μεσονύκτιον] μέσον τής νυκτός D*
16.35: ἀπέστειλεν οί στρατηγοί] συνήλθον οί στρατηγοί έπί τό
αυτό εις την αγοράν καί ἀναμνησθέντες τόν σεισμόν τόν
γεγονότα ἐφοβήθησαν καί άπέστειλαν D syhmg
17.8: τόν όχλον καί τούς πολιτάρχας ] τούς πολιτάρχας καί τόν
όχλον D gig syp
19.9: ὡς δέ τινες ] τινὲς μάν συν αυτών D
22.29: ευθέως ουν ] τότε D
28.18: οἵτινες ] + πολλά 614 pc syh** [D is not extant at this
point.]
Variants such as these (and they are many) suggest nothing more
(and nothing less) than an intention on the copyist’s part to reproduce
the content of the text before him, which is evidently judged to be of
some value, but a lack of concern with its precise wording: any
words will do provided that they represent the sense of the text with
reasonable accuracy; if they express that sense with greater vividness
and give the narrative a greater connectedness, so much the better.
This is very different from the traditional regard for a sacred text,
which must be preserved intact, and points to what may probably be
regarded as the origin of the Western text, or rather texts, all
characterized by a free and therefore not identical paraphrasing of the
original. It will appear6 that there was a period when Acts was in
circulation but was not regarded as a canonical text; it is for this
5See I.5, 16f., 19.
6See p. lxviii.
xxii COMMENTARY ON ACTS

reason that there is a far greater range of textual variation in Acts

than occurs in the gospels and in the Pauline letters. This account of
the origin of the Western text does not in fact differ as much from
that of Prof. B. Aland7 as may at first sight appear, but it does omit
the hypothesis of something like a Western text, in which many
‘wild’ variants were edited into a unified edition which subsequently
split up into the variant forms known to us today in D, the Old Latin,
some Syriac versions, and a few other sources. We have no good
ground for positing a standard Western text. This observation of
course bears also on the work of Boismard and Lamouille. It leaves
open the probability that the Western text will have not infrequently
preserved the original text, most frequently of course when, as often,
it is in agreement with the Old Uncial text and occasionally when the
two are in disagreement.8
In practical terms, this means that the critic will pursue the
‘eclectic’ method, for the original wording may turn up anywhere
(though it does turn up most frequently in the Old Uncial MSS).
Textual divergence was probably at its greatest in the earliest period,
and this means that the divergent texts (as known to us) will
(whatever the dates of the codices in which they are contained) be
early texts, and sometimes their readings will be original when the
Old Uncial texts have been revised under the influence of classical
rules and good literary taste.9 It is important that several lines of
study concur to establish the picture of Acts as a relatively early
work that achieved canonical status late; see also IV: acts in
history.

Bibliography
B. Aland, EThL 62 (1986), 5–65.
K. Aland, ‘Alter und Entstehung des D-Textes im Neuen Testament', in S.
Jeneras (ed.), Miscellania Papirologica, Barcelona 1987.
K. Aland, Text und Textwert der griechischen Handschriften des Neuen
Testaments, III Die Apostelgeschichte, Berlin and New York 1993.
C. K. Barrett, FS Black (1979), 15-27.
J. N. Birdsall, FS Klijn, 39-45.
Μ. E. Boismard and A. Lamouille, Texte Occidental des Actes des Apôtres,
Paris 1984.

7‘Entstehung, Charakter und Herkunft des sog. westlichen Textes untersucht an der
Apostelgeschichte’, EThL 62 (1986), 5-65. See I.25-29.
8See also Ropes on the origin of the Western text, Begs. 3.ccxc-ccxcvii.
9See here especially the work of G. D. Kilpatrick, notably in the posthumous
publication of his textual essays: The Principles and Practice of New Testament
Textual Criticism. Collected Essays ofG. D. Kilpatrick, ed. by J. K. Elliott, BEThL 96,
Leuven, 1990.
 INTRODUCTION xxiii

J. W. Childers, NTS 42 (1996), 55-74.

A. C. Clark, The Acts of the Apostles, Oxford 1933.
J. K. Elliott, NTS 34 (1988), 250-58.
E. J. Epp, The Theological Tendency of Codex Bezae Cantabrigiensis in
Acts, SNTSMS 3, Cambridge 1966.
T. C. Geer, Bib 69 (1988), 27-46.
T. C. Geer, NovT 31 (1989), 39-47.
T. C. Geer, JSNT 39 (1990), 59-76.
E. Haenchen and P. Weigandt, NTS 14 (1968), 469-81.
R. P. C. Hanson, NTS 14 (1968), 282-6.
R. F. Hull, JBL 107 (1986), 695-707.
A. Jousseu, Die koptischen Versionen der Apostelgeschichte, BBB 34, Bonn
1969.
Ad. Jülicher, ZAW 15 (1914), 163-88.
R. S. Mackenzie, JSNT 6 (1980), 522-76.
R. S. Mackenzie, JBL 104 (1985), 637-50.
C. Martini, Kremer, Actes 21-35.
C. D. Osburn, JSNT 44 (1991), 39-55.
D. C. Parker, JSNT 15 (1982), 97-112.
D. C. Parker, Codex Bezae: An Early Christian Manuscript and its Text,
Cambridge 1992.
T. Petersen, CBQ 26 (1964), 225-41.
J. H. Petzer, NTS 39 (1993), 227-45.
W. A. Strange, The Problem of the Text in Acts, SNTSMS 71, Cambridge
1993.
B. H. Streeter, JTS 34 (1933), 232-41.
Μ. Wilcox, Kremer Actes 447-55.
B. Witherington, JBL 103 (1989), 282-4.
xxiv COMMENTARY ON ACTS

II. SOURCES

The question what sources were used by Luke in Acts 1-14 is

considered in 1.49-56.1 There are now few who think that it is
possible, by the detection of doublets, to find in chs. 2-5 a pair of
written sources.2 The apparent doublets can be otherwise explained,
and it is improbable that the church of the earliest years kept written
records of its activities suitable for literary editing. Luke drew on
such oral traditions as were available; he may occasionally have
found something in writing (see e.g. 1.15, with the note). More can
be said about chs. 6-14, though here too it is well to be cautious with
regard to written sources. Whoever the author of Acts may have been
he had at his disposal a quantity of information about Paul, and this
may well have begun with an account of Paul’s conversion (9.1- 19).
The author’s apparent contact (not necessarily at first hand) with
Philip the Evangelist (21.8) may have provided information not only
about Philip’s own activities (8.5-13, 26-40) but also about early
Caesarean traditions. There was probably a cycle of stories about
Peter (9.32-43; perhaps 10.1-11.18; perhaps 12.3-17; earlier stories
such as 3.1-10 may have been included). More important however
as we move into the later chapters, is the Antiochene tradition, which
runs back from the foundation of the church in Antioch (11.19-26)
as far as the story of the Hellenist widows (6.1)3—with a great deal
of material, some from other traditions, mentioned above, inter-
polated into it. Antiochene material runs forward also. 11.27-30
manifests the Antiochene church’s independence, in that it is able to
minister to its mother, and (see I.54f.) the story in chs. 13, 14 of the
appointment of Barnabas and Saul, and of their mission in Cyprus
and Asia Minor, is told from an Antiochene standpoint. The first
question to be considered here is whether the Antiochene tradition
continues into ch. 15.
This seems probable.4 The Jerusalem Council is precipitated by
events in Antioch, where, according to the narratives in Acts, there

1It is important now to add a reference to J. Taylor, Les Actes des deux Apôtres
(Études Bibliques, Nouvelle Séne 23), Paris, 1994; also (p. xxxii) a further volume.
2The best statement of this now largely abandoned view is that of A. Hamack, The
Acts of the Apostles, ET London, 1906, pp. 172-86.
3Though this verse and the use of the word Hellenist are probably Luke’s own; see
the note on the verse, I.305-9.
4See below, pp. 696f., 710f.; and the detailed notes on many of the verses in ch.
15.
 INTRODUCTION XXV

had been a mixed church including both Jews and Gentiles since
11.20f. Acts 15 begins in Antioch and ends in Antioch (15.30-41),
but it does not seem probable that the whole of the intervening
account was derived from Antioch. According to Acts 15.1-5 the
leading Antiochene representatives were Paul and Barnabas; this is
confirmed in the parallel account5 in Gal. 2.1-10 where Paul himself
gives a record of the Council which differs in some ways from that in
Acts; the latter probably incorporates Jerusalem memories as well as
Luke’s own editing. For further discussion of the sources of ch. 15
see pp. 710f.
With ch. 16 we encounter a new phenomenon.6 In a number of
passages the narrative is set in the first person plural, which prima
facie suggests that the story is being told by one who was present.
The verses in which the significant pronoun or verb form occurs are
the following:7 16.10-(14)-17; 20.5-8, 13-15; 21.1-8, 11, 12, 14-
18; 27.1-8, 15, 16, 18, 20, 27, 29, 37; 28.1, 2, 7, 10-16. Between
these verses the narrative is sometimes carried forward in a way that
shows that a story in the first person is being continued, though this is
not always so. The most natural interpretation of these passages is
that in them the story is being told by one who was present and took
part (though possibly only a reporter’s part) in the events described.
How this interpretation should be understood, and whether it is in
fact correct, will be considered below (see pp. xxvii-xxx). Our first
task must be to survey the materials. ‘We’ appears first at 16.10,8
when Paul, in Troas, is summoned by a dream vision to extend his
mission to Macedonia. Accordingly, we sought to leave for Mace-
donia. The person, real or fictional, who says this, is in Paul’s
company; it is reasonable to think that he joined Paul if not at Troas
itself at least in that area. It seems unreasonable to think (though it
has been suggested) that he himself was the man of Macedonia, in his
narrative transforming himself into a vision. He accompanies Paul to
Macedonia, reaching Philippi at v. 12. The first person continues up
to v. 17 (its absence from 16.14 is insignificant), the point at which
Paul exorcises the Python spirit, and with Silas is thrown into prison.
We appears no more in the chapter. This could mean that the person
responsible for it had left Paul’s circle, but a much more probable
cause is that whereas, up to this point, all the party have been
involved in the events described, henceforth the story is focused
upon Paul and Silas, who are arrested, beaten, imprisoned, delivered

5Some find the parallel to Gal. 2.1-10 not in Acts 15 but in Acts 11.30, or
elsewhere. This is not probable; see pp. lxf., and elsewhere in the notes on ch. 15.
6See however the reading of D (p w mae) at 11.28 and Bultmann’s conjecture at
13.2; I.564,604.
7In addition at 16.8 the Latin version of Irenaeus has nos venimus (for
κατέβησαν).
8But see notes 6 and 7 above.
xxvi COMMENTARY ON ACTS

by the earthquake, cared for by the converted gaoler, and as Roman

citizens are set free by the apologetic magistrates. The ‘first person’
was out of the story and was therefore not mentioned. Did he, when
Paul moved on, remain in Philippi? The question is proper whether
we are thinking in terms of history or of fiction. He does not appear
in the rest of Paul’s Macedonian ministry (17.1-12), or in his
mission to Athens (17.13-34). Throughout ch. 18 the story is told in
the third person (the work in Corinth, the journey to the east, and the
return to the west), and so it is in ch. 19 (Ephesus). We returns at
20.5. Sopater and others waited for us in Troas; we sailed away from
Philippi and joined them at Troas. Had the ‘first person’ been in
Philippi all this time? It may be so; it may be that the author wishes
us to think so; it may be that as a writer of fiction he thinks that
variety would be pleasant and reintroduces his alternative reporting
style. We continues up to the Christian supper in v. 8, drops out in the
story of Eutychus, and is resumed when the voyage is taken up again
in v. 13. Does this mean that the miracle story of the raising of
Eutychus was drawn from a different source, perhaps not an eye-
witness source, a less trustworthy source? It may be so. We continues
in w. 13–15; then stops. This may mean that not only the speech of
20.18- 35 but the whole Miletus episode is from a different source.
This again seems quite probable, though it cannot be proved.
The journey is resumed in 21.1, and we returns, found in every
verse up to v. 8. Its absence from v. 9, which simply tells objectively
of Philip’s four daughters (who could well be there because they
were known to us), is not significant (though the contact between us
and Philip may be). In the absence of a pronoun it is implied in v. 10
(made explicit in E <a> gig syhmg, which have ἡμῶν, explicitly
absent in 1175, which have αυτών), and continues up to v. 18,
where Paul meets James and the elders, with an unimportant
omission in v. 13. We thus sees Paul through to his destination; it
then disappears. Did the ‘first person’ leave Jerusalem, so that
information about Paul’s adventures there and in Caesarea had to be
drawn from another source? This could well be so, but we remember
16.19-40, where the Philippian troubles are described in the third
person, perhaps for no other reason than that they concerned only
Paul and Silas.
We returns in ch. 27, in the account of the voyage and shipwreck,
again with possibly significant intermissions. It runs without a break
up to v. 8, the arrival of the ship at Fair Havens. It is not resumed till
after the breaking of the storm in v. 15. The intervening passage
recounts the disagreement between Paul and the nautical authorities
on the propriety of continuing the voyage; was this interpolated into
narrative that did not originally contain it? or does the absence of the
first person simply mean that this part of the story is about Paul and
not about us? We continues from v. 15 to v. 18. Its absence from
 INTRODUCTION xxvii

vv. 19, 20 does not seem to be significant, but vv. 21-26 are a
possibly independent story about Paul, comparable with w. 9-14.
We occurs in vv. 27, 29, but after that virtually disappears, since the
count of the ship’s company in v. 37 could stand on its own.
In ch. 28 we is found in vv. 1, 2, 7, 10, that is, in the bare account
of the residence in Malta; it is not found in the stories of Paul’s snake
bite and his cure of the father of Publius. These could be drawn from
a different source. Verses 11-16 bring Paul to Rome, and from this
point the story is focused on him, and the first person plural
disappears.
What is to be made of these We-passages? As was said above, the
prima facie inference to be drawn from them is that the person who
wrote them was present at the events he describes; this however may
be, and has been, understood in different ways.
(1) The traditional understanding of them, which goes back as far as
the Muratorianum,9 is that the author of Acts as a whole was an
eyewitness of the events concerned, breaking into first person
narration at those points at which he was himself a member of
Paul’s party.
(2) A second view is that the author of Acts is to be distinguished
from the person responsible for a source that he was using; it was
this latter person who used the ‘we’ and was present; his first
person narration was retained by the author of the book though it
was no longer strictly appropriate.
(3) The third possibility is that the ‘We’ is fictitious, included in the
text for some editorial purpose, perhaps simply in order to add to
the verisimilitude of the narrative.
An admirable account of the debate between those who have
maintained these opinions has been given by C.-J. Thornton,10 and
there is no need to retell the story here. It is impossible also to follow
Thornton’s careful and thorough examination of the Ich-Erzählung in
ancient literature. It must suffice to quote the conclusion that he
reaches on the basis of this examination. ‘Die Wir-Erzählungen der
Apostelgeschichte enthalten nichts, was antike Leser nicht für völlig
realistisch gehalten hätten. Sie konnten darin nur einen Bericht über
die wirklichen Erlebnisse des Autors erblicken. Hätte der Autor die
in Wir-Form geschilderten Reisen gar nicht mitgemacht, so wären
seine Erzählungen darüber—auch nach antikem Verständnis—
Lügen’ (p. 141). So far Thornton’s conclusion seems to be fully
justified by the evidence cited. He has however a good deal more to
say. The writer of the We-passages may—should—be accepted as
one of Paul’s travelling party. It does not follow that we are dealing
9Lines 35-37: Lucas optimo Theofilo comprendit quae sub praesentia eius singula
gerebantur ... See 1.44.
10Der Zeuge des Zeugen, Tübingen, 1991.
xxviii COMMENTARY ON ACTS

throughout with ‘reinen Tatsachenberichten’, There is in fact a

‘Skala von Fingiertheitsgraden’, and it has many steps. We can
accept the author’s participation in Paul’s journey to Rome ‘ohne
damit automatisch die Autentizität der Paulusreden an Bord oder der
Schlangenbiss Episode auf Melite in Kauf zu nehmen' (p. 142). Here
it seems better to take up the observation made earlier. The first
person, used three times (ἐπέγνωμεν, ἡμîν, ημᾶς) in 28.1, 2, drops
out entirely in vv. 3-6, the story about the snake; it is resumed in v.
7. Similarly, some (see above) though not all of the speeches
attributed to Paul in ch. 27 are in the setting of third person narrative.
It is a more plausible view that there are places (even within the
account of the journey to Rome) which did not come to the author of
the book as the words of an eyewitness. This leads to a further point.
If 28.1, 2 and 28.3-6 came to the author of Acts along different
channels we are dealing not with various degrees of credibility
within an Ich-Erzählung but with the combination of different
sources, of which only one was a We-source. This immediately
suggests the question of the relation of the author-editor of Acts to
the author of the We-source. Thornton maintains their identity, but
the arguments by which this conclusion is reached are less impress-
ive. They are bound up with the attribution of the work to Luke, the
companion of Paul (Col. 4.14; Phm. 24). This is a matter that will be
considered below (see pp. xxixf.); Thornton lays great stress on the
witness of Irenaeus, which he traces back to the beginning of the
second century, and, taking the date of Acts to be in the time of Titus
or the early years of Domitian, points out that at this time not only the
historical Luke but others such as Timothy must still have been alive,
so that pseudonymity can hardly have been possible, while in a book
with a dedication anonymity would have been out of the question.
Ascription to Luke implies a special relation to the history and thus
to the eye-witness. All this is interesting and, more important, may be
true, but it is not fully convincing. In particular it runs into the
difficulty of supposing that the author of Acts as a whole was
familiar with Paul. On this difficulty see below, pp. xlivf. and
elsewhere, where it is argued that the author, though he greatly
admired Paul, did not fully understand him and at some points
misunderstood him. Thornton recognizes the difficulty, and deliber-
ately excludes it from consideration. ‘Die Frage nach dem Verhältnis
von lukanischer und paulinischer Theologie etwa muss ich ganz
ausklammem; das müsste Gegenstand einer eigener Studie sein, die
im Licht der folgenden Untersuchungen zu wesentlich anderen—und
methodisch gesicherteren – Ergebnissen fuhren würde, als sie heute
zumindest in der deutschen Forschung weithin akzeptiert werden’ (p.
200). Such a study might however lead to confirmation of a fairly
widely accepted view, and to no more than qualified acceptance of
Thornton’s opinion. That Acts as a whole was written by one of
 INTRODUCTION xxix

Paul’s immediate circle is very difficult to believe; that the author,

whoever he may have been, was able to draw on one or two sources
derived from that circle—the We-passages and perhaps some oth-
ers—is probable.
On this view we are left with the difficult problem of explaining
why the final author saw fit to leave the first person in the passages
taken from the presumed source. So far as one can tell he did not in
general treat his sources with such wooden rigidity as might lead him
to retain the precise wording of this one. If it was a source of such
exceptional authority that he felt that he must transcribe it word for
word one would have expected him to say what the authority was. It
may be that it was the only source available to him that came directly
from the Pauline party; this would make it distinctive and he might
have supposed that the distinctive mode of expression would make
this clear. In the end however we have to balance against each other a
factual problem (the misrepresentation of Paul’s thought and action)
and a verbal problem (the retention of ‘We’ passages in the midst of
‘He/they’ passages); and it seems reasonable to give the factual
problem priority. It is perhaps best to suppose that the We-source
was little more than an itinerary, a list of places visited, with some
notes on lodging and means of transport. This applies well to most of
the passages listed above (pp. xxv-xxvii). Thus: ‘They waited for us
in Troas; we sailed from Philippi to Troas, a five-day journey; we
sailed to Assos to pick up Paul, who went by land; then Mitylene,
Chios, Samos, Miletus;’ and so on. When he could do so, the author
of Acts supplemented this outline with stories that he collected in the
places mentioned. This suggestion works least well with the opening
verses which give some details of ‘our’ stay in Philippi (cf. other
more personal notes at 20.36-38; 21.5, 6, 10-14). But the author of
the We-source was not compiling a Mediterranean shipping time-
table, or even a guidebook.
Between 21.28 and 27.1 there is no first-person narrative. This
long passage includes Paul’s visit to the Temple with the four vow-
makers; the attack on him by the crowd and the intervention of the
Romans; his speech to the crowd, renewed violence, and further
intervention by the Romans, who learn that Paul himself is a citizen;
Paul’s appearance before the Sanhedrin, the plot to kill him, and his
transference by the tribune to Caesarea; his accusation by the Jews
before the Governor, who leaves him in custody for his successor to
deal with; Paul’s appeal to Caesar and his statement in the presence
of Agrippa as well as Festus. There is a great deal of information
here, most of which cannot be traced to any specific source or
historically checked. One may wonder whether the eye-witness who
composed the ‘We’ passages remained in Jerusalem between 21.18
and 27.1, where he reappears to accompany Paul on the voyage to
Rome. This is not impossible. There is no reason to think that he was
XXX COMMENTARY ON ACTS

no longer in Philippi after 16.17: he simply ceased to write in the first

person because he was describing what happened to Paul and Silas
when they were not in his presence but in prison and dealing with
officials. He may have remained in Jerusalem and picked up what
information about Paul’s affairs he could get. As (presumably) a
Gentile he would not be admitted to the Temple; what happened
there he could describe only in the third person.11 Thereafter Paul
was in Roman hands and probably for the most part inaccessible,
though on the whole his captivity does not seem to have been strict
(23.16; 24.23; 26.31, 32) and a friend might have been able to gather
information about events at which he was not himself present. The
speeches he would no doubt compose himself. All this is no more
than a possibility. It must however be added that if it is not accepted,
we have no means of investigating the origin of the material
concerned; we can say only that if it is not fiction it12 must be Pauline
tradition, going back to some member or members of the Pauline
circle, and (in part) to Paul himself. If it is accepted, the problem
arising out of Thornton’s work on the ‘We’ passages is to some
extent at least alleviated, and we may suppose that Luke (if we accept
the name, but here too Thornton is not so easy to follow) was the
author of a considerable part of Acts, indeed of most of the second
part of the book. The person responsible for Acts 16-28 must in any
case be thought of as a traveller, sufficiently familiar with Corinth
and Ephesus, probably with other cities also, to have picked up local
traditions and memories of Paul and his missions. Much or all of the
contents of chs. 17, 18, 19 will be accounted for in this way, as well
as some of the paragraphs inserted into the ‘We’ passages in chs. 27,
28.

ADDITIONAL NOTES
The Acts of the Two Apostles See note 1.
The work of Μ. E. Boismard and A. Lamouille on the text of Acts
was considered in I.24f. and above, pp. xixf. It was followed by three
volumes by the same authors on the sources of Acts (Les Actes des
deux Apôtres). These were published in 1990; I was not yet aware of
11Was he right about the vow? Would Paul have taken part in this Jewish ceremony?
See pp. 1000f. and the notes on 21.21,24,25. He might perhaps have done so. He was
prepared to become a Jew to the Jews (1 Cor. 9.20), and this was, or might have
seemed to be, a private act which he could perform without committing Gentile
Christians to anything, whereas the Decree (15.29) committed them, if they would be
saved, to minimal legal observance. On the other hand, it is doubtful whether Paul
would have voluntarily undertaken a legal obligation in order to prove what was
certainly not the whole truth.
12Or some of it; it is to some extent possible to distinguish between tradition and
redaction.
 INTRODUCTION xxxi

their existence when I finished the manuscript of Volume I in

January 1991. They are to be followed by a historical commentary,
of which J. Taylor has now published the part that deals with Acts
9.1-18.22 (Paris, 1994) and a further volume (p. xxxii). In his Avant-
Propos he outlines the thesis of Boismard and Lamouille, on which
his own commentary is based. What is offered here is a summary of
his outline. The textual work referred to above is presupposed in this
account of sources.
The reconstituted Western text of Acts is the nearest approach we
have to the original text of Acts, the work of Luke himself. The
Alexandrian text is a revision of the original text made by another
hand. There are thus two editions of Acts. They contain also many
non-Lucan revisions, additions, and redactions. Luke had not only a
reviser but a precursor, a document which already had the form of a
life, passion, and resurrection of Jesus followed by the story of the
earliest church. This Boismard and Lamouille call Act I; Act II is the
work of Luke himself; Act III is the work of the Lucan reviser. But
Act I also made use of a yet more ancient work, already of the same
pattern, and centred on the work of Peter. This is DocP (= Document
Pétrinien). What we have here is not so much a theory of sources as
of several levels of composition. It is possible to go further still and
to distinguish a Journal de Voyage (JV), comprising most of the We-
passages, and a Document Johannique, emanating from a group of
disciples of John the Baptist who took John to be the Messiah.
The present writer can only say that he finds this elaborate
hypothesis not incredible, for it is quite possible, but incapable of
proof or of disproof, and therefore beyond serious discussion. There
is of course much that is independent of the hypothesis, and of
historical and exegetical value, in Taylor’s commentary.

Which was the Fourth Gospel? JSNT 54 (1994), 3-28.

In this article R. C. Morgan argues that Jn was known to and used by
Luke. This relates much less to Acts than to the Third Gospel (if the
term may be allowed), but there are a few points, such as the
Ascension, in which it is profitable at least to consider the matter
from the stand-point of Johannine priority. Even here, however, we
are dealing with theological rather than literary relations.

Bibliography
Μ. E. Boismard and A. Lamouille, Les Actes des deux Apôtres, Paris 1990.
H. J. Cadbury, NTS 3 (1957), 128-32.
L. Cerfaux, Rec. 2.63-91.
xxxii COMMENTARY ON ACTS

J. Dupont, The Sources ofActs, ET London 1964.

J. A. Emerton, JSS 13 (1968), 282-97.
A. Hamack, The Acts of the Apostles, ET London 1906.
C. J. Hemer, Tyndale Bulletin 36 (1985), 79-109.
J. Jeremias, ZNW 36 (1937), 205-21.
R. A. Martin, NTS 11 (1965), 38-59.
S. E. Parker, ‘The “We-Passages” ’, in The Book of Acts 2.545-74.
S. Μ. Praeder, NovT 29 (1987), 193-218.
J. Taylor, Les Actes des deux Apôtres (Actes 9.1-18.22), Paris 1994.
J. Taylor, Les Actes des deux Apôtres (Actes 18.23-28.31), Paris 1996.
C.-J. Thornton, Der Zeuge des Zeugen, Tübingen 1991.
J. Wehnert, Die Wir-Passagen in der Apostelgeschichte, Göttingen 1989.
Μ. Wilcox, The Semitisms ofActs, Oxford 1965.
 INTRODUCTION xxxiii

III. ACTS AS A HISTORICAL DOCUMENT

Read on the surface, as generations of Christians have read it, Acts

presents the history of the Christian mission in its first three decades.1
It is right to say ‘of the Christian mission’ rather than ‘of the
Christian church’, for Acts shows little interest (it would be wrong to
say, no interest) in the structure and development of the church as an
institution. The history is only a sketch; its author probably intended
to confine himself to one roll of papyrus, matching his earlier volume
on the story of Jesus, which in turn matched other works on the same
theme. The outline of the sketch is familiar. Everything turned on the
fact that the crucified Jesus was raised from death by God, reas-
sembled his disciples, and fulfilled his promise of sending to them
the gift of the Holy Spirit. His parting from them is narrated, and the
descent of the Spirit on the Day of Pentecost leads immediately to the
beginning of the mission in Jerusalem (cf. Lk. 24.47). There follows
not a next-day event but what is obviously intended as a specimen of
early Christian activity: there is a miracle of healing, a crowd
assembles, and the Christian message is proclaimed once more.
Another event, doubtless again a representative rather than a unique
event, follows: the unauthorized, and unwelcome, preachers are
arrested and brought before the Jewish Council. They are warned not
to repeat their offence, a warning which they evidently have no
intention of heeding. What Luke tells us at this stage about the
primitive community is not that it had such-and-such an organiz-
ation, but that its members prayed and took their meals together, sold
their property and shared the proceeds; this leads to the shameful
discovery that a couple have lied about the price obtained for their
property. Their fate occasions fear of divine retribution but not fear
of the religious and civil authorities. Preaching continues; there are
further arrests and a beating; and evident intention to continue the
mission.
This closes the first part of the story (chs. 1-5). The next begins
with a second regrettable discovery: the charitable distribution of
supplies proves to be inequitable. The apostles have enough to do
without giving time to this task and seven others are appointed. One
of them, going far beyond this appointment, preaches and engages in
controversy to such effect that he too is arrested and brought before
the Sanhedrin. His long speech is followed by his martyrdom, and

1F. V. Filson, Three Crucial Decades, London, 1964.

xxxiv COMMENTARY ON ACTS

the outbreak of general persecution, which has the effect of pushing

the mission out beyond Jerusalem, as the Christians escape from their
persecutors. Further preaching and miracle working lead to the
conversion of Samaritans and of an Ethiopian on his way back from
Jerusalem to his own country. The arch-persecutor also leaves
Jerusalem in order to deal with Christians in Damascus, but on the
way, encountering Christ himself, he becomes a Christian and
returns to Jerusalem, where, not without difficulty, he is united with
the main body of Christians and works as a mission preacher till
opposition enforces his withdrawal from the city (6.1-9.30).
A third phase (9.31-12.25) begins when Peter also leaves Jerusa-
lem and after performing miracles on the way reaches Caesarea
where he is led into contact with a Roman centurion. The mission
preaching is repeated and the centurion, Gentile though he is,
unbaptized as he is, becomes a Christian and receives the Spirit.
Peter returns to Jerusalem, receives the inevitable criticism from
those who thought it right to maintain the old barriers between Jews
and Gentiles, and overcomes it by telling the story. The mission to
the Gentiles receives a fresh, independent beginning in Antioch,
where the church is founded and develops in independence, man-
ifested in a gift to the mother church in Jerusalem. The third phase is
wound up in Jerusalem, where persecution initiated by King (Herod)
Agrippa I is ended by the discomfiture of the tyrant.
The scene shifts to Antioch (13.1), where the mixed Jewish-
Gentile church commissions Paul and Barnabas to take the mission
further. They travel to Cyprus and to central southern Asia Minor,
preaching, working miracles, making disciples, establishing chur-
ches, eventually returning to their base and reporting to the church.
At this point a vital problem appears. It is undoubtedly true that
Gentiles have been converted by the preaching of the Gospel of
Jesus. What next? There are some who take the view that these
converts, if they are to be saved members of the people of God, must
be circumcised and thereafter observe the law of Moses. Others think
that this is not necessary but imposes an uncalled-for burden upon
the Gentiles. This was too big an issue to be settled in Antioch; it
must be discussed and decided in Jerusalem. After some debate the
decision is reached: Gentile converts do not have to be circumcised
and keep the whole Law; they must however observe a few
regulations. This has the effect of setting Paul (and no doubt others)
free to pursue the mission without restriction; and, with companions
and assistants, he does so (chs. 13-15).
The rest of the book contains the record of Paul’s work throughout
the north-éastem quadrant of the Mediterranean. He repeatedly meets
opposition but everywhere there are converts and churches are
established in such places as Philippi, Thessalonica, Corinth, Ephe-
sus. Eventually he returns to Jerusalem, where his arrival creates
 INTRODUCTION XXXV

renewed difficulty. Jewish Christians have been given the impression

that Paul in his travels throughout the Diaspora has been teaching
Jews not to circumcise their children, not to keep the Law, has been,
in fact, practising and preaching apostasy from Moses. The Christian
authorities are sure that this report is not true, but Paul must
somehow demonstrate its falsehood. The suggestion is made that he
should accompany and pay the expenses of four men who were due
to be released from a vow. Paul accepts the suggestion, but the plan
misfires. He is accused not only of attacking the Law but of
profaning the Temple by bringing Gentiles into it. He is rescued from
an angry mob by Roman soldiers, and from this point to the end of
the book is in Roman hands. The Romans do not release but they do
protect him. At length he appeals to Caesar and is sent to Rome. It is
a stormy and dangerous passage, but in the end he arrives, and the
Gospel is preached in the capital of the Empire, and even if Jews
close their ears the Gentiles will listen (chs. 16-28).
‘Read on the surface’ this is a plain tale. It is history, of a popular
kind. That means that it is biographical, focusing on a few out-
standing characters, episodic in style,2 highly coloured with wonder-
ful and exciting incidents, and not given to philosophizing about the
nature and meaning of history or about the theological content of the
message. Luke’s interest in the spread of the Gospel into the non-
Jewish world is plain, and so is his sense of achievement when he can
narrate the great missionary’s arrival at Rome—a goal probably in
his mind as early as 1.8 (‘... up to the end of the earth’). More
searching inquiry, however, and a reading that digs beneath the
surface, unearth problems which the reader cannot ignore. In reading
5.1-11, for example, he will ask whether Ananias and Sapphira were
in fact guilty of nothing more than money-grubbing deceit; does
there not perhaps lie behind their action a divergent view of the way
in which die church should handle its corporate finances? In ch. 6 we
seem at first sight to have a simple problem in the field of social
concern, speedily resolved by administrative adjustment. Was it so
simple? Who were Hellenists and Hebrews, and how did they differ?
The word Hellenist is used twice more, in 9.29 and 11.20, each time
with a different meaning. Can the difference between the two groups
have been purely linguistic? How was it that when all the Christians
were obliged to leave Jerusalem, the apostles (who would, one would
have thought, be regarded as ring-leaders) were excepted (8.1)?
Perhaps, notwithstanding the Temple disturbances of chs. 4 and 5,
they were less obnoxious to the authorities. The conversion and
baptism of an Ethiopian eunuch (who could not have been a
proselyte) might seem, as an example, to settle the question whether
2See especially E. Plümacher, Lukas als hellenistischer Schriftsteller (Göttingen,
1972); ‘Die Apostelgeschichte als historische Monographie’, in Kremer, Actes,
457-66. See also W. C. van Unnik in Kremer, Actes, 37-60.
xxxvi COMMENTARY ON ACTS

Gentiles might become Christians without circumcision; but Peter’s

visit to Cornelius is described as an unprecedented step. It leads to
the conclusion, ‘To the Gentiles also God has granted that repentance
that leads to life’ (11.18); if this was agreed why was another
conference necessary to discuss the question whether Gentile con-
verts should be circumcised (15.1, 5)? This introduces Paul, who, in
the Acts narrative, is present at the conference and assents to and
helps to disseminate a decree to which in his extant letters he never
refers, a decree which it is very difficult to harmonize with the
theology of those letters. Apart from this the account of Paul’s work
to which Acts proceeds raises few serious problems, though some of
the speeches attributed to him, especially the Areopagus address
(17.22-31), are scarcely consistent with his thought as otherwise
known.
These problems, and others, are discussed at appropriate points in
the Commentary. What must be attempted here is not a repetition of
what is done elsewhere but an integrated statement which may lead
to a general evaluation of the book before us. If this is to be done
there can be no doubt that we should begin with the problems that are
created by the account of the Council in ch. 15. This is not only
because these problems are of unusual importance but because they
lead, in many cases, to problems in other parts of the book, and
because the Council has often, and rightly, been described as the
centre of Acts. It is this in itself, and also (as is pointed out on pp.
709f.) as the prime and determining example of a pattern of narration
that occurs frequently. It is one of Luke’s ways of expressing the
victory of the Gospel. A difficulty arises; it is addressed by the
Christian community; a solution is found which not merely solves
the difficulty but leads to a further expansion of the Christian
movement. Acts 15 provides a crucial example of this pattern which
may be said without exaggeration to constitute the pattern of the
whole book. In the early chapters, having provided a basic statement
of Christian faith and life in the primitive Jerusalem context, Luke
collects such examples as he can find of the way in which the Gospel
spread out of its original setting into the non-Jewish world.3 There
now arises the supreme difficulty, in the assertion that the mission to
the Gentiles can be legitimate only if it is a Jewish mission; that is,
converts to Christianity must be circumcised. They can become
Christians only if they also become Jews, members of the people to
whom the promises were spoken and the Messiah came. If this
conviction had been upheld the world-wide mission would have
come to an end and the distinctive Christian message would have
ceased to exist. This supreme difficulty is solved in the Council, but
the result is not confined to the relief of the church in Antioch

3See I.50-52.
 INTRODUCTION xxxvii

(15.31); the result runs through the rest of the book which describes
the increasing success of the mission to the Gentiles and reaches its
climax when Paul proclaims the Gospel unhindered in Rome.
The difficulties in the account of the Council arise partly within
Acts itself, and partly when Acts is compared with the Pauline
letters, especially Galatians, where also a Jerusalem Council is
described. The internal difficulties may be briefly outlined as fol-
lows. See further detail in Sections 38, 39.
1. Verses 1 and 5 prepare the reader for a violent debate, but with the
exception of v. 7 and its reference to πολλή ζήτησις there is no
debate. All the participants are in agreement; there is no opposi-
tion.
2. In view of 11.18 it is not clear why there should be a Council; in
this verse it seems to be agreed that God is willing to grant to
Gentiles as they are, that is, without circumcision, the repentance
that leads to life. The issue is already settled.
3. When Peter speaks he seems to claim too much. That he should
refer back to the conversion of Cornelius in ch. 10 is natural
enough;4 but with what right can Peter describe the Law as an
intolerable burden which neither he nor his Jewish ancestors have
been able to bear?
4. It is surprising that Paul and Barnabas, appointed (according to
15.2) its representatives by the church in Antioch, should have
virtually nothing to say. They report (15.12) the signs and portents
done by God through them, but they bring no theological
argument to bear on the question whether circumcision should or
should not be demanded of Gentile converts. Has 15.12 been
inserted into a source that did not originally contain it? What
interest does Luke think that these two represent?5
5. When James speaks he agrees with Peter that circumcision should
not be required, pointing out that Peter is supported by the words
of the prophets; when however he quotes Amos to this effect he
uses a passage which in the LXX text (which he uses) makes his
point, but in the Hebrew text has a quite different meaning. This
use of the LXX is, to say the least, surprising.
6. The issue of the Council is a decree, suggested by James in v. 20
and incorporated (v. 29) in a letter (vv. 23-29).6 The inter-
pretation of this decree, and the grounds on which it rests, are
themselves disputed. The letter, including the decree, is addressed
τοῖς κατά τὴν ’Αντιόχειαν και Συρίαν και Κιλικίαν άδελφοῖς
τοῖς ἐξ ἐθνῶν; it is however conveyed by Paul, Barnabas, and
4Though this story itself is not without some serious problems; see 1.491-8.
5A related question is whether Luke did or could regard Paul and Barnabas as
apostles; see pp. xlivf., lxxxix and I.101, 644, 666f.
6On the textual problems that arise here see pp. 735f.; they need not be discussed at
this point.
xxxviii COMMENTARY ON ACTS

others to Antioch (v. 30), then to Derbe, Lystra, and other cities of
that area (16.1, 4); nothing is said of Syria and Cilicia, but it is
assumed that the decree is of wider application than the letter
itself suggests.
7. Finally, it is not clear exactly what was discussed at the Council.
In w. 1,5 it seems plain that the issue is that of salvation: without
circumcision, οὐ δύνασθε σωθῆναι. It is however often main-
tained7 that the decree itself deals with a different matter. It does
not present conditions on which Gentiles may receive salvation
but lays down procedures that will facilitate fellowship, and
especially fellowship at table, between Jewish Christians and
Gentile Christians. The latter must (the word ἐπάναγκες in v. 28
is not always noticed and given due weight) observe certain
regulations regarding food and other practices. Does the Coun-
cil—at least in Luke’s account of it—change course part way
through?
The questions that arise when Acts 15 is compared with the Pauline
literature may now be considered.
1. It must first be observed that Paul also describes a Council in
Jerusalem, in Gal. 2.1-10. It is hard to doubt8 that Paul and Luke
are describing the same event. Paul, Barnabas, James, and Peter
are present in both; according to Paul, Titus and John were there
too. Luke does not mention them but does not say that they were
not there. There was, according to Galatians, a strong move to
bring Paul and his Gentile Christians into bondage (to the Law);
this Paul resisted. The ‘Pillars’ recognized his mission, but there
was an agreement that he should go to the Gentiles, they to the
Jews, and a request that he and his colleagues should remember
the poor. Neither this agreement nor this request is contained in
Acts.
2. The Acts Council results in a Decree; there is no reference to this
in Galatians, or indeed in the Pauline corpus. The Decree rejects
idolatry and fornication; so does Paul, but without reference to the
Decree and its authority. The Decree explicitly forbids the use of
εἰδωλόθυτα; Paul explicitly permits it (so long as it does not
result in unloving treatment of others). Whether the Decree is
compatible with Pauline theology at large need not be discussed at
this point; it is enough to note that he seems to be unaware of it
and on at least one point to contradict it.
3. The Acts Council arises out of a visit of Judaean Christians to
Antioch. The Council in Galatians is followed by a visit to
Antioch of certain people who came from James and were

7See pp. 695-7.

8Though some do; see for example those mentioned in Kümmel, Einleitung 263.
 INTRODUCTION xxxix

responsible for the separation of Peter, Barnabas, and other

Jewish Christians from the Gentile Christians of Antioch.9
4. When the two narratives are taken together, each considered in its
own context, that of Acts and that of the Pauline corpus respec-
tively, the question of Paul’s status and the roots of his theology is
acutely raised. For Paul’s apostleship in Acts see p. lxxxix; in
only two verses (14.4, 14) is he described as an apostle; he is not
so described in ch. 15. Too much should not be made of this;
neither is Peter. Paul himself insists most strongly on his apostle-
ship (e.g. 1 Cor. 9.1f.), and in Gal. 2.6 he affirms that the men of
repute ‘added nothing to him’ (ούδέν προσανέθεντο)—gave him
no additional authority, or added nothing to his Gospel. They
recognized that he had been entrusted with the ‘Gospel of the
uncircumcision’ as Peter with that ‘of the circumcision’, and
perceived the grace that had been given to him. It is however
striking that in v. 8 the word άποοτολή is used with reference to
Peter, not with reference to Paul (though it may be held that it is
implied). It seems that in the Galatians narrative Paul’s status,
though at least partially affirmed, is one of the questions under
discussion.10 Acts 15 is silent about this.
It will be shown below (pp. 709-11) that, as far as the structure of
Acts 15 is concerned, the upshot of these observations is that Luke
has himself produced the apparently well told story of the Council,
though he has not created it out of nothing. He knew (correctly) that
there had been a gathering in Jerusalem in which Peter, James, Paul,
and Barnabas had taken part. He knew that there had emerged from
Jerusalem a decree which regulated the place of Gentile Christians in
the people of God. He knew that men who were or claimed to be
envoys of the Jerusalem church caused trouble in Antioch, and
perhaps elsewhere, by stressing the claims of the old religion. He
knew a number of arguments11 by which the mission to the Gentiles
might be justified and Gentiles admitted without the full requirement
of the Law. These did not include the great Pauline theological
argument, but they were not unworthy of consideration. He
believed—wrongly—that the Law in its fullness must be a burden
intolerable even to Jews. He knew (and had recounted the fact) that
Gentiles had received, in the gift of the Spirit, precisely the same
experience of salvation as Jewish Christians. Who could refuse to
such believers the water of baptism? He knew that as the mission to
the Gentiles had been pursued miracles had taken place. Would God
have granted such tokens of his power and favour if he had
disapproved of the proceedings? And finally Luke believed that
9Some think that in this Paul is not following a chronological order; the events of
Gal. 2.11-14 preceded those of 2.1-10. This is unlikely.
10Cf. the Western text of Acts 15.2 (see p. 701).
11Not (as Haenchen 121 says) only one—the signs and wonders.
xl COMMENTARY ON ACTS

Scripture itself made it clear that it was God’s intention to include in

his saved people ‘the rest of men’ (Amos 9.12, LXX), that is, the
Gentiles. These arguments Luke distributed among the participants
in the Council, as he thought, suitably.
The distribution of arguments, however, is sufficient to show that
the centrality of the Council in Acts (like the narrative itself) is
Luke’s own work. The mission to the Gentiles is his chief concern,
and he wishes to show that it received the backing of all elements
within the church. The few who dissented were speedily silenced or
convinced. It was not so in Paul’s lifetime. There was (as Galatians
shows) a measure of agreement between Paul and the Jerusalem
apostles, but it was by no means complete agreement. F. C. Baur12
rightly pointed out that if the Jerusalem apostles had fully accepted
Paul’s Gentile mission they would have joined it rather than
dissociating it from their own mission (Gal. 2.9). The very existence
of Galatians proves that the Council did not put an end to conflict.
The welcome Paul receives in 21.18-25 is, even in Luke’s narrative,
lukewarm, and from that point onwards there is not a word to suggest
that the church in Jerusalem and the church in Caesarea had any
interest in Paul’s fate. Luke associates Paul with Stephen (cf. 22.20),
that is, he was a legitimate, harmless, missioner to the Gentiles. But
Paul was not a Hellenist;13 he was a Hebrew. Luke did not write as he
did with the intention of conveying a false impression of the church
in its first age. He wrote as he did because his understanding was
coloured by the period in which he lived. It was a period in which
consensus had at last been reached.14 The strife of the first three
decades, often, as the Pauline epistles show, bitter and unrelenting,
had worked itself out and a period of peace ensued. In addition to the
fact that, badly as some of them failed, the Christians acknowledged
a common loyalty to Jesus and acceptance of his authority, which not
only gave them a common axis by which to orient their lives but also
laid upon them some sort of obligation to love one another, there
were several particular factors that contributed to the overcoming of
dissent and the establishing of consensus. By the end of the 60s Paul,
Peter, and James had all, it seems, died as martyrs.15 This removed
two highly contentious figures and one who, though possibly not
contentious in himself, could easily be used as a standard-bearer for a
third Christian group; it also made possible, perhaps inevitable, a
different evaluation of them. That they had once disagreed over
weighty matters now seemed less important than that they had
unitedly borne witness to Christ with their blood. They were united
in death; it was hard to believe that they had ever been separated in
12Paulus, 1125; 2142f.
13On this term see I.308, 309; also FS Borgen, 20-25.
14It was a Hellenist consensus; see FS Jervell 9f.
151 Clement 5.3-7; Eusebius, HE 2.23.4-18 (Hegesippus); Josephus, Ant. 20.200.
 INTRODUCTION xii

life. Again, Jerusalem, once the scene of dispute and the centre from
which Judaizing missions bore down upon the Pauline churches, had
virtually ceased to exist as a Jewish and Christian religious centre.
Sacrificial worship probably did persist in the Temple between AD
70 and 135,16 but this did not alter the fact that the Christians had left
the city, taking refuge in Pella.17 The anti-Pauline missions that had
emanated from Jerusalem had lost their base of operations; in any
case, they (but this is a comprehensive term, calling for differ-
entiation) had defeated Paul,18 whose radical theology and mis-
sionary principles were no longer understood and practised even by
those who admired him, such as Luke (who could take the Decree as
the core of his faith) and the author, or authors, of the Pastorals. The
Gentile mission and the Gentile church were fully accepted, and no
one who counted demanded circumcision. The Gentiles for their part
were willing to accept the Decree that had been drawn up many years
previously, and they continued to observe it for centuries to come.19
This was the situation in which Luke lived and wrote. He took it
for granted. This makes much better sense than to suppose that he
invented a fictitious story designed to conceal the horrid truth about
the church’s past. If Luke was practising concealment he was an
incompetent practitioner: traces of division, of which he was himself
perhaps hardly conscious, show through his narrative. These traces
are much better explained if we think of Luke as writing honest
history but writing it in an atmosphere different from that of the
period that he described. It is not merely that contemporary written
records were few, perhaps even as some would say non-existent, and
that memories, then as now, could be fallible. Any narrative needs an
appropriate frame of reference if it is to be intelligible and to give an
accurate impression of events. To enter with controlled imagination
into a past frame of reference is the historian’s hardest task; Luke did
not achieve it with complete success. He was describing contentious
events within a framework provided by a period of consensus. This
was his problem, and, in a different sense, it is ours, as we attempt to
understand and assess his work. Varying his language only a little,
one learns from F. C. Baur to recognize tendency and tension, the
one masking the other. Baur was fundamentally right to see in

16See K. W. Clark, NTS 6 (1960), 269-80.

17Eusebius, HE 3.5.3, probably drawing on Hegesippus. S. G. F. Brandon, The Fall
of Jerusalem and the Christian Church (London, 1951), 168-73 thinks that they took
refuge not m Pella but in Egypt.
18In this (but in little else) I disagree with Μ. D. Goulder’s A Tale of Two Missions
(London, 1994). He thinks that Paul won.
19For a somewhat different, and very important point of view, see J. Jervell, Luke
and the People of God (Minneapolis 1972) and The Unknown Paul (Minneapolis,
1984). In FS Jervell 1-101 have given some indication of points at which, with great
respect, I differ from him.
xlii COMMENTARY ON ACTS

primitive Christianity tension: division and conflict. No one agreed

with him more firmly than J. B. Lightfoot.20 And there was in Acts a
contrary force which had the effect of throwing a smoke screen over
the battlefield. For Baur’s insight one cannot be too grateful. His
error was not in perceiving tension and tendency; it lay rather in
chronology and in the evaluation of the tendency. Conflicts which
Baur placed in the second century, with the books that displayed
them, belonged in fact to the first; and the tendency that shrouded the
conflicts was not wilful concealment but the—mainly uncon-
scious—effect of an environment marked by consensus rather than
dispute.
We shall return to this theme (see pp. cxii-cxviii); for the present
we have a cue for turning aside to themes that have a conventional,
and proper, place in an Introduction. Who wrote our book, and when,
and where? These questions are relegated to a relatively subordinate
place for the good reason that there is little that can be said about
them.
Traces of the knowledge and use of Acts were listed in 1.34-38.
Acts cannot be proved by quotation to have existed before the second
half of the second century. It may have been known to Justin; it was
rather more probably known to the author of the Epistola Aposto-
lorum. It is tempting to see traces of it within the NT itself, in Mk.
16.9-20 and 2 Tim. 3.11, but these passages are adequately
explained as due to the use of parallel traditions. External evidence
will not take us to a date earlier than c. AD 150, but the lack of earlier
evidence does not mean that the book was not written before that
date. It seems (see above) to have been written at a time of both
inward and outward peace, and there is evidence in remarks about
Roman provincial administration and provincial officers21 that sug-
gests a date within the first century. The period of consensus within
which Acts appears to have been written (see above, pp. xlf.) was
ended by the development of gnosticism, of which signs appear
already in the NT, and indeed in Acts (see 20.20,27,29, 30), and the
last years of Domitian (AD 81-96) were marked by some oppressive
measures. Luke’s first volume, the Third Gospel, dependent as it is
on Mark, can hardly have been written before 80; quite possibly
some years later. So far it may seem probable (though anything but
certain) that Acts was written in the late 80s or early 90s. This dating
is complicated by several factors.

20See e.g. his Galatians p. 374, and the comment by W. G. Kümmel in Das Neue
Testament im 20. Jahrhundert (Stuttgarter B-S 50; Stuttgart, 1970), 73. Lightfoot
corrected Baur’s dates but did not wholly disagree with his understanding of early
Christian history. Μ. D. Goulder (see n. 18) makes a welcome restatement of the
fundamental points.
21See e.g. the notes on 17.6f; 19.31; and much more in chs. 21-28.
 INTRODUCTION xliii

One is constituted by the ‘We’ passages; it will be better to

consider these under the heading of authorship rather than date; see
above, pp. xxvii-xxx, and below, pp. xliv-xlix. If the author was one
of Paul’s travelling companions he cannot have been writing much
later than the 80s. Another arises out of the fact that Acts comes to an
end at the close of Paul’s two year ministry in Rome (28.30, 31).
Why does Luke take the story no further? Why in particular does he
not finish the story of Paul by taking it up to his death—his
martyrdom, if Clement of Rome and later writers are to be believed?
The simplest answer to these questions is that Luke has brought the
story up to date. He tells no more because there is no more to tell.
This is an attractive view. Luke was without doubt an admirer of
Paul; he loved a good story, and the account of a martyrdom that
would have matched Stephen’s is the sort of thing he would have
liked to write. It is an attractive view, but it is not convincing. It runs
into an insuperable obstacle in the fact that Acts presupposes Lk.,
and Lk. presupposes Mk., which was not written till about 70. Lk.
moreover implies the fall of Jerusalem in AD 70 (Lk. 21.20-24). The
later date of Acts is not dependent on our being able to explain why
the book ends where it does, but explanations may be suggested. It is
fair to observe that the promise and commission of 1.8 are fulfilled
by Paul’s preaching in Rome; they would not be more completely
fulfilled by his death. Again, Luke was writing a work of edification;
Paul’s death may not have been an edifying story. He may have been
deserted, even betrayed, by those who should have stood by him. 2
Tim. 4.16 (πάντες με ἐγκατέλιπον) may or may not be historical, but
it is more likely to rest on tradition than on fancy; the story of
desertion might have been repeated (or this might have been Paul’s
only defence). Perhaps there was no dramatic scene; the Romans
locked him up and left him to rot.
It is sometimes argued that the confusion over the dates of Judas
and Theudas (5.36, 37) arose through a misreading of Josephus, Ant.
20.102 (see 1.293-295—in line 18 of p. 295 correcting nothing to
noting). This would make Acts later than the publication of the
Antiquities in AD 93. The argument has little weight.
A more serious matter is Luke’s failure to use or to make any
reference to Paul’s letters, some of which are referred to in 1 Clement
(usually dated c. AD 96). Is it conceivable that Luke, if writing about
AD 90, would not know them? at least have heard of them? It is a
difficult question (see FS Mitton, 2-5), but hardly determinative for
the dating of Acts.
As to the place of writing almost any guess will do. Rome has been
suggested, mainly because the author of the ‘We’ passages accom-
panies Paul to Rome; also because this would furnish an explanation
of the book’s ending—there was no need to write more because the
readers already knew the story from Paul’s arrival to his death. Luke
xliv COMMENTARY ON ACTS

himself is said to have been an Antiochene;22 this may or may not be

true but in any case tells us nothing about where he wrote. Ephesus,
or a Pauline community in Macedonia, Achaea, or Asia Minor have
been suggested.23 But Kümmel’s judgement is correct: ‘Wo der Verf.
schrieb ... lässt sich nicht mehr feststellen.’
The author, who writes prefaces to both gospel (Lk. 1.1-4) and
Acts (l.l),24 reveals himself as a person, and it will be possible on the
basis of these prefaces and the two works as wholes, to make certain
observations about him, but his name is not mentioned. The earliest
sources to name him (Irenaeus, the Anti-Marcionite Prologue, the
Muratorian Canon) call him Luke, and it is clear that they think of
him as the member of the Pauline circle who is referred to at Col.
4.14 (the beloved physician); Philemon 24; 2 Tim. 4.11. If Thornton
is right in his claim, for which there is much to be said, that the
author of the book must be also the author of the source expressed in
the first person plural,25 we must conclude that the author of the book
was one of the Pauline circle, and it would be reasonable to follow
Thornton (see p. xxviii) in the opinion that Luke wrote Acts (and the
Third Gospel). Against this conclusion the only substantial argument
consists in the errors found in the Acts account of Paul. These must
not be exaggerated. The epistles that come from his own hand
represent Paul as precisely the kind of person who appears in Acts: a
constant traveller, a fearless and effective preacher, a founder and
pastor of churches, a Jew who had the gift of ready access to
Gentiles. The geographical and chronological details provided by
Acts fit reasonably well with the data provided by the letters (see
Dates and Places, below). But the differences and problems referred
to above are more than sufficient to cast doubt on the identification of
our author with a Pauline travelling companion. Did such a com-
panion never see Paul at work writing ‘weighty and strong’ letters,
and observe their effect on the recipients? Was he unaware of the
turbulent and complicated relation between Paul and Corinth? Did he
understand Pauline theology so ill that he could picture Paul not only
helping to compose but distributing a decree which told Gentiles that
if they would be saved they need not indeed be circumcised but must

22Jerome, Comm, in Matthaeum, Praefatio 26-39: Tertius Lucas medicus natione

Syrus Antiochensis; Anti-Marcionite Prologue to Luke: Ὲστιν ὁ άγιος Λουκάς
’Αντιοχεύς Σύρος τω γένει, ιατρός τὴν τέχνην.
23Kümmel, Einleitung 154.
24Cadbury’s commentary on Lk. 1.1-4 (Begs. 2.488-510) is a classic and still
worthy of careful study. See also, in addition to commentaries, L. Alexander, The
Preface to Luke's Gospel (SNTSMS 78, Cambridge, 1993), with a valuable bibliogra-
phy (and index).
25The same conclusion was reached by Hamack, Luke the Physician (ET London,
1911); The Acts of the Apostles (ET London, 1909).
 INTRODUCTION xiv

abstain from εἰδωλόθυτα and food that did not satisfy the kashrut
regulations? It is certainly true that in every Christian generation
there have been those who found Paul’s thought difficult and were
misled by their misunderstanding of it (2 Peter 3.16), but this is not
an adequate explanation. It must have been clear even to the dimmest
theologian that Paul was engaged in a running battle with (e.g.) the
Galatian and Corinthian disturbers of the peace. The argument, based
on his vocabulary, that the author was a medical doctor, which would
support the reference to Col. 4.14, was disproved by Cadbury.26 It is
not that Paul does not use terms that are used by Greek medical
writers; he does from time to time do so. It is rather that these terms
are used by other non-medical writers also; they did not constitute a
special technical language, understood and used only within, the
profession. Luke’s use of technical vocabulary suggests, if anything,
that he was not a doctor but a sailor. Apart from tins, his varied style
shows him to be a competent writer but does little or nothing to
identify him. That he was capable of writing good idiomatic Greek is
shown by his preface to his first volume (Lk. 1.1-4) and by a number
of passages in the second, especially in places where he puts words
on the lips of educated men. Such are for example the speech of the
town clerk in Ephesus (19.35-40), that of Tertullus (24.2-8) and
Paul’s reply (24.10-21), Festus’s words to Agrippa (25.24-27);
compare the letter of the tribune Claudius Lysias (23.26-30). The
question of Semitism in his style has been much discussed. The view
that he was in the first half of Acts translating a continuous Aramaic
source had been generally if not universally abandoned. There are
however a few passages where the hypothesis of mistranslation of
Aramaic is plausible, possibly even convincing, but these have more
to do with Luke’s sources than with his own background and
ability.27 Anyone may make a mistake in translating Aramaic, or
perhaps misread the text he is translating; more relevant to an
assessment of Luke as an author is the mildly Semitic atmosphere
that can be detected in many parts of the earlier chapters. This could
be due to the translation of Semitic documents, but is much better
explained as due to the influence of environment and of the Greek of
the LXX. This may have been partly subconscious but was probably
to a great extent conscious; Luke thought it suitable to write of
Palestinian events in a way that evoked the Palestinian background
and, more important, to continue the record of God’s dealings with
his people in language and in a manner that suggested his relation
with the elect people of earlier times.
This appears in the Septuagintalisms which are a not uncommon

26See Cadbury, Style 39-72; also JBL 45 (1926), 190-209.

27See e.g. 1.15, 22; 2.47; 3.16; 4.25; 7.35; 9.28; 10.14, 36f., 40; 12.3.
xlvi COMMENTARY ON ACTS

feature especially of the earlier chapters of Acts.28 These are not so

much a phenomenon of translation as of appropriate, ‘biblical’ style.
It will be necessary here to list only some of the most important turns
of phrase and vocabulary that bear witness to Luke’s awareness of
the OT and of the relation to it of his own work. It is seldom that he
allows himself to write intolerable Greek, but some of his idioms
attest their origin, though often only by relative frequency.
Καἰ ἐγένετο (or ἐγένετο δὲ) suggests the familiar Hebrew the
verb with the waw consecutive that would follow in Hebrew is
represented in various ways. Thus:
4.5: ἐγένετο δέ with accusative and infinitive, συναχθῆναι τούς
άρχοντας ...
9.3: ἐγένετο with accusative and infinitive, αυτόν έγγίζειν
9.32: ἐγένετο δέ with accusative and infinitive, Πέτρον
κατελθεῖν
9.43: ἐγένετο δέ with the infinitive, μεῖναι
10.25: ὡς δὲ ἐγένετο with accusative and the genitive of the
infinitive, τού εἰσελθεῖν τόν Πέτρον
11.26: ἐγένετο with dative, καί, and infinitive, αυτοῖς καί
συναχθήναι
14.1: ἐγένετο δέ with accusative and infinitive, είσελθεῖν αύτούς
16.16: ἐγένετο δέ with accusative and infinitive, παιδισκὴν
ύπαντήσαι
19.1: ἐγένετο δέ with accusative and infinitive, Παύλον
κατελθεῖν (and ἐν τῷ with accusative and infinitive)
21.1: ὡς δέ ἐγένετο with accusative and infinitive, ἀναχθῆναι
ήμάς
21.5: ὅτε δέ ἐγένετο with accusative and infinitive, ἡμᾶς
έξαρτίσαι
28.8: ἐγένετο δέ with accusative and infinitive, τόν πατέρα
κατακεΐσθαι
Hebrew idiom, reflected in the LXX, is recalled by the use of
verbs, often as participles, which are unnecessary to the sense. So for
example ἀναστάς (- άντες) at 1.15 (ἀναστάς Πέτρος); 5.6, 17, 34;
8.27; 9.18, 39; 10.13, 20, 23; 11.7,28; 13.16; 14.20; 15.7; 22.10,16;
23.9.29 άποκριθείς (- θέντες) is used similarly, reflecting the use of
which sometimes means no more than he spoke up, intervened
28See Wilcox; W. K. L. Clarke in Begs. 2.66-105; J. de Zwaan in Begs. 2.30-65; H.
F. D. Sparks in JTS 1 (1950), 16-28; Moulton in Μ. 1.18; Howard in Μ. 2.411-485;
Turner in Μ. 4.45-63; BDR § 4 (‘Für einen feierlichen und würdevollen Stil erschien
die Sprache der LXX als sehr passend.’).
29In some of these passages it may be argued with some force that the participle
plays a significant narrative role and does not necessarily suggest a Semitic back-
ground.
 INTRODUCTION xlvii

in conversation. Thus 4.19; 5.29; 8.24, 34; 9.13; 19.15; 25.9.30

Another word that sometimes seems superfluous is ἅρχεσθαι, espe-
cially the aorist participle: 1.1; 2.4; 11.4 (ἀρξάμενος), 15?; 18.26;
27.35.31
(καὶ) ιδού, recalling the Hebrew is used more frequently
than is usual in Greek: 1.10; 2.7 (οὐχ ἰδoύ[sic]); 5.9, 25, 28; 7.56;
8.27, 36; 9.10, 11; 10.17, 19,21,30; 11.11; 12.7; 13.11 (καὶ νῦν ἰ),
25, 46; 16.1; 20.22 (καί νῦν ἰ.), 25 (καί νυν ἰ); 27.24.
The use of periphrastic tenses is not un-Greek, but there are more
such forms in Acts than would be expected in a piece of ordinary
Greek prose. Howard (Μ. 2.452) lists occurrences of the imperfect of
εἰμί with the present participle as follows: 1.10, 13, 14; 2.2, 5, 42;
8.1, 13, 28; 9.9, 28; 10.24, 30; 11.5; 12.5,6, 20; 14.7; 16.9,12; 18.7;
21.3; 22.19, 20. It will be noticed that occurrences disappear with the
Jewish element in the story; from ch. 22 onwards Paul’s dealings are
mostly with Romans.
The use of ἐν τῷ with the infinitive in a temporal sense probably
reflects the LXX rendering of with the Hebrew infinitive. Howard
(Μ. 2.450f.) lists the following passages: 2.1; 8.6; 9.3; 11.15; 19.1,
rightly excepting 3.26; 4.30, where the meaning is not temporal and
the use can be regarded as classical.32
Causal use of από occurs at least three times in Acts: 11.19; 12.14;
22.11. This recalls the LXX use of ἀπό to render causal
In non-biblical Greek verbs of speaking are normally followed by
a dative of the person addressed. In Hebrew prepositions are
used, and these are represented in the LXX by πρός. In Acts, cf. 1.7;
2.29, 37, 38; 3.12, (25); 4.1, 8, 19, 23, 24; 5.8, 9, 35; 7.3; 8.20, 26;
9.10, 11, 15; 10.28; 11.14, 20; 12.8, 15; 13.15; 15.7, 36; 16.36, 37;
17.15; 18.6, 14; 19.2(bis), (3<a>), (25D); 21.37, 39; 22.8, 10, 21, 25;
23.3, (30); 25.16, 22; 26.1, 14, 28, 31; 28.4.21, 25.33 This is a feature
of Luke’s style maintained in the later chapters. It may be due
ultimately to the influence of the LXX but can hardly be said to be a
conscious attempt to affect a biblical style where the context makes
that style suitable.
Emphasis on the action of a finite verb is often expressed in
Hebrew by the addition of the infinitive absolute of the same verb.
The LXX translators sometimes represented this by the use (with the
finite verb) of a participle of the same, or a kindred, verb or by the
use of the dative of a cognate noun. The former construction occurs
30As well as the participle the finite verb is sometimes similarly used; sometimes the
verb has, or may be held to have, narrative significance.
311.22; 8.35; 10.37; 24.2 may be taken as part of the narrative.
32Howard (Μ. 2.451) quotes Sophocles, Ajax 554, ἐν τω φρονεῖv γάρ μηδέν
ἣδιστος βίος, to show that ‘the temporal use of ἐν τω c. inf. is not impossible Greek.’
There is certainly a temporal element m the thought here.
33The list is provided by Turner in Μ. 4.54. He does notice that the reading in 19.3 is
insecure.
xlviii COMMENTARY ON ACTS

in the NT only in actual quotations of the LXX; the latter occurs in

Acts 2.30; (4.17 in many MSS); 5.28; 23.14. See Μ. 2.443f.; also
Moule (IB 177., with the quotation of Plato, Symposium 195b,
φεύγων φυγῇ).
Another Hebrew characteristic is the use of a noun in the genitive
instead of an adjective; in Acts see (6.11, D 614 lat); 8.23; 9.15.
For further detailed discussion see Μ. 1.18, and references under
Hebraism and Septuagint; Μ. 2.411-485; Μ. 4.45-63; Wilcox: H. F.
D. Sparks, JThS 1 (1950), 16-28; BDR §4.34 The strongest impres-
sion is given by some examples that occur very infrequently. Thus
12.3, προσέθετο συλλαβεῖν τόν Πέτρον, clearly represents the
Hebrew idiom 10.14, ουδέποτε ἔφαγον πᾶν κοινόν recalls
the Hebrew 7.20, αστείος τω θεῷ is probably a Hebrew
form of superlative; cardinal for ordinal numeral at 20.7 is a
Septuagintal Hebraism; predicative εις (if such it is) at 19.27 recalls
(though the context is wholly Greek) predicative the use of εἰ in
indirect questions (1.6; 7.1; 19.2; 21.37; 22.25) may be of Semitic
origin mediated through the LXX.
Luke was a competent writer who, if he was not himself a Jew,35
had made himself familiar with the OT in its Greek form, and was
able, when it suited his purpose, to write in a style that suggested the
continuity of his theme with the OT. In telling the unit stories, or
episodes, of which his book is composed, he had great skill; Peter’s
visit to Cornelius (10.1-48), for example, and Paul’s conversion
(9.1-19), are complicated stories but they are told not only with
clarity and effectiveness but with compelling interest. He shows less
skill in planning the whole, but here allowance must be made for the
fact that he was dependent on sources, often a matter of distant and
imprecise recollection. It is probable that he wrote by the light of
nature rather than under the control of accepted rhetorical princi-
ples—if indeed there were such principles for the writing of
history.36 He gave picture after picture of things he had discovered to
have happened in the generation before his own until he came to the
ministry of Paul which he was able to tell as a connected whole,
though here too the story is constructed in the form of a series of
34See fn. 28.
35That Luke was a Gentile is widely believed, partly on the ground of his traditional
identification with the Gentile physician of Col. 4.14; see however E. E. Ellis, The
Gospel of Luke, NCB, London, 1966.
36Cadbury (Making 224) refers to Cicero, De Oratore 2.15 (61-63): Videtisne,
quantum munus sit oratons historia? haud scio, an flumine orationis, et varietate
maximum, neque tamen earn reperio usquam separatim instructam rhetorum prae-
ceptis. sita sunt enim ante oculos. Namquis nescit, primam esse historiae legem, ne
quid falsi dicere audeat? deinde ne quid veri non audeat? ne qua suspicio gratiae sit in
scribendo? ne qua simultatis? Haec scilicet fundamenta, nota sunt omnibus. The
‘rules’ referred to by W. C. van Unnik in the very important paper (‘Luke’s Second
Book and the Rules of Hellenistic Historiography’, in Kremer, Actes 37-60) relate
much more to the substance than to the forms of history.
 INTRODUCTION xlix

vivid scenes, and we move from one city to another, and then from
one court room to another, and it is seldom that the connecting links
are given in much detail.
If we ask why Luke wrote, secular motivation is not to be set aside
entirely. Luke was born to write this kind of book and it is impossible
to doubt that he enjoyed writing it. The enjoyment was partly a
Christian enjoyment: it was good to enjoy fellowship with Christians
of the past as well as with Christians of the present. But it was only
partly a Christian enjoyment—though there is nothing un-Christian
in getting pleasure from telling a good story. Or in reading one; and
we may suppose that Luke was pleased that readers should enjoy
reading his book.37 That Luke wrote for pleasure, his own and his
readers’, is a motivation not to be overlooked and not to be ashamed
of; but alone it is not adequate as an account of the origin of Acts.
The book has been described as apologetic, and that in various
senses. That it is in the most general sense of the word a piece of
apologetic is undoubtedly true; in this sense apologetic shades into
evangelism, and there can be no question that Luke wished to
commend the Christian faith as true, and as truth that all should, for
their own good as well as simply because it was true, accept. The
reader could not miss the injunction to save himself from the
perverse generation by which he was surrounded (2.40), and the
instruction, ‘Believe in the Lord Jesus and you will be saved’
(16.31); nor could he fail to note the repeated assertion that the
crucified Jesus had been raised by God from the dead, an assertion
which, if true, must constitute the greatest possible claim on his
attention and belief. In this general sense, of statement and com-
mendation, the book (following as it does upon an account of the life
and teaching of Jesus) may be described as apologetic. Here and
there it goes further than this, notably in 14.15-17 and 17.22-31,
where Christian belief is set over against the religion of paganism.
Polytheism and idolatry are refuted, and the Christian doctrine of
God is defended by reference to his forbearance; if his justice is
impugned it may be defended in terms of his patience in the past and
his intention in the future, and presumably the near future, to carry
out a universal judgement.38 Jewish religion is not treated in the same
way as pagan religion,39 because Judaism, resting as it does on the
OT, is true religion. But true Judaism, that is, Judaism as understood
by Christians, is defended over against Judaism which so far
misunderstands itself as to resist and persecute Christians. Paul
repeatedly insists that he is simply a Jew who sees the fulfilment of

37R. I. Pervo, Profit with Delight: The Literary Genre of the Acts of the Apostles
(Philadelphia, 1987).
38Luke has successors in Justin and other apologists of the second century.
39See below, pp. xcviif.
1 COMMENTARY ON ACTS

Judaism and the OT in the life, death, and resurrection of Jesus (e.g.
24.14; 26.22f.). It follows that there is both an affirmation of Judaism
and an apologetic directed against its abuse and misinterpretation
(e.g. 13.45-47).
It is not incorrect to describe Acts as a work of apologetic, but this
is not an adequate description. In particular the view, sometimes
held, that Acts is an apology addressed to the Roman judiciary and
intended to show that Christianity is a movement which right-
thinking Roman officials will view with tolerance is unconvincing. It
is true that, according to Acts, Roman courts, when correctly
informed, show no disposition to persecute Christians: the proconsul
in Cyprus, Sergius Paulus, actually becomes a Christian (13.12); the
magistrates in Philippi, when they learn that Paul and Silas are
Roman citizens, immediately release them (16.38, 39); the proconsul
Gallio in Corinth is not interested in the case brought against Paul
(18.15); the town clerk in Ephesus defends him (19.37,38); and from
the time when Paul is rescued by Roman soldiers from the Temple
mob (21.31, 32) the Roman tribune and the provincial governors
treat him decently and protect him from Jewish violence up to the
moment when Agrippa and Festus agree, ‘He could have been
released if he had not appealed to Caesar’ (26.32). Luke would no
doubt have been pleased to hear that a Christian on trial had been
able to use some of these precedents, but the book as a whole is not to
be thought of as serving such a purpose; no Roman court could be
expected to wade through so much Jewish religious nonsense in
order to find half-a-dozen fragments of legally significant material.
The same argument proves even more conclusively that the book was
not written as a brief for the defence at Paul’s trial; what would be
the use of chs. 1-12 for such a purpose?
The theme of apologetic has been inverted in the suggestion40 that
Acts was an apology addressed to the church on behalf of the
Empire. Christians (such would be its theme) have no need to fear
and ought not to oppose the Roman administration. When Roman
officials are informed of the facts they will give a fair trial and will
not believe calumnies. The church may settle down to live in
confidence whatever time is left before the End. Again, there is
something to be said for this as a subordinate interest of Luke’s, but it
will not account for more than a relatively small part of the book.
An important observation to be made here is that Luke had
provided his church (wherever that may have been) with its NT. It
probably had no more. We cannot suppose that Luke wrote his
gospel with the notion that it should be published in one of four
parallel columns in a Synopsis. He used Mk and collected what other

40See P. W. Walaskay, And so we came to Rome, SNTSMS 49 (Cambridge, 1983).

 INTRODUCTION li

material he could find;41 he included what he thought should be

included and this meant that Mk was no longer needed and could fall
out of use. The Third Gospel would provide what Luke’s fellow
Christians (locally) needed to know about the life and teaching of
Jesus. The second volume, Acts, correspondingly contained what the
church (in the 80s or early 90s) needed to know about the apostles
and their teaching. The disuse of Paul’s epistles in Acts is a familiar
and often discussed fact. Whether this was due to ignorance or to
deliberate omission,42 Acts was there to serve as Apostle supplement-
ing Gospel. It was what Luke’s contemporaries needed to know
about the apostles—and Paul. This leads to two observations.
The first was made by W. C. van Unnik,43 who wrote of Acts as the
confirmation of the gospel. The record of the life and expansion of
the church confirms that the claims made in the gospel for the person
of Jesus and the effect of his work were true. The two books thus
form a unit, and the purpose of the second must be viewed in the light
of the first.
This presupposes (what Luke no doubt took for granted) that the
earliest period of Christianity was an exemplary and classical period;
this conviction supplies perhaps the strongest because the most
practical reason for the writing of Acts. Luke wished to hold up
before his readers a set of Christian ideals which would show them
what their own Christian life should be and at the same time supply
them with a strong motivation for following the example. He
probably wrote primarily for church leaders, who are given espe-
cially in Paul’s Miletus speech (20.18-35) a clear account of their
pastoral responsibilities. The nearly contemporary (perhaps some-
what later) Pastoral Epistles encouraged the ministers to whom they
are addressed to preach: Preach the word, keep at it in season, out of
season, reprove, rebuke, exhort (2 Tim. 4.2). Christian leaders
towards the end of the century might reply to such exhortations, Yes,
but what must we say? Luke replies, You will find no better
examples, no clearer instruction, than in the sermons of Peter and
Paul that I am providing.44 He might similarly claim that he had
provided an outline of the kind of behaviour to be expected of the
Christian rank and file: They continued steadfastly in the teaching of
the apostles and the fellowship, in the breaking of bread and the
prayers (2.42).
Beyond these matters one may inquire whether there were any

41No one has in the last fifty years pointed out to me any reason why I should
abandon the view of Luke’s sources I sketched briefly in ExpT 54 (1943), 320-23; see
FS Smit Sibinga, 96.
42See FS Mitton, 2-5.
43NovT 4 (1960-1), 26-59, ‘The “Book of Acts”—the Confirmation of the
Gospel’; see also my exploration of the converse of this in FS Neirynck 1451-66.
44Cf. Dibelius 165: ‘This is how the gospel is preached and ought to be preached!’.
lii COMMENTARY ON ACTS

specific theological truths that Luke wished to communicate. For his

theology as a whole and in some detail see below, pp. lxxxii-cx. One
theological point merges with a historical; it is not mentioned by
Luke, probably because it seemed to him part of the nature of things,
but it was fundamental to his work. For the view that Jesus foretold
his suffering and that after his suffering he would be vindicated by
God, but did not in this vindication differentiate between resurrection
and a coming of the Son of man in glory I may refer to Jesus and the
Gospel Tradition (London, 1967), pp. 77-83. After his crucifixion
Jesus appeared to his disciples, raised from death; but there was no
coming with the clouds of heaven. One may guess with some
probability, but cannot prove, that the disciples, compelled by events
to differentiate where Jesus himself had not done so, believed that the
second act, which would complete the first, would follow quickly. It
did not do so. Had it done so, there would have been no church
history. It was the ‘delay of the parousia’ that made it possible for the
church to have a history, and for the church’s historian to write it. It
does not seem that the delay, though it must have puzzled many, led
to a crisis of belief. Luke did not write to assuage apocalyptic
enthusiasm, or to comfort those who were disappointed in their
hopes. When he wrote, some measure of delay had been accepted,
and Luke himself sets out what were probably widely acknowledged
terms of acceptance. After the crucifixion, Jesus had been raised up
by God; he had ascended into heaven; he had sent to his followers
(and continued to send as more followers were added) the gift of the
Holy Spirit. Empowered and directed by the Spirit, the believers
would act as witnesses to Jesus till their testimony had reached the
end of the earth (1.8). This goal Luke probably saw as representa-
tively achieved in the arrival of Paul at Rome, but there is no hint in
his writing that the End could now be expected at any moment. He
knew that some thirty years had elapsed between Paul’s arrival in
Rome and the point at which he was writing; and if God could wait
thirty years, he could no doubt wait a good deal longer. This was
probably the accepted position; there was no need to argue for it.
Luke does however (without saying explicitly that he is doing so)
show how the various features of Christian truth cohere with one
another. This was something that a gospel alone could not do; his
two-volume work was needed. There is an end, including a judge-
ment, to look forward to, and not only Areopagites but all must
prepare for it by repentance. For those who repent there is forgive-
ness and the gift of the Spirit, manifested in such gifts as prophecy
and speaking with tongues. The believers live in unity and godly
love, and bear witness to Jesus.
This is in fact Luke’s theology and he does not see fit to expand it,
though there are many points that invite expansion. Who was Jesus,
who plays such an important part in the history of God’s dealings
 INTRODUCTION liii

with men? How did his death effect forgiveness and release from
sin? How is the Holy Spirit related to Father and Son? What is the
proper constitution of the church, and what happens when it
immerses converts in water, and when its members assemble ‘to
break bread’? Luke gives only the barest hints towards answers to
these questions. The basic theology of Judaism he accepts but
scarcely develops. He believes as the apostles did, and does not seek
to go beyond their pronouncements. To do so indeed would cut
across his purpose, which is to record what they said and did, not to
point out what they ought to have said and done. On the theology of
Acts see further below, pp. lxxxii-cx.
The lessons Luke presses upon his readers are not speculative but
practical. The Christian faith is the truth, the truth of God: magna est,
et praevalebit. There is therefore no need for Christians to fear, even
though they are attacked on every side as Paul was, even though, like
Stephen and James, they must die for their faith. God does not fail his
servants whether in life or in death, and his ability to preserve them,
even in acute danger, proves that the truth of the Gospel will not be
suppressed but will eventually triumph. But woe to those who, like
Ananias and Sapphira, trifle with truth and are less than sincere in
their adherence to the community and its principles. Their fault is not
so much that they have attempted to deceive the rulers of the
community; they have lied not to men but to God. The apostles do
indeed have their own dignity, but this is rather because they serve as
an indispensable link between Jesus and the consequences that
flowed from his work than because they occupy positions of
authority and must therefore be obeyed. Luke has not forgotten Lk.
22.24-27; and indeed he knows (or at least tells) hardly anything
about the Twelve. It is right that those who are appointed to share
their ministry (as elders) should be unpaid. Let them rather work for
their living and be ministers in their spare time; they will then be
independent and able to minister to the needs of others.45
All that has been said here about the characteristics, interests, and
intentions of Luke as an author bear on the question of the historical
value of his book. For details on a number of specific points see the
appended note on Dates and Places. On such matters, Luke, on the
whole does fairly well. He is not unfamiliar with the roads and
shipping routes that he describes; his references to secular history are
usually in reasonably close agreement with the facts as otherwise
known. He comes down badly over Judas and Theudas (5.36, 37),
but is probably not far out in his references to Claudius, Sergius
Paulus, Agrippa I and II, Gallio, Felix, and Festus. Oddly it is on
Christian matters that he is most open to criticism. His account of
Hebrews and Hellenists, of Paul and the Council, is distorted by the

45On the supposed Frühkatholizismus of Acts see below, pp. xciii-xcvii.

liv COMMENTARY ON ACTS

refracting medium of continuing church life through which he views

the past.

Dates and Places

None of the events in Acts can be dated precisely; some can be given
a relative dating and placed in chronological order, and a few can be
given an approximate absolute date. It will be best first to run through
t e main events which could be given dates—if we had the necessary
h
information.46
Acts begins with reference to resurrection appearances of Jesus, his
Ascension, and the gift of the Holy Spirit on the Day of Pentecost.
These are represented by Luke as historical events which therefore
had dates, but these are only to be obtained (if obtained at all) by
reckoning forwards from events in the ministry of Jesus or backwards
from events recorded in Acts. There is nothing in the narratives
themselves to suggest a date, but of course Luke thinks of them as
preceding all the other events in the book. In 4.6 Luke refers to Annas
the High Priest and to Caiaphas.47 In fact Caiaphas was High Priest
from AD 18 to 36, Annas having been deposed in AD 15. We have in
the text no means of fixing a date more precise than 18-36.
At 5.34 Gamaliel is said to intervene in the Sanhedrin’s discus-
sion. This was Rabban Gamaliel I, according to Acts 22.3 the teacher
of Paul. He lived in the first half of the first century. No precise date
can be inferred from this reference. Gamaliel however is said in vv.
36, 37 to refer to Theudas and Judas, in that order. It is very probable
that the revolt of Theudas took place in about AD 44, that of Judas in
AD 6. If Luke’s account is taken as it stands Gamaliel must have
been speaking after AD 44 (and into the bargain have inverted the
order of the two revolts). It is virtually impossible to believe that
Gamaliel has been reported correctly; no inference can be drawn
regarding the date of the events in Acts 5.
The reference to Simon the Magus (8.9-24) provides no date.
Even if second-century Simonianism originated with the Magus this
does not tell us the time at which his contact with Philip, Peter, and
John took place.
The narrative of Saul’s conversion (9.1-19) does not refer to a
date. The only evidence by which the event can be dated is given by
the reference to it in Gal. 1.15, 16.48 Here the conversion is followed
by a journey into Arabia; how long he stayed in Arabia and, on his
return, in Damascus Paul does not say. Then after three years (from
the conversion, from the return to Damascus?) Paul went up to
46Details will be found at the appropriate points in the Commentary; no attempt is
made in this Introduction to cite evidence in full.
47The name of Annas appears also at 5.17, but only in p mae.
48No help can be drawn from 1 Cor. 15 or Phil. 3.
 INTRODUCTION lv

Jerusalem and stayed there only a fortnight before going to Syria and
Cilicia. Then after fourteen years (from the conversion, from the
previous visit?) he went back to Jerusalem for the so-called Apos-
tolic Council. If we knew the date of the Council we could work back
from this,49 but even so there would be uncertainties. Three plus
fourteen is seventeen but years were commonly counted inclusively,
so that three might be (part of a year + a year + part of a year);
perhaps little more than one and a half years. Similarly fourteen years
might be twelve and a half years. And we do not know whether the
three must be added to the fourteen or counted within it. The total
span might be anywhere between twelve and a half and seventeen. If
the Council is to be dated in 47,50 the conversion cannot be later than
c. 35; it might be considerably earlier.
In 11.28 we read of a famine ἐφ’ ὅλην τὴν οικουμένην which
happened ἐπὶ Κλαυδίου. Claudius ruled from 41 to 54; there is reason
to think that the famine may have occurred in 47, though Hemer (165)
prefers 45-46. The bearing of this date on the chronology of the
Christian movement will depend on conclusions reached with regard
to the Council in Acts 15. Was there one Council or were there two,
one on the question of circumcision, one dealing with the famine?
What was the point of Gal. 2.10? T was eager to take up this sugges-
tion?’ or, ‘I had already been keenly pursuing relief work?’ There is
much to be said for a Council in 46 or 47; see below, p. lxi.
Most of Acts 12 is concerned directly or indirectly with a person
described by Luke as King Herod; he was of the Herod family but his
name was in fact Julius Agrippa (= Agrippa I). Some details of his
career are given in I.573f. There is no means of dating with certainty
the death of James or the imprisonment of Peter, but it is to be noted
that Peter’s escape took place at or before Passover (12.4). Herod’s
feud with the Tyrians and Sidonians cannot be dated but his death
very probably happened in 44, on or near 5 March, possibly on or
near 1 August. If he died on 5 March Peter’s Passover escape must
have happened in 43; if on 1 August, in 44. These events therefore
belong to a time before the famine of 11.28; Luke is not writing in
chronological order.
The next possible date arises in 13.7 with the reference to the
proconsul of Cyprus, Sergius Paulus. Unfortunately though there is
epigraphical evidence that may refer to this man the reference is
uncertain and dates (apart from the name of the reigning emperor,
which may be [Ga]ius or [Claud]ius) are not given. There is no
means of dating the ‘first missionary journey’; see below on the
question whether it preceded or followed the Council.
In Acts the Council follows the journey, in ch. 15. Nothing in this

49See p. lxi.
50See p. lxi.
lvi COMMENTARY ON ACTS

chapter supplies a date apart from the words ascribed to Peter, which
look back to the Cornelius episode of ch. 10, and the reference made
by Barnabas and Saul to the miracles that had accompanied their
missionary work. These references would be useful if we could date
the events mentioned; this we cannot do. The relative chronology of
the Council is of great importance but it cannot be even tentatively
established until all possible points of absolute chronology have been
considered and some questions of topography considered too. See
below, pp. Ivii-lxi.
Luke relates Paul’s visit to Corinth (18.1-18) with secular events
that can be dated with some probability. When Paul reached Corinth
he found there the Jewish couple Aquila and Priscilla who had
recently arrived because Claudius had expelled the Jews from Rome.
The date of this expulsion is discussed on pp. 861f. The view taken in
the Commentary is that AD 49 may be accepted with very consider-
able confidence; this has the advantage of being consistent with
another date given by the account of Paul’s ministry in Corinth; see
below. The alternative date (see p. 862) for the expulsion is 41; if this
is accepted we are likely to conclude that Paul visited Corinth first in
AD 41 and again in or about AD 50; see Lüdemann (17-20).
In 18.11 Paul is said to have stayed in Corinth one year and six
months; in 18.12 there is a reference to Gallio, proconsul of Achaea;
in 18.18 Paul is said to have stayed on in Corinth ημέρας ἱκανάς.
This is vague, and we do not know whether or not these days were
included in the eighteen months of v. 11, or how many ἱκανάς
means. The date of Gallio’s proconsulship can be determined with
some accuracy ; see p. 871, where, on the basis of an inscription, it is
concluded that Gallio ‘probably became proconsul of Achaea in 51
(summer), possibly, if, as was less usual, he held office for two years,
in 50’. It is further shown to be reasonable, though by no means
certain, that the encounter between Paul and Gallio took place
towards the end of Paul’s stay in Corinth and early in Gallio’s
proconsulship. Accepting this view (with due reservation), and
naming months for simplicity and brevity of statement, we may
suggest that Paul was brought before Gallio in September 51, having
arrived in Corinth in March 50. This would be consistent with the
recent arrival of Aquila and Priscilla after they, as Jews, had been
expelled from Rome in 49. These dates are not certain, but they are
more nearly certain than anything else in the life of Paul.
The next two references give no clue to absolute chronology and
may be simply mentioned without discussion. In 19.8 Paul spends his
first three months in Ephesus discoursing in the synagogue. In 19.10,
after moving from the synagogue, he teaches for two years in the
school of Tyrannus. A total stay of two years and three months is
consistent with 20.31 where Paul recalls to the Ephesian elders that
he had spent three years (τριετίαν) among them. Using inclusive
 INTRODUCTION lvii

reckoning two and a quarter years would cover three years. These
references are followed by another which tells only of extent of time.
Paul spent three months in Greece (20.3). From Greece he went to
Philippi whence he sailed μετά τὰς ημέρας τῶν ἀζύμων to Troas,
the voyage taking five days (20.6). In Troas the travellers spent seven
days (20.6) and with the local church met ‘to break bread’ ἐν τῇ μια
των σαββάτων; Paul was to leave next day (20.7, 13). From these
data and from the (probable) date of Passover it has been argued that
the events fell in AD 57. This would be an interesting contribution to
chronology; for reasons why it should not be accepted see p. 952.
Ananias, referred to as High Priest at 23.2; 24.1, was in office c.
AD 47-59. There is no doubt that the events of these chapters fell
within this period, so that the reference to Ananias provides no more
than confirmatory evidence.
At 24.27 the succession of Festus to Felix is mentioned. The date
of Festus’ accession is discussed on pp. 1116-18. The matter is
contentious. The view taken here is that though some would place the
succession as late as AD 60 the balance of probability is that it
occurred much earlier, perhaps as early as 55. This fits well (see
below, p. lxi) with the chronology of Acts (and of Paul) provided that
the two-year period (διετία) is taken to refer not to Paul’s imprison-
ment but to Felix’s term of office.
King Agrippa II (son of Julius Agrippa I, 12.1-23), accompanied
by his sister Bernice, appears with Festus in 25.13 – 26.32. He seems
to have become king not immediately upon his father’s death but in
AD 52, or a little later. He died probably in AD 92/3 (see NS
1.472-83). For a short time (c. 64-66) Bernice was married to King
Polemon of Cilicia. These dates cause no problems but add no
precision to the chronology of Acts.
Acts 27.9 refers to the Fast, that is, the Day of Atonement (10
Tishri). This was already past; sailing was therefore dangerous. This
has been held to mean that in the year of Paul’s voyage Tishri 10 must
have fallen late; it was late (on 5 October) in AD 59, later than in 57,
58,60,61,62. From this it has been inferred that the voyage took place
in 59. The inference is very insecure; see p. 1188, and on 28.11.
The two verses (27.9; 28.11) when taken together give rise to a
problem. Even if it is assumed that the ship did not leave Fair Havens
till after 5 October one would suppose that Malta was reached by the
end of October; three months bring us to the end of January, a very
unlikely date for resuming the voyage. One can only say that this
note of time makes one view it and all the other notes of time relating
to the voyage with a measure of suspicion.
According to Acts 28.30, Paul, under guard in Rome, spent two
years lodging at his own expense. The two years would be chrono-
logically helpful if we knew when they began or ended; we do not.
See further on 28.30, 31.
lviii COMMENTARY ON ACTS

Before anything further can be said about chronology some of the

routes taken by Christians, and especially by Paul, must be con-
sidered. Very few of these give rise to any problems.
The first journey in Acts is that undertaken by Philip in ch. 8. See
1.402, 423f. There is a textual ambiguity in 8.5, and we do not know
where Philip began the journey of 8.26. There is a straightforward
road connection from Azotus to Caesarea (8.40).
There is no difficulty in Paul’s journey from Jerusalem to Dam-
ascus (9.1, 2); Acts does not mention his visit to Arabia (Gal. 1.17);
he would naturally retrace his steps when he returned to Jerusalem
(9.26). Peter would take Lydda (9.32) on his way to Joppa (9.36);
Caesarea lay further north on the coast. Phoenicia, Cyprus, Antioch
(11.19) are easily identified on a map; so is Tarsus, in Cilicia (11.25).
There are no complications in the route followed by Paul and
Barnabas in chs. 13 and 14, or in their visit to Jerusalem for the
Council and subsequent return to Antioch.
Paul, accompanied by Silas, returned to Derbe and Lystra of the
first journey not by sea but through Syria and Cilicia ( 15.41; 16.1). It is
at this point that a real problem arises, in 16.6-10. This must be for the
moment deferred and considered with 18.22, 23. Many other topo-
graphical references follow in Acts and little would be gained by
adding here a list of place names. For the most part there is no
difficulty, and information is given in the notes. There is a problem of
text and interpretation at 17.14, sufficiently discussed in the commen-
tary, and the meaning of τα ἀνωτερικὰ μέρη (19.1) is disputed. This
question will be raised below. Another difficulty arises at 21.15-17;
this is discussed in the commentary, as is the identification of the
island on which Paul and his companions were wrecked (28.1).
We have to consider 16.6-10; 18.22, 23; 19.1. For many details
see the notes. It is important here to take the evidence of Galatians
fully into account. The fundamental question is simple to state: What
is meant by Galatia (and the Galatians)? In the epistle (Gal. 1.1; cf.
3.1) it seems that the reference is most probably to the territory in
north central Asia Minor inhabited since the third century BC by
Gallic tribes rather than to the territory added to this when the whole
was made in 25 BC into a Roman province.51 One objection
commonly made to this view is that nowhere in Acts is Paul said to
have established churches in this northerly area, whereas there is a
detailed account of his work in the additional, southern, part of the
province. This was visited in the course of the ‘first missionary
journey’ in Acts 14.1-25 and revisited at the beginning of the
second, 16.1-5. If ‘Galatia’ in the epistle is taken to refer to this,
southerly, area, the epistle could have been written at any time after
ch. 14; if it were written after 16.1-5, Gal, 4.13 could be taken to

51The matter is disputed; there is a good summary in Kümmel, Einleitung 258-65.

 INTRODUCTION lix

refer to the former of two visits.52 If the epistle was written soon after
ch. 14, either Acts 15 cannot be identified with Gal. 2.1-10 or the
sequence of chs. 13, 14,15 is mistaken. The other view of Paul’s use
of Galatia, Galatians, however, leaves us with the problem of
finding a mission to Galatia, and we must consider Acts 16.6 (τήν
Φρυγίαν καί Γαλατικήν χώραν; cf. 18.23, τὴν Γαλατικὴν χώραν
καί Φρυγίαν). The meaning of these phrases is discussed in the
commentary, together with the meaning of the aorist participle
κωλυθέντες. The first observation—simple but very important—to
make is that before 16.6 is reached Paul has already passed through
southern, provincial Galatia (16.1-5). Galatic territory must be
different from this. Taking κωλυθέντες as the aorist participle it is
one must suppose that Paul passed through Phrygia and Galatic
territory because he had been forbidden to turn westward into the
province of Asia. He must therefore have gone northwards until he
was compelled to travel towards Troas along the frontier between
Bithynia (which he was not allowed to enter53) and Mysia, which he
passed through or beside (παρελθόντες). Between southern provin-
cial Galatia and this Bithynian-Mysian stretch lay τήν Φρυγίαν καί
Γαλατικήν χώραν. It is a vague expression, and probably reflects the
fact that Luke had no first-hand or even second-hand information
about it and what happened there.54 It cannot be said that because no
missionary work is described in this region, therefore none hap-
pened. 18.23 (τήν Γαλατικήν χώραν καί Φρυγίαν) is important,
first, because the reversed order reinforces the sense of vagueness
and lack of first-hand knowledge, and secondly because it proves that
there had been missionary work in the area, since on his new journey
Paul was ἐπιστηρίζων πάντας τούς μαθητάς.55 The obscure refer-
ence in 19.1 to the ἀνωτερικά μέρη through which Paul passed on
his way to Ephesus probably explains Paul’s lack of personal contact
with the churches of the Lycus valley. The curious route of 16.6-10
may be a rationalization—or spiritualization—of Paul’s surprising
failure to make at once for the great city of Ephesus.56
We return to chronology in the light of the questions raised in the
last paragraph. Acts 13 and 14 are a block of Antiochene tradition.

52But this is not necessarily implied by the words τό πρότερον.

53ούκ εἴασεν αυτούς τὸ πνεύμα Ἱησού. Some think that this means only that
missionary activity was forbidden; the travellers were allowed to make their way
through Bithynia but not to preach. This is not a natural interpretation of
πορευθῆναι.
54The We-passages begin at Troas, 16.10.
55This is fully consistent with Gal. 1.2; if there were disciples there were
ἐκκληοίαι.
56On peut voir alors que la raison pour laquelle l'itinéraire de Paul est si difficile à
reconstituer provient du fait que Act 16, 6-8 est une composition artificielle de Luc à
partir d'éléments puisés à ses sources et qu’il s’efforce d’harmoniser’ (Taylor 234f.).
Details follow on pp. 759-62.
lx COMMENTARY ON ACTS

There are probably some inserted pieces but on the whole they
present a complete missionary circuit from Antioch back to Antioch,
with a strong claim to historicity. This however does not guarantee
the point at which the circuit is inserted in the general outline of Acts.
Galatians gives an account of two visits to Jerusalem made by Paul,
in 1.18f. and in 2.1-10; in the corresponding period in Acts there are
three, in 9.26-30, in 11.30 (12.25), and in 15.1-29.57 In Galatians the
peace of the church is disturbed by a visit of envoys from, or
purporting to come from, James (of Jerusalem): 2.12. This takes
place after the Jerusalem meeting. In Acts the disturbance from
Jerusalem takes place before the Council: 15.1, 2, 24.
Two points are hard to doubt: that Paul in Galatians did not omit
any of his visits to Jerusalem—to have done so would have been as
foolish as dishonest; and that Galatians 2 corresponds to Acts
15—the persons are the same and the question is the same. There is
no better explanation than that Luke has made two visits out of one.
In Galatians 2 there is a meeting at which Paul’s law-free Gospel was
confirmed with the proviso that he should ‘remember the poor’,
which, he said, he was and had been eager (ἐσπούδασα) to do.
Luke, we may suggest, had a tradition of a Council at which the
Gospel without circumcision was confirmed, and another tradition
according to which Paul and Barnabas brought relief for the poor. He
concluded, wrongly, that two purposes meant two visits, two meet-
ings.
If this suggestion is accepted we ask, When did the one visit take
place, at 11.30, 15.1, 2, or at some other time, and what was its
relation to the ‘first missionary journey’? If the one Jerusalem visit is
put at the time of the famine visit (11.30) the first journey followed it.
If we choose the alternative date the order is reversed. The epistles
do not suggest that the terms of the mission had all been settled
before the mission began; the reverse is more probable. Events had
thrown up unforeseen problems; what in fact was to be done with
Gentile converts who now existed?58 If the interpretation of the
word Galatia given above is correct the epistle was not written
till after Acts 16.1-6—perhaps not till some time after, for time
must be allowed for the difficulty to arise in Galatia: ἐτρέχετε
καλώς τίς ὑμάς ἐνέκοψεν; (Gal. 5.7). If undue weight is laid upon
57It may be that Acts 18.22 also refers to a visit to Jerusalem.
58Lake (Begs. 5, 203f.) attempts to solve the problem of the relation between
Galatians and Acts by means of source criticism; this is not convincing (though his
exposition—5.195-212—is a masterpiece of clarity and should never be overlooked).
See also Taylor (225): ‘Notons enfin qu’il serait futile [in view of the use of sources]
d’essayer de dater le ‘Concile de Jérusalem’ en calculant le temps que Paul aurait mis
à faire son premier voyage missionaire, ou son second, et en tenant compte de la date
de la famine sous Claude qui l’a précédé, ou du proconsulat de Gallion qui l’a suivi, et
ensuite d’essayer de concilier la chronologie ainsi obtenue avec celle donnée par Paul
dans les Galates.’
 INTRODUCTION lxi

Gal. 4.13 and the verse is taken to refer to a second visit to Galatia,
the letter might have been written after 18.23; it has even been
suggested that Gal. 2.1-10 will refer to the (supposed) Jerusalem
visit of 18.22. The letter however was written at a time when Paul
was unable to visit Galatia (Gal. 4.20)—making perhaps his rapid
journey east without time even to linger in Ephesus. In the circum-
stances he may have lost hope of completing his collection in Galatia
(1 Cor. 16.1).
If the Council was in 46 or 47 Paul could easily have reached
Corinth by March 50 and left in September 51. The next date that can
be even argued about59 is that at which Festus succeeded Felix (see
above, p. lvii). If this was 55, and if the διετία refers to Felix’s
governorship, Paul’s movements fit neatly into the interval. He could
‘sail, by one of the last boats of the summer, to Syria (Acts 18.21f.).
How long he spent in Antioch (Acts 18.23) we do not know;
probably he renewed his travels with the return of favourable
weather in the spring of the next year—52. His journey through ‘the
region of Galatia and Phrygia’ (Acts 18.23) and through the ‘upper
country’ (Acts 19.1) might well last most of the following summer;
he would reach Ephesus (Acts 19.1) in the autumn. The length of his
stay there is given by Acts 19.8 (three months) and 19.10 (two years),
confirmed by 20.31; the Pentecost therefore that he was anxious to
keep in Jerusalem will have been that of 55, and his three months in
Greece (Acts 20.5) must have come to an end a few weeks earlier.’60
He will have left for Rome not long after the arrival of Festus and
reached the city early in the next year—in 56, or as many years later
as the date of Festus’ accession requires.61

Bibliography
P. J. Achtemeier, CBQ 48 (1986), 1-26.
L. Alexander, The Preface to Luke's Gospel, SNTSMS 78, Cambridge
1993.
C. K. Barrett, FS Neirynck, 1451-66.
C. K. Barrett, FS Jervell, 1-10.
E. Bickermann, Studies in Jewish and Christian History, Leiden 1986.
F. F. Bruce, ANRW 2.25.3 (1985), 2570-603.
H. J. Cadbury, JBL 45 (1926), 190-209.

59I exclude the attempt based on 20.7 (above, p. lvii).

Romans 4. If a later date is taken for the accession of Festus the διετία may be
60
taken to apply to Paul’s imprisonment or some time added to his travels—or both.
61For a different chronology see the summary in Taylor (333-5).
Ixii COMMENTARY ON ACTS

H. J. Cadbury, The Making of Luke-Acts, London 1958 (New York 1927).

S. Dockx, NovT 13 (1971), 261-304.
F. V. Filson, Three Crucial Decades, London 1964.
R. Μ. Grant, VigCh 46 (1992), 105-111.
T. Holtz, ThLZ 100 (1975), 321-32.
J. Jervell, Luke and the People of God, Minneapolis 1972.
J Jervell, The Unknown Paul, Minneapolis 1984.
E. Larsson, NTS 33 (1987), 205-25.
D. L. Mealand, ZNW 82 (1991), 42-66.
J. Murphy-O’Connor, Paul, a Critical Life, Oxford 1996.
R. I. Pervo, Profit with Delight: the Literary Genre of the Acts of the
Apostles, Philadelphia 1983.
E. Plümacher, Lukas als hellenistischer Schriftsteller, Göttingen 1972.
E. Plümacher, in Kremer Actes 457-66.
P. Pokorny, ZNW 64 (1973), 233-44.
W. Μ. Ramsay, The Church in the Roman Empire before A.D. 170, London,
New York and Toronto, 10th ed, n.d.
W. Μ. Ramsay, The Cities of St Paul, London 1907.
W. Μ. Ramsay, St Paul the Traveller and Roman Citizen, London 101908.
H. F. D. Sparks, JTS 1 (1950), 16-28.
W. Stegemann, Zwischen Synagoge und Obrigkeit. Zur historischen Situa-
tion der lukanischen Christen, FRLANT 152, Göttingen 1992.
A. Suhl, NTS 38 (1992), 430-47.
G. Theissen, NTS 21 (1975), 23-39.
E. Trocmé, Le 'Livre des Actes' et l'Histoire, Paris 1957.
W. C. van Unnik, NovT 4 (1961), 25-59.
W. C. van Unnik, in Kremer, Actes, 37-60.
P. W. Walaskay, And so we came to Rome, SNTSMS 49, Cambridge 1983.

A number of the works mentioned under § IV are relevant here also.

 INTRODUCTION lxiii

IV. ACTS IN HISTORY

I have described Acts as the history of the church in a time of conflict

written in a time of consensus (see pp. xlf.). It was the consensus
that provided Luke’s frame of reference and it affected the way in
which he wrote, making it impossible for him to understand fully,
and so to describe accurately, the story of a church almost tom in two
by dissension.1 The Christian ship had sailed out of the storm into
calmer waters, and Luke found it easier to picture the storms of the
Mediterranean and Adriatic (which he may well have experienced)
than the ecclesiastical troubles of the 40s and 50s. Were it not for a
few verses one might suppose that Luke and his colleagues had come
to believe that peace would last for ever. But there is enough to show
that Luke knew better; indeed, the troubles had already begun. Luke
knew that the prophecy he had put on Paul’s lips was beginning to be
fulfilled: T know that after my departure grievous wolves will come
in among you, not sparing the flock, and out of your own number
there will rise up men speaking perverse things so as to draw away
after themselves disciples’ (20.29f.). This prediction—or vaticinium
post eventum—corresponds with surprising exactness to events that
are described elsewhere in the NT. Faced by the failure of the
Christian mission (for ‘the world does not listen to us’, 1 Jn 4.6)
some members of the Johannine community had ‘gone out’ into the
world (1 Jn 2.9); they had learned to speak out of the world’s
vocabulary (έκ τοῦ κόσμου), and accordingly the world paid atten-
tion to them (1 Jn 4.5). But this meant, John said, that they were
teaching false doctrine (Luke’s διεστραμμένα), and they had ceased
to practise love (1 Jn 2.22, 23, 26; 3.7-10, 14, 15). Evidently they
were drawing disciples after them. It is doubtful whether John the
epistolographer very greatly loved these whom he accused of lack of
love. He would not pray for them (1 Jn 5.16), and the church was
split into mutually exclusive groups: Diotrephes excommunicated
the Elder and his church members (3 Jn 9, 10) and the Elder
excommunicated Diotrephes and those who thought as he did (2 Jn
9-11).2 The force, the doctrine, that split the church was gnosticism;
1Inevitably one refers to F. C. Baur, especially Die Christuspartei in der kor-
inthischen Gemeinde (1831); Über Zweck und Veranlassung des Römerbriefs (1836);
and Paulus (‘1845; 21866); but now see also Μ. D. Goulder, A Tale of two Missions
(1994). Lightfoot’s Excursus on ‘St Paul and the Three’ (pp. 292-374 in his
Galatians) is also important.
2See my Essays on John (1982); some, notably R. E. Brown (The Community of the
Beloved Disciple, 1979), have taken this into much more detail. Not all the details are
convincing, and in any case there is no point in going into them here.
lxiv COMMENTARY ON ACTS

this is what Johannine Christians went out into the world to learn,
and with them it gave rise to docetism, answered by the watchword,
‘Jesus Christ in the flesh’.3 Against this Luke had set his warning on
Paul’s lips; he probably saw his own consensus environment as part
of the good past that he was holding up for imitation, scarcely aware
of the bad past with its different conflicts and dangers.
Of the development of this situation we have little contemporary
evidence. Clement of Rome is aware of renewed troubles in Corinth,
but they appear to be troubles of the old Corinthian kind, arising out
of rivalry, lawlessness, and indiscipline, but local. A better source,
probably a little later than Acts, is die Pastoral Epistles. Here too, as
in Acts 20, Paul is represented as a prophet who foresees coming
problems (1 Tim. 4.1; 2 Tim. 3.1; 4.3). These have a gnostic element
(note especially the βεβήλους κενοφωνίας καί αντιθέσεις τής
ψευδωνύμου γνώσεως of 1 Tim. 6.20) and a Jewish element (1 Tim.
1.7-9); one might say a Jewish-gnostic element (Ti. 1.14,
Ίουδαικοῖς μύθοις). Judaism is a problem also in the Ignatian
letters, and Ignatius himself was a gnostic of sorts. Serious division
lurks behind what can be known of the work of Hegesippus,4 whose
episcopal lists were designed to demonstrate and perpetuate apostolic
unity between the principal sees. The continuity of the lists will
prove that each see taught in agreement with the aposties; the
agreement between the sees will show that each confirms the rest.
Hegesippus, so far as we can tell, did not know Acts;5 this
introduces us to an obscure but interesting period in the history of
the book, written not far from AD 90 (see above, pp. xliif.) but
dropping out of use in a period in which its steady emphasis on
Christian beginnings, marked (as Luke represents them), by harmony
and sound teaching, might seem to have been particularly useful.
‘Die Apostelgeschichte finden wir vor Irenäus überhaupt nicht
bezeugt.’6 This is probably an overstatement; see I.34-45. But it
does not exaggerate greatly. The possible allusion in Polycarp 1.2
and the strange echoes in the Epistula Apostolorum do not carry
much weight. A more difficult question arises over Justin (see
I.41-44). It cannot be proved either that Justin did or that he did not
32 Jn 7—with the untranslatable present participle ερχόμενον.
4See Eusebius, HE, especially 4.22.3. In every succession and in every city (ἐν
ἐκάστη δέ διαδοχῇ καὶ ἐν ἑκαστη πόλει) so it is, as the Law and the Prophets and the
Lord teach. Eusebius preserves and expands the lists.
5Parallels have been seen between the account in Acts of the death of Stephen and
Hegesippus' account (in Eusebius, HE 2.23.4-18) of the death of James the Just.
James speaks of Jesus as the Son of man (2.23.13); he is thrown down, and prays for
those who stone him (16). But, in Hegesippus, the Son of man sits, not stands, at God's
right hand; James is thrown down from the Temple, and his prayer for the forgiveness
of those who stone him (ἄφες αὐτοῖς οὐ γάρ οἴδασι τί ποιοῦσιν) resembles Lk. 23.34
rather than Acts 7.60. Does Hegesippus belong to the time when Luke’s two books
were separated and the gospel was better known than Acts?
6H. von Campenhausen, Die Entstehung der christlichen Bibel (1968), 152.
 INTRODUCTION lχν

know Acts. There are indeed parallels of a sort. ‘The mission of the
apostles after the ascension of the Lord [Acts 1.8-10, 1 Apol 50.12
and Acts 1.9-11, Trypho 68.5] is amplified by the fact that they are
ἰδιῶται [Acts 4.13, 1 Apol 39.3] and that it is right to obey God, not
man [Acts 5.29, Trypho 80.3]. There are references to food which is
common or unclean [Acts 10.14, Trypho 20.3] and the designation of
Christ as judge of living and dead [Acts 10.42, Trypho 118.1]. The
word is received with joy and glory is given to God [Acts 13.27, 48,
1 Apol 49.5]. In both Acts and Justin may be found the widespread
concept of the άγνωστος Θεός and the rhetorical plea of sanity, “I
am not mad, most excellent Festus” [Acts 26.5, Trypho 39.4].’7 On
the whole, these are commonplaces, which indicate a common habit
of mind. The same goes for what is a Christian commonplace, the use
of the OT. At Trypho 87 Justin quotes Joel 2.28, 29, but P. Prigent8
rightly observes, ‘Le texte de Joël 2.28s. ne dépend pas d’Actes
2.17ss.’. In addition to specific points such as those listed above we
should note also the general cast of Justin’s mind. Like Luke, he has
accepted the notion that Christians have something to learn from the
Greek tradition; there were in it ‘Christians before Christ’, though
their work has to be accommodated to the OT, which constitutes the
word of God, Scripture, as the Greek writings cannot do. This makes
Justin, as Luke was, a moderate and tolerant man. Again, the limits
of his tolerance are much the same as Luke’s. Jewish Christians
Justin is more ready to accept than were some of his contemporaries.
True, they must not be allowed ‘to compel the faithful Christian
Gentiles to live according to the Law of Moses’ (Trypho 47, with a
parallel to Acts 15.1, 5). But they may keep the Law themselves; so
long as they do not impose their ways on others, they are to Justin
Christian brothers. Even the Jews (though Justin can describe them,
perhaps with some exaggeration, as cursing Christians in their
synagogues) are not beyond the range of reasonable and temperate, if
not exactly friendly, discussion, as the mere existence of the
Dialogue with Trypho shows. Justin is working within the limits of
the Decree of Acts 15.29; he disapproves of Jewish pressure only if it
goes beyond this. He does not however refer to the Decree.
More striking than Justin’s disuse of Acts is the fact that he
ignores Paul either completely, as is probable, or so nearly com-
pletely that the odd reference is almost more a disparagement than a
mark of respect. The neglect of Paul is underlined, and perhaps to
some extent explained, by the fact that Justin wrote a book,
unfortunately lost, against Paul’s great supporter Marcion. Even
Eusebius seems not to have seen the book, though he refers to it in

7E. F. Osborn, Justin Martyr (1973), 135; Osborn’s footnote references are given in
square brackets.
8Justin et l'Ancien Testament (1964), 114.
lxvi COMMENTARY ON ACTS

HE 4.11.8. Justin, he says, ‘wrote also a treatise against Marcion, in

which he mentions that at the time he composed it the man was alive
and well known. He speaks thus.’ At this point the reader of Eusebius
naturally expects a quotation from the book in question, but Eusebius
continues (quoting), ‘... and a certain Marcion of Pontus, who is
even now still teaching his followers to believe in some other God
greater than the Creator ..This passage is to be found in Justin, 1
Apol. 26. It is possible that Eusebius mixed up the quotations; it is
possible that Justin remembered what he had written in the Apology
and used it also in the book against Marcion (or vice versa). But it
remains true that though Eusebius tells us that Justin wrote a work
against Marcion he gives us no information about its contents, or any
evidence that confirms his statement. Irenaeus however also refers to
Justin’s book and in quoting Irenaeus Eusebius (HE 4.18.9) quotes
Justin. The passage is in Adv. Haereses 4.6.2. In the same passage
Eusebius also quotes from Adv. Haereses 5.26.2, where Irenaeus
quotes Justin. In Eusebius’s quotation there is nothing anti-marcio-
nite, but Irenaeus’s paragraph ends with words that could be an
attack on Marcion. It seems probable (though not proved) that Justin
wrote against Marcion; the book may have disappeared because the
case against Marcion seemed to have been more comprehensively
and powerfully made by Irenaeus himself and by Tertullian; there
was no need for the earlier work.
Justin’s disuse of Paul may be explained by his strong opposition
to Marcion. Marcion claimed to have the support of Paul, so that
Paul, also dangerously favoured by Valentinus and other gnostics,
could only be regarded as a divisive and disturbing force. This does
not however explain Justin’s strange combination of general parallels
with particular disuse of Acts, which must be taken as part of a
general disuse in the first half of the second century. It is tempting to
draw the conclusion that Acts was written in the middle of the
century as a mediating and conciliatory work, or at least, with John
Knox,9 to think that the double work, Luke-Acts, reached its final
form at that time and was directed against Marcion. This conclusion
however in its simplest form will not stand in view of the earlier
discussion (see pp. xliif.) of the date of Acts. But that Acts was
rediscovered, or perhaps deliberately revived, in the controversy with
Marcion and the search for a new consensus, a new stability, is a very
probable view. Itself the monument to an earlier consensus, dis-
tinguishable from but not wholly dissimilar to that which was sought
and ultimately achieved in the second century, it was admirably
suited for this purpose. Marcion did not include Acts in his canon; he
may not have known it, but there is at least something to be said for

9Marcion and the New Testament (1942).

 INTRODUCTION lxvii

the view that he knew it and rejected it.10 Evidence is provided by

Tertullian. In De Praescriptione 22 Tertullian does not mention
Marcion by name but there can be no doubt whom he has in mind
when he writes, ‘... it is proved in the Acts of the Apostles that the
Holy Spirit did come down. Now they who do not receive that book
of Scripture (quam Scripturam qui non recipiunt) can neither belong
to the Holy Spirit, seeing that they cannot acknowledge that the Holy
Spirit has been sent as yet to the disciples, nor can they pretend to
claim to be a church themselves ...’ In Adv. Marcionem 5.2 he
writes, ‘Since the Acts of the Apostles thus agree with Paul it
becomes apparent why you [Marcion] reject them (ea [Acta] respuas-
tis). It is because they declare no other God than the Creator; whilst
the promise of the Holy Spirit is shown to have been fulfilled in no
other place than in the book of Acts (quam de instrumento
Actorum).’ It is precisely in this sense that Irenaeus (see above) uses
Acts against Marcion: it shows that the apostles proclaimed no other
God than the God of the OT, who was the creator of the world. The
gnosticism and the anti-Judaism of Marcion are answered at one
blow; and at the same time Paul is shown to be in agreement with the
other apostles.
These observations may be taken a stage further. It is known that
Marcion wrote a book called the Antitheses. It is unfortunately lost,
but its contents have been partially reconstructed on the basis of
quotations and references, supplied almost exclusively by Marcion’s
enemies.11 It is certain that the Antitheses contained much more—a
theological exposition of Marcion’s position and material explana-
tory of passages in Marcion’s gospel, Luke, and in the Pauline
epistles—but it is not wrong to describe it, as Hamack does, as
Marcion’s response to the rejected Acts. ‘Am Anfang der “Antithe-
sen” hat sich Μ. im Anschluss an Gal. 1.2 über das Verhältnis von
Paulus und den Uraposteln ausgesprochen und die Uraposteln char-
akterisiert sowie “die falschen Brüder”.’ (81). Some have gone
further and taken the Antitheses to have been part of the new
Marcionite Scriptures. This view is based on Adv. Marcionem 1.19
where Tertullian speaks of the Antitheses as the Marcionite ‘sum-
mum instrumentum’, and of the Gospel and the Law as ‘instru-
menta’. It is however to be rejected.12 The Antitheses is an ‘instru-
mentum quo denique [the Marcionites] initiantur et indicantur [v.l.,
indurantur] in hanc haeresim’; that is the book serves as an introduc-
tion to, not part of the text of Scriptures. Hamack adds (155*, 284*) a
further point, which deserves to be taken more seriously than it is by
10A. Hamack, Marcion: das Evangelium vom fremden Gott (1921), 153*, ‘Marcion
hat die Apokalypse und die Acta als falsche, d.h. als Bücher des Judengottes
verworfen.'
11Hamack, Marcion, 68-135.
12With J. Knox, Marcion, 6.
lxviii COMMENTARY ON ACTS

Knox (121). About AD 400 the Marcionite church was still active in
Mesopotamia; it may be assumed that it was in possession of its
sacred literature and that the Marcionite NT and the Antitheses were
current and were read both by Marcionites and by those who entered
into controversy with them. Among the latter was a local bishop,
Marutha of Maipherkat (Maipheracti, Martyropolis, Tagsit). His
work confirms what is otherwise known of Marcion’s NT, and he
adds (Hamack’s translation), ‘Das Buch der Πράξεις haben sie [the
Marcionites] vollständig aus der Mitte geräumt und statt seiner ein
anderes Buch ... eingeschoben, das sie Sākā = Summa ... nennen,
so dass es sei gemäss ihren Meinungen und Lehren.’ On this Hamack
comments (155*f.), ‘Auch nach diesen Worten liegt die Annahme
nahe, dass die Ausstossung der Apostelgesch. ausdrücklich begrün-
det war; das Buch aber, welche sie an ihrer Stelle hatten, können nur
doch wohl die ‘Antithesen’ sein; denn diese haben auch nach Tert.
eine Auseinandersetzung mit den Uraposteln enthalten, ja wahr-
scheinlich in ihrer Einleitung. Darüber hinaus aber sind die “Anti-
thesen” wirklich ein “Ersatz” für die Apostelgeschichte; denn mit
dieser begründete die Kirche die Concordanz zwischen dem Alten
und dem Neuen Bund und zwischen den Uraposteln und Paulus; mit
den Antithesen aber begründete Μ. die Discordanz zwischen diesen
Grössen.’ The words Concordanz and Discordanz are to be noted.
If it is correct that Marcion wrote the Antitheses in order to
introduce his NT, and if in doing so he conceived himself to be
replacing Acts (and there are powerful though not overwhelming
arguments in favour of both propositions) we have an important
pointer to the position of Acts at this time. It was thought of not as a
part of the NT as this, consisting of gospels and epistles, was
gradually coming into being; it served rather as an introduction to the
new collection. A number of considerations are consistent with, and
thus give support to, this conclusion.
It is suggested above (pp. 1f.) that what Luke in his two-volume
work produced was a primitive NT. It contained everything that a NT
should contain. The life and teaching of Jesus were there, together
with an account of his death and resurrection. The work and teaching
of at least some representative apostles and preachers were there,
together with the necessary check on the continuity between Christ
and the church. What more could a church want? The local church
within and for which Luke wrote was probably for a time content, but
it is evident that the church as a whole wanted a good deal more. It
wanted all the gospels it considered orthodox, four not one, even
though the multiplicity of gospels was eventually to prove a theologi-
cal problem.13 More influential than theological problems was the

13See O. Cullmann, ‘Die Pluralität der Evangelien als theologisches Problem im

Altertum', in V & A 548-65.
 INTRODUCTION lxix

fact that local churches had local gospels and were unwilling to give
them up. And some churches had letters written by apostles, or other
authorities. 'Our own precious piece of Paul’s handwriting is worth
more than Luke’s stories.’ And churches passed on copies of ‘their.’
letters to other churches, Corinth for example to Rome.14 Luke’s
‘Shorter New Testament’ (but he had nothing with which to compare
it) was out-dated and discarded.
Not wholly out-dated, and in the end not discarded. Half of the
total work was a gospel, and gospels were of evident authority and
one knew what to do with them. They were stamped with the
authority of the Lord himself and could not simply be cast aside. So
Lk. became part of a developing four-gospel canon. This meant
however that it was separated from Acts, and that Acts now stood
alone and (unless Lk. was placed last among the four—as in the
Curetonian Syriac) lost the connection that gave it its original
meaning and raison d’etre as the ‘confirmation of the gospel’.15 As
soon as this happened gospel and Acts began to have different
histories. Acts had in fact very little history; the epistles gave a more
direct contact with their authors and, as was pointed out above, began
to circulate and to be accorded a measure of authority which the
interesting stories of Acts did not have. It was now not apparent what
purposes Acts would serve. Did a reader with theological interest
desire to sit at Paul’s feet? There was a great deal more for him in
Romans than in, say, Acts 13 and 17. Did he need guidance in regard
to moral problems? He could find them discussed explicitly in 1
Corinthians, whereas they scarcely reached the surface in Acts. Acts
moreover was anonymous; how different from, say, Galatians; or, if
you lived in Asia, works ascribed to John!
That Acts lacked the authority of other parts of the NT is clearly
proved by the textual phenomena.16 There is no part of the NT in
which textual variants are so many and free. The Western text is, as
all know, not geographically Western at all. All over the ancient
world, not only in areas in which Greek was spoken, but where Latin,
Syriac, and Coptic were used, the text of Acts was handled freely.
Again and again changes are made with no apparent concern other
than that of introducing a measure of variety, or simply because the
copyist did not trouble to remember precisely what he had seen in his
exemplar. There is a Western text in other NT books but nowhere
does it show the same wild variety as it does in Acts—a clear sign
that Acts lacked the strict verbal authority that other Christian books
possessed.
The fact is that the church of the next two or three generations
14See 1 Clement 47.1f.; Ignatius, Eph. 12.2; Rom. 4.3; Polycarp, Phil. 3.2; 11.2
(quoting 1 Corinthians).
15See van Unnik’s article referred to on p. li.
,6I.3-29; II.4.
lxx COMMENTARY ON ACTS

after its composition did not know what to do with Acts. It was not a
gospel; it was not an epistle; it did not fit into any known category of
Christian literature, of which indeed there was very little in any form.
If the notion of a New Testament over against an Old had been
established it might have been classed with the historical books, such
as Samuel and Kings. But it was part of the unique event that had
brought the Christian movement into being. In separation from the
gospel it first came into use as an introduction to the new group of
books that formed an incipient NT, or perhaps simply as an
introduction to the Apostle over against the Gospel section of the NT,
showing how the two parts belonged together. It was no good to
Marcion, who supplanted it with his Antitheses, and thereby gave his
orthodox opponents the cue that they needed. If Marcion did not like
Acts, Acts must be anti-Marcionite. And so it was, as first Irenaeus
and then Tertullian saw. It showed Paul, in unity with the other
apostles, engaged in preaching one God and one Gospel, the one God
who was both the Creator of the material universe and the Father of
the Lord Jesus Christ the Redeemer. Here was the biblical, apostolic,
non-Marcionite, non-Valentinian ground on which all Christians
might unite. Acts might be part of the NT, standing in its appropriate
place to demonstrate that the story of Jesus issued in the church and
that the church arose out of the word of Jesus.
And, from the time of Irenaeus and the Muratorian Canon, Acts
was part of the NT.17 It continued however to stand a little apart from
the other books. The canon was not thought of as supplying an
authorized history of Christian origins; it existed primarily as a basis,
and yardstick, for Christian preaching,18 and there is little to suggest
that Acts was frequently preached on. It remained something of a
puzzle, neither a gospel, stamped with the authority of the Lord
himself, nor an epistle bearing an apostolic signature. And what
could a theologian make of stories? The theologians of the early
centuries did not see that there might be a positive answer to this
question.
There is little material available for writing a history of the use of
Acts in the first centuries.19 What there is has been admirably
collected by P. F. Stuehrenberg.20 His list contains 147 entries, but
the number, impressive at first, says more for Dr Stuehrenberg’s
diligence and scholarship than for the exegetical activity of the pre-
Reformation writers. Most of the works listed have never been

17See 130-34, 44-48.

18See FS Schweizer 5-21.
19Though more than W. Gasque, A History of the Criticism of the Acts of the
Apostles (1975) takes note of.
20‘The study of Acts before the Reformation. A Bibliographic Introduction’, NovT
29 (1987), 100-136. See also F. Stegmüller, Repertorium Biblicum Medii Aevi
(1940-1980).
 INTRODUCTION lxxi

published and can be consulted only in manuscript. Most of them are

known to us only in fragments containing brief notes on a few
scattered verses. Many of them are known only through quotations in
Catenae, notably in J. A. Cramer, Catena in Acta S.S. Apostolorum,
Oxford 1838 (Volume ΙII of Catenae Graecorum Patrum in Novum
Testamentum). Thus from the third century Stuehrenberg is able to
cite a comment on Acts 5.4 from Dionysius of Alexandria; this is in
Cramer and comes from a commentary not on Acts but (probably) on
Ecclesiastes, and from Origen comments on 1.16 (given by Migne)
and on 4.33; 7.4, 52; 21.38 (given by Cramer). Origen however is
known to have written homilies on Acts; he wrote on almost the
whole Bible. From the early centuries the only works of which we
can read a continuous text are the Homilies of Chrysostom, of which
there is no satisfactory critical Greek text, and the exposition of
Bede. Of Chrysostom’s Homilies Stuehrenberg (110) writes, ‘John
Chrysostom’s fifty-five sermons on Acts were extremely influential
throughout the Middle Ages, frequently serving as the basis upon
which subsequent commentaries were constructed. These sermons,
composed ca. 400, were “the only complete commentary on Acts
that has survived from the first ten centuries’’ (Quasten, ΙII, pp.
440-441).’ There is little that is original in Bede’s commentary but it
is of considerable value in having preserved material from a number
of older writers, notably Jerome, Gregory, and Arator. The last
named should probably be given individual mention for his epic
poem, De Actibus Apostolorum. This is not quite a commentary;
whether it is a paraphrase of Acts which draws special attention to
baptism or a work on baptism with special reference to Acts is
perhaps open to discussion. There is a recent edition of Arator’s 2326
hexameter work by R. Hillier.21
Chrysostom’s Homilies are practical sermons from which we learn
as much about life in Antioch in Chrysostom’s time as about the times
of the apostles, but there is a good deal of plain practical exposition
and application of the text. It is notable that the infelicities that trouble
modern readers of the Greek text of Acts and lead to hypotheses such
as that of an Aramaic original seem to give the Greek orator little or no
trouble. Perhaps we are too sensitive. Bede also was a writer of
common sense and practical interest—and good at apt quotation.
Luther saw Acts as essentially a reproduction of Pauline theology.
It was not simply a piece of history and an example of good works,
expressed in the sharing of property and in mutual concern. ‘Sondern
darauff sol man mercken Das S. Lucas mit diesem Buch die gantze
Christenheit leret bis an der Welt ende das rechte Heubtstück
Christlicher lere nemlich Wie wir alle müssen gerecht werden allein

21R. Hillier, Arator on the Acts of the Apostles. A Baptismal Commentary, Oxford
Early Christian Studies, 1993.
lxxii COMMENTARY ON ACTS

durch den glauben an Jhesu Christo on alles zuthun des Gesetzes

oder Hülffe vnser werck.’22 It is hard to acquit Luther of reading Acts
through spectacles more Pauline than Luke himself used. Luke did
have an interest in history, and his one reference to justification
though put on Paul’s lips is not cast in quite the Pauline mould (see
on 13.38f.). Luke does however insist on faith as the primary
Christian requirement (e.g. 15.9). Calvin with a full dress commen-
tary on Acts,23 is able to take exegesis more seriously than many
subsequent expositors—sometimes more seriously than even Luke
himself.24 He also takes historical questions seriously, and is perhaps
the first of modern commentators, occasionally, and surprisingly,
anticipating Enlightenment and post-Enlightenment writers.
It is impossible here to give a full account of more recent work on
Acts; also unnecessary, since useful accounts are already available.
The following should be noted:
W. Gasque, A History of the Criticism of the Acts of the Apostles,
BGBE 17, Tübingen, 1975.
‘A Fruitful Field’, Interpretation 42 (1988), 117-31.
E. Grässer, ‘Die Apostelgeschichte in der Forschung der Gegenwart’,
ThR 26 (1960), 93-167.
‘Acta-Forschung seit 1960’, ThR 41 (1976), 141-94; 259-90; 42
(1977) 1-68.
E. Haenchen, 29-63, 124-41.
RGG 1.507f.
F. Hahn, ‘Der gegenwärtige Stand der Erforschung der Apos-
telgeschichte’, Theologische Revue 82 (1986), 177-90.
J. W. Hunkin, British Work on the Acts. Begs. 2.396-433.
W. G. Kümmel, ‘Das Urchristentum, I, II, IIΙ, IV’, ThR 14 (1942),
81-95, 155-73; 17 (1948), 3-50, 103-42.
Nachträge zu Teil Ι-ΙII, 18 (1950), 1-53.
‘Das Urchristentum I, IT, ThR 22 (1954), 138-70, 191-211.
‘Das Urchristentum I, IIa, b, c, d, e’, ThR 48 (1983) 101-28; 50
(1985), 132-64, 327-63; 51 (1986), 239-68; 52 (1987), 111-54,
268-85.
‘Zwei Gesamtdarstellungen des Urchristentums’, ThR 52 (1987),
401-9.
A. C. McGiffert, The Historical Criticism of Acts in Germany, Begs.
2.363-95.
A. J. and Μ. B. Mattill, A Classified Bibliography of Literature on
the Acts of the Apostles, Leiden, 1966.
E. Plümacher, ‘Acta-Forschung 1974-1982’, ThR 48 (1983), 1-56;
49 (1984), 105-69.
22Vorrede auff der Apostel Geschichte, in the 1545 edition (Stuttgart, 1989).
23See the translation by J. W. Fraser and W. J. G. McDonald, two volumes, 1965
and 1966.
24See FS Corsani, 312-26.
 INTRODUCTION lxxiii

The reports contained in Theologische Rundschau are of very great

value; it is as needless as it is impossible to work on the same scale in
this Introduction.
Haenchen (29-63) in his survey of ‘die historisch-kritische Acta-
forschung’ divides his account into the following sections: A. Die
Epoche der Tendenzkritik; B. Die Epoche der Quellenkritik; C. Die
Kritik bis 1945; die Formgeschichte (1. Phase); D. Die Kritik nach
1945; Formgeschichte (2. Phase). In the edition of 1977 a supplemen-
tary section was added, Die Arbeit geht weiter. Whether Haenchen’s
description of the first period, which focuses on F. C. Baur, Schneck-
enburger, and Overbeck, is most happily described as that of Tend-
ency-criticism is open to question. The description is of course under-
standable; it was one of Baur’s principles that ancient sources must be
examined for their tendencies, that is, in English idiom, one must ask
what axes their authors had to grind. It is in the light of this that their
historical worth must be assessed. This is truly observed; but there is a
prior, though related, observation, namely that our sources (alongside
Acts must be placed the Pauline epistles) do not tell the same story,
and the historian’s task is not to harmonize them but to set them over
against one another. This is criticism; and it is when the reader
observes that the first Christian decades were not a period of universal
peace25 that he begins to ask why Acts gives the impression that they
were, and to inquire what tendency led Luke to write as he did.
Tendency may prove to be the wrong word, if it suggests an intention
to present things as other than they were. Misleading impressions may
be given as a result of incomplete knowledge, or of unconscious
presupposition. There is ample room here for debate, and ‘tendency’
makes the debate theological as well as historical. Baur had his own
explanation of the phenomenon he observed, an explanation that was
already being questioned and modified in the middle years of the
nineteenth century. His most important contribution was the observa-
tion (anticipated to some extent by Sender and Michaelis)26 that the
phenomenon existed. The Pauline epistles made it clear that there was
a conflict between Pauline Christianity and Judaizing Christianity,
whether the latter is to be put down to the original apostles or to other
Jewish Christians. This is not the point at which to attempt an answer
to the question, What actually happened? The first question, which
Baur was perhaps the first to see clearly, is, What are we to make of
Acts? It is important, and often overlooked or indeed contradicted,
that Baur could write as late in his career as 184527 that Acts remains
‘eine höchst wichtige Quelle für die Geschichte der apostolischen
Zeit,’ though he adds immediately ‘aber auch eine Quelle, aus welcher
25See above, note 1.
26See W. G. Kümmel, The New Testament: The History of the Investigation of its
Problems (ET 1973), 120.
21Paulus, p. 13; reprinted in the second edition (ed. E. Zeller, 1866), p. 17.
lxxiv COMMENTARY ON ACTS

erst durch strenge historische Kritik ein wahrhaft geschichtliches Bild

der von ihr geschilderten Personen und Verhältnisse gewonnen wer-
den kann.’ What Baur found in Acts was neither pure Paulinism (as
determined by the epistles) nor pure Judaizing (the requirement that
Gentile converts should be circumcised was clearly rejected, for
example). It was not unnatural to think that the middle course between
the two extremes was a compromise; and this was not incorrect. Baur
was mistaken however in thinking that the compromise was devised
and was being commended by Luke and his contemporaries in the
second century.28 Luke did not invent the compromise; already in the
first century he found it in operation and did not question that it had
been intended by all the chief representatives of the preceding dec-
ades. It is in something like this sense that a sentence of Overbeck’s,
quoted by Haenchen (37) is correct: ‘Ihr [der Apostelgeschichte]
Heidenchristentum ist nicht das paulinische, aber noch weniger ist ihr
Judaismus das urapostolische.’ Luke confuses the Pauline Gentile
mission with another, and ignores the Judaism that he alludes to in
15.1,5—probably because it had come to an end.
Baur’s tendencies were those that must appear as soon as the notion
is abandoned that all apostolic and semi-apostolic writings must (if
rightly understood) present a harmonious picture of a peaceful and
blameless apostolic age. For unfettered historical inquiry applied to
Acts (and of course to the rest of early Christian literature) all sub-
sequent students are indebted to him (and to Loisy after him). But
estimates of tendency are apt to be subjective, and Haenchen is right in
picking out Quellenkritik as the next stage in the study of Acts. A
possible answer to the question why some things are not mentioned in
Acts may be not that Luke chose to keep quiet about them but that his
knowledge was limited by the sources available to him.
Haenchen takes his account of Quellenkritik as far back as Schleier-
macher, and indeed further. A more detailed account is given by
J. Dupont,29 and the question of sources has been discussed earlier in
this Introduction, and in I.49-56. Luke probably used a considerable
number of sources (if that is the proper name for them), collecting
information where he could. He was probably himself a well-travelled
man, who sought information in the Christian groups he visited; there
is no reason why he should not have obtained written information by
letter. The only more extensive pieces that may be singled out are the
Antiochene traditions—one cannot tell whether they were in written
or in oral form—and the We-passages.
Sections C and D of Haenchen’s summary of work on Acts are
headed Formgeschichte; the first phase deals with die Kleinen
Einheiten, the second with die Komposition. With a few deviations

28On this see the commentary on ch. 15.

29 J. Dupont, The Sources of Acts (ET 1964).
 INTRODUCTION lxxv

these sections are devoted to the work of Haenchen’s great predeces-

sor, Μ. Dibelius, whose essays on Acts were collected and edited by
Heinrich Greeven in a volume, Aufsätze zur Apostelgeschichte
(1951).30 The term Formgeschichte is unfortunate because though
there is resemblance between Dibelius’s work on Acts and the
contributions made by him and others to the form criticism of the
gospels there is not the same concern with the form of the units that
are studied. Dibelius’s concern is to distinguish between and to
separate out the traditional material used by Luke and his editorial
redaction of it—redaction criticism, much used since his time, would
in fact be a better term for what Dibelius himself practised. Speeches
may indeed be separated from narrative, and it is sometimes possible
to see the literary devices by which speeches have been accommo-
dated to the narrative settings in which they have been placed. A
good example is found in ch. 3, where a speech which is on
fundamentally similar lines to that of ch. 2 is fitted to the story of the
healing of the lame man at the Temple gate (vv. 12, 16). A
particularly clear (though not in fact convincing—see 1.491-8)
example of Dibelius’s method is provided by his treatment of the
conversion of Cornelius in ch. 10. At the root of this episode
(according to Dibelius) there was a simple pious story of a conver-
sion; this was taken up and used as an important step in Luke’s
account of the expansion of the Gospel into the Gentile world. Not
only was Peter’s short statement of the Gospel added but also the
story of Peter’s vision, and the setting was constructed so as to bring
out the radical step involved in Peter’s visit to and, in the end,
baptism of a Gentile household. A somewhat more convincing
example is provided within this one by the use made of Peter’s
vision, which seems in itself to refer to the relaxation of food laws
but is used in Acts not with reference to Gentile food but to the
Gentiles themselves: no man is to be regarded as unclean (10.26).
The total effect of Dibelius’s work, which after 70 years (counting
from the earliest of his essays) remains of great importance, was to
move the historian’s use of Acts a stage or two further back. It is
impossible to give an immediate general estimate of the whole book,
or even to distinguish between more and less trustworthy sources.
The contents of the book must be analysed and classified, rather in
terms of content than of form: speeches distinguished from narratives
(for speeches contained within narratives may have had a completely
different origin) and summaries from particular episodes. In regard to
each unit the attempt must be made to distinguish between a
traditional basis and the editorial use that has been made of the
tradition. It must be emphasized that this is a far more problematical
process than some have recognized; it goes too far to describe it as in

30ET, Studies in the Acts of the Apostles, 1956.

lxxvi COMMENTARY ON ACTS

itself and necessarily guesswork, but strict control has to be exercised

in order to prevent criticism from descending into guessing. In the
Third Gospel Luke’s work can be detected by a comparison with
Mk., less confidently by comparison with Mt. In Acts the only
comparative material is provided by the Pauline epistles, and this
material is narrowly limited in extent. The necessary control (so far
as control is possible) is provided by study of Lucan interests (but
may not Luke’s interests be based on, derived from, his knowledge
of tradition?) and Lucan style (but the variety of styles that Luke
could command has often been noted). Also to be noted is the
editor’s arrangement of his material.
Mention of Dibelius has led, inevitably, to what may be called
redaction criticism; to this we shall soon return. The supplementary
chapter in Haenchen’s 1977 (7th—16th in the Meyer series) edition
(pp. 124-41), entitled ‘Die Arbeit geht weiter’, is not analysed in the
same way as the original chapter but surveys the work of a number of
authors. The literature on Acts that has appeared since 1977 (roughly
the period occupied by the writing of this commentary) has been
diverse in range and immense in quantity. The survey articles by
Grösser and Plümacher referred to above (p. lxxii) cover some of it as
well as earlier material. No claim is made here to have surveyed the
whole, and no attempt will be made to present an adequate account,
which would remove all balance from the commentary. A few lines
of study will be briefly illustrated.
Redaction criticism has already been described, and is practised,
though to a limited degree and with a measure of scepticism, in this
commentary. Scepticism is necessary because in Acts redaction
criticism turns to so great an extent on subjective judgement, as was
pointed out above. Perhaps the best representative of the redactional
method is G. Lüdemann (see I xiii), whose ‘Kommentar’, which
omits much of the information traditionally found in commentaries,
consists in an attempt to distinguish throughout Acts between
tradition and redaction, and on this basis to assess its value as a
historical source. One of the examples picked out by Lüdemann
himself to illustrate, in his Introduction (pp. 17-20), the method he
employs is the account in Acts 18 of Paul’s visit to Corinth (see
below in this Commentary pp. 858-82). Redactional features stand
out at once: Paul’s Sabbath preaching in the Synagogue and the
favourable picture of Gallio. The use of Jews as a point of contact for
the Christian preacher and the representation of Roman authorities as
tolerant of the Christian movement are frequently recurring themes.
These points may indeed be redactional here, but it is worth noting
that the synagogue, rightly approached, did afford a ready-made
audience, and if some Roman officials (not necessarily of course
Gallio) had not shown a fair-minded approach the apologetic stories
in Acts would have been worthless. But there is also (Lüdemann
 INTRODUCTION lxxvii

thinks) traditional material: Paul’s work with Aquila and Priscilla,

and their recent arrival in Corinth as a result of Claudius’s expulsion
of the Jews from Rome; the arrival of Silas and Timothy from
Macedonia; Paul’s use of the house of Titius Justus; the conversion
of the ruler of the synagogue, Crispus; the ‘trial’ before Gallio; the
reference to the ruler of the synagogue, Sosthenes. ‘Nicht sprachliche
Gründen führen zu der Annahme von Traditionen, wohl aber die
Konkretheit der obigen Nachrichten und—wichtiger—der ...
Befund, dass ein nicht unerträchtlicher Teil der Nachrichten wenig-
stens z. T. durch die Paulusbriefe bestätigt wird’ (p. 18). Not that
Lüdemann neglects linguistic considerations. The above quotation is
taken from his brief introductory summary. His full treatment (under
the heading Redaktion) of 18.1-17 begins (p. 203), ‘V.1: Zu meta
tauta vgl. die luk. Parallelen 7.7 (cit.); 13.20; 15.16 (cit.); Lk 5.27
(diff. Mk); 10.1; 12.4 (diff Mt.); 17.8; 18.4. V.2-3: Die Verse tragen
sprachlich-syntaktisch luk. Kolorit: vgl. onomati, to genei (vgl. 4.36;
18.24), AcI [= Accusative and infinitive] mit dia eingeleitet (bis);
chorizesthai nimmt choristheis (V. 1) auf. Der gehäufte Partizipial-
stil ist gleichfalls redaktionell.’ And much more.31 Even when a good
deal of redactional material is eliminated (though it would be
illogical to assume that anything due to Luke’s editorial work on a
source is unhistorical) Lüdemann is able to maintain (p. 24) that ‘Die
Apg bleibt neben den Paulusbriefen eine wichtige Quelle für die
Geschichte des frühen Christentums.’
From redaction criticism it is natural to move by way of a book
described by its author, P. F. Esler,32 as socio-redaction criticism, to
the sociological study of Acts. Sociological study is not unrelated to
historical criticism, for historical study if it is to be complete must
always include a social element; the best historians have always
known that they must inquire into the way in which men and
women—not only kings and queens and their ministers but ordinary
folk—conducted their lives, and that even power struggles on the
largest scale were conducted by means of personal and admin-
istrative agencies. It is however undoubtedly true that in recent years
greater stress has been laid upon investigation of the personal lives of
Christians in NT times and that the methods of sociology have been
employed. Esler looked back to pioneering work by G. Theissen,33
W. A. Meeks,34 and others, but developed his sociological observa-
tions in the light of Lucan theology. Much of his book is social
31For the use of linguistic characteristics as marks of Lucan redaction see especially
the commentary of Weiser.
32Community and Gospel in Luke-Acts, SNTSMS 57, 1987; see now also The First
Christians in their Social Worlds, 1994; J. H. Neyrey, ed., The Social World of Luke-
Acts, Peabody, Mass., 1991.
33The Social History of Pauline Christianity (1982), a translation of articles
published some years earlier.
34The First Urban Christians (1983).
lxxviii COMMENTARY ON ACTS

history, and his observations on the Law, the Temple, the Poor and
the Rich, and Rome are of great value. What calls for attention here
is Esler’s debate with the ancient historian E. A. Judge (who has also
made very valuable contributions to the study of early Christian
social history) about the applicability of the methods of sociology in
NT history, and indeed to ancient history in general. Unfortunately it
is impossible to follow the debate in detail here. ‘Reading between
the lines of Judge’s article,35 one senses that his real worry with
sociological exegesis is that its exponents will attempt to plug holes
in first-century data by drawing upon relevant features of the
comparative materials they apply to the New Testament text’ (Esler,
p. 14). Esler and Judge are in fact not so far apart as might at first
appear. Esler’s reply to Judge’s ‘worry’ is in essence that models and
types should be more accurately defined and more carefully used
than they are by some sociological practitioners who turn to ancient
history, and his own observations in ancient social history as it
concerns the NT are, as was noted above, of considerable value. The
NT (like e.g. the Histories and Annals of Tacitus) is an old book, and
there are no easy transitions (beyond the basic facts of human nature)
between its world, and its forms of society, and our own. Socio-
logical procedures can help to open up and sharpen questions for us,
but the answers are to be found in the patient observation and
collection of data found in ancient texts, literary and non-literary, and
in the results of archaeological exploration, and it must not be
expected that answers will be found by sociological analogy, if
indeed answers are to be found anywhere.
More recent years have seen the development of methods of
literary study, developed first, for the most part, in the realms of
secular literature and applied to the NT as a whole. These have more
to do with hermeneutics than with history,36 and find more scope in
other parts of the NT than in a matter-of-fact work such as Acts. This
is not to say that they are of no value, though the thought of Acts lies
on the surface and does not require elaborate hermeneutical arts to
draw it out.
Not that there are no literary questions to ask about Acts. One may
ask, for example, to what class of literature it belongs. This is a
question that was raised above (pp. xlviii—li) and here it must suffice
simply to mention a few terms, all of which cover part of the truth,
none the whole truth. It looks like history,37 biographical history, the

35Journal of Religious History 11 (1980), 201-217.

36See R. C. Morgan and J. Barton, Biblical Interpretation (1988), where they are
dealt with in ch. 7 (pp. 203-68). ‘Some of the different ways that interpreters read a
work of art today may prove more suggestive for theological interpretation than a
historical scholarship which is less interested in the aesthetic and moral significance of
great literature’ (p. 203).
37See Plümacher, Lukas als hellenistischer Schriftsteller.
 INTRODUCTION lxxix

historical monograph.38 It has been described as an apology, and that

in several senses, and like a hellenistic romance. Luke had found his
own way of communicating the Gospel and bearing witness to
Christ; that is perhaps as near as we can get to a characterization of
Acts. To say that the book is sui generis may seem like running away
from a difficult and disputed problem of classification, but it is in fact
true. No Christian wrote a book like it; the apocryphal Acts are at
best like a Western text run even wilder.39
Also to be briefly mentioned here is the literary analysis of
narrative, in relation not only to the narrator but also to the reader. It
is possible to distinguish between the actual author of a piece of
literature and the implied author, between the actual readers and the
implied readers. The actual author of the Sherlock Holmes stories is
Conan Doyle, a historical figure, about whom a good deal is known
(far more than about most ancient authors). The implied author,
implied by the form and content of the stories, is Dr Watson, a person
known only from the stories themselves, though from them quite
well; we know something of his medical training, his military
service, his marriage, his loyal though not always highly intelligent
assistance to his friend. In some kinds of literature this distinction is
illuminating; hardly so in Acts, where the actual author is known, if
at all, only by identification with the implied author, who shows his
hand only in the We-passages (if indeed there). Of the actual readers
of Acts, again, we know little. Some of the evidence has been set out
above (I.30-48). There may have been earlier readers, indeed it is
highly probable that there were, but we can hardly give a name to one
before Irenaeus. There is an excellent discussion of the implied
readers in J. B. Tyson, Images of Judaism in Luke-Acts (Columbia,
South Carolina, 1992). We may infer the characteristics of the
implied readership from the extraneous pieces of information40 which
the author thinks it necessary to supply, and from those facts which
the author assumes that the reader will know. Tyson’s conclusions
(pp. 35f.) are worth quoting.
1. Our reader is generally a well-educated person with a
rudimentary knowledge of eastern Mediterranean geography
and a familiarity with the larger and more significant Roman
provinces.
2. The implied reader is familiar with some public figures,
especially Roman emperors. He has some knowledge about
James and his position within the primitive Christian commu-
nity ...

38Again, Pliimacher.
39FS Black (1979), 27.
40See S. Μ. Sheeley, Narrative Asides in Luke-Acts, JSNTSupps 72 (1992); R.
Tannehill, The Narrative Unity of Luke-Acts, Philadelphia, I, 1986, II, 1990.
lxxx COMMENTARY ON ACTS

3. The implied reader is not expected to know any language

other than Greek ...
4. The implied reader is knowledgeable about public affairs

5. ... has a working knowledge of common Greek and

Roman measurements and coinage.
6. ... has a working knowledge of both pagan and Jewish
religions, an aversion to some pagan practices, and an attraction
to Jewish religious life. But he is probably not Jewish and is not
well informed about certain significant aspects of Jewish reli-
gious life.
7. ... is familiar with the Hebrew Scriptures in their Greek
translation and acknowledges their authoritative status but is not
familiar with those methods of interpretation that find the
fulfilment of the scriptures in Jesus.
From these observations Tyson infers that the implied reader ‘is
similar to those characters in Acts that are called “Godfearers” ’ (p.
36).41 The two centurions (of Luke 7 and Acts 10; 11) serve as good
examples.
This sort of observation is interesting and helpful in the study of
Acts. It is however fairly described as a further development of
historical study, an aspect of historical criticism little noticed in the
past and now profitably brought to the fore.
The same comment may be made on the use, of which much is
now being made, of the rhetorical principles taught and practised in
antiquity. The works of the orators are an important part of ancient
literature, and instruction in oratory was an important part of ancient
education. These facts are not a new discovery, but the focusing of
attention on them is new, and has borne a good deal of fruit in the
study of the NT. Not so much fruit however in the study of Acts. As
we have seen, Luke writes in such a way as to cause his work to be
counted along with and judged alongside the work of ancient
historians. His history is a very readable work, clear, fascinating,
illuminating, and not inaccurate. But this is not because Luke follows
the rules of the rhetoricians. On the whole his manner corresponds
closely with the words of Shakespeare’s Mark Antony: T am no
orator, as Brutus is ... I only speak right on.’ Special attention has
been given to the speeches recorded in Acts, e.g. to the speech of
Tertullus (24.2-9), which has been described as a masterpiece of
forensic oratory, and Paul’s reply (24.10-21). These however, like
most of the speeches in Acts, are too short to make the rules of
rhetoric applicable. Rules that are suitable to (e.g.) Demosthenes’ De

41For the problems arising out of this term, which are not discussed by Tyson, see
1.499-501.
 INTRODUCTION lxxxi

Corona (112 pages in OCT), cannot be reasonably adapted to

speeches, or summaries of speeches, only a few lines in length.

Bibliography
C. K. Barrett, FS Corsani, 312-26.
H. W. Bartsch, ThLZ (1972), 721-34.
W. Bieder, Die Apostelgeschichte in der Historie, ThSt 61, Zürich 1960.
H. J. Cadbury, The Book of Acts in History, London 1955.
H. von Campenhausen, Die Entstehung der christlichen Bibel, Tübingen
1968.
Μ. Dibelius, Studies in the Acts of the Apostles, ET London 1956.
P. F. Esler, Community and Gospel in Luke-Acts, SNTSMS 57, Cambridge
1987.
P. F. Esler, The First Christians in their Social Worlds. London and New
York 1994.
W. Gasque, A History of the Criticism of the Acts of the Apostles, BGBE 17,
Tübingen 1975.
W. W. Gasque, Interpretation 42 (1988), 117-31.
A. George, Études sur l'oeuvre de Luc, Paris 1978.
A. Harnack, Marcion: das Evangelium vom fremden Gott, Leipzig 1921,
21924.
R. Hillier, Arator on the Acts of the Apostles, Oxford 1993.
E. A. Judge, Journal of Religious History 11 (1980), 201-217.
R. J. Karris, CBQ 41 (1979), 80-97.
A. F. J. Klijn, NovT 10 (1968), 305-312.
J. Knox, Marcion and the New Testament, Chicago 1942.
W. S. Kurz, Bib 68 (1987), 195-220.
A. J. and Μ. B. Mattill, A classified bibliography of literature on the Acts of
the Apostles, Leiden 1966.
A. J. Mattill, FS Bruce (1970), 108-122.
A. J. Mattill, CBQ 40 (1978), 335-50.
W. A. Meeks, The First Urban Christians, New Haven and London 1983.
F. Neirynck and F. van Segbroeck, EThL 59 (1983), 338-49.
S. Μ. Sheeley, Narrative Asides in Luke-Acts, JSNTSupp 72, Sheffield 1982.
F. Stegmüller, Repertorium biblicum Medii Aevi, Madrid 1940-1980.
P. F. Stuehrenberg, NovT 29 (1987), 100-136.
C. H. Talbert (ed.), Luke-Acts, New Perspectives, New York 1984.
G. Theissen, The First Followers of Jesus, ET London 1978.
J. B. Tyson, Images of Judaism in Luke-Acts, Columbia (South Carolina)
1992.

A number of the works mentioned under §3 are relevant here also.

lxxxii COMMENTARY ON ACTS

V THE THEOLOGY OF ACTS

So far as Acts makes a specific contribution to the theology of the NT

this is to be found not so much in the treatment of particular doctrines
as in an understanding of the possibility and course of Christian
history and especially of the Christian mission. It may nevertheless
be worth while to collect some of Luke’s views in relation to a
number of important themes before attempting to sum up the general
convictions that underlie his presentation of Christian origins.1
(a) Eschatology. To the modern student of theology biblical
eschatology is a problem. It was not a problem to Luke; he knew
what to make of it and had already established a position which was,
at the end of the first century, quite satisfying. It had long been
necessary to recognize the existence of an interval between resurrec-
tion and parousia. In the beginning this may not have been expected
or, when it happened, understood. The crucifixion (it was believed)
would be followed by the vindication of Jesus, and this might be
expressed in terms of Daniel 7 (a coming with the clouds) or of
Daniel 12 (a rising up of those who sleep in the dust). Luke saw a
clear distinction; it appears both in the Third Gospel and in Acts.
Jesus was raised from death by God; after a period of forty days
during which he appeared to various disciples he went up into
heaven, his ascent modelled on Daniel’s picture of the coming of the
Son of man put into reverse; thence at file right moment he would
return. All these events might be described as eschatological in the
sense that they belonged to the winding up of God’s purpose for the
world. God himself had determined that the last act of the play
should be a complex one, and had already declared this in the OT.
The basic text is given a prominent place in the account of the gift of
the Spirit on the Day of Pentecost. Altering the text of Joel 3.1-5
(LXX) Luke writes of the last days, which cover all Christian history.
They are initiated by the gift of the Spirit, described by Luke before
the text is quoted, and their end will be heralded by cosmic
phenomena (the darkening of the sun, and so forth) which have
manifestly not yet happened; the great and glorious day of the Lord
has not yet arrived. The church exists in the intervening period, its

1For an admirable and very much fuller analysis of Lucan theology see F. Bovon,
Luke the Theologian, Thirty three years of Research (1950-1983), ET Allison Park,
1987; J. A. Fitzmyer, Luke the Theologian: Aspects of his Teaching, London, 1989;
see also the Bibliography.
 INTRODUCTION lxxxiii

life determined by what has already happened and by what is still to

come. The period is provided by God so that witness may be borne to
Jesus (Acts 1.8); Luke describes the record of this witness from
Jerusalem as far as Rome, but there is no hint, as his book draws to a
close, that the end may be expected soon. That there will be an end,
which will include the return of Jesus and judgement carried out by
him, is plainly assumed. It is categorically affirmed at the beginning
of the book (1.11) and several times repeated, especially in the earlier
chapters (2.19-21; 3.20, 21; 10.42; 17.31; (26.6)); it is implied also
by some of the passages that refer to salvation, for this is in part
salvation from the judgement that will fall upon the sinful human
race. The coming of the Messiah waits upon repentance and faith
(3.20, 21), which Luke hopes to encourage, but it is clear that his
interest is in the present and he makes little use of eschatological
threats to encourage response. In this respect Acts is to be contrasted
with Luke’s gospel; on this see especially Wilson (Gentiles 67-80:
‘The two strands in Luke’s eschatology’). His conclusion is correct.
Tn Acts we have a further development of one of the two strands we
found in Luke’s Gospel to the exclusion of the other. Luke has
moved away from belief in an imminent end. One of his methods of
doing this is to schematise and objectify the eschatological timetable.
Another is to substitute Ascension theology, the present activity of
the exalted Lord in his Church, for belief in an imminent end. This is
done not so much by dogmatic statement as by the concentration on
this element in Acts. The time-scheme of Acts allows for a hiatus
between the Resurrection and the Parousia in which the Church can
exist and grow’ (p. 80). See also the excellent observations of E.
Grösser in Kremer, Actes 99-127, especially his conclusion on p.
127. Consideration of Luke’s eschatology leads immediately to a
second theme.

(b) The Holy Spirit. According to 2.17, 18 the gift of the Holy
Spirit is (after the resurrection and ascension of Jesus) the first part of
the fulfilment of the eschatological hope; it initiates the period in
which (see above) the church can ‘exist and grow’. References to the
Spirit are as frequent as references to the parousia are few. Luke
writes ‘non tam apostolorum, quam Spiritus sancti Acta describens,
sicut prior liber Acta Jesu Christi habet’ (Bengel, 389). This is
already made clear in 1.6-8 (it is not for disciples to know God’s
eschatological programme but they will be empowered for witness
by the gift of the Spirit) and confirmed in 2.1-4 (after a period of
waiting the church receives its full active being in the gift of the
Spirit at Pentecost). It is not strictly consistent with this definitive,
founding gift of the Spirit that from time to time afterwards
Christians are said to be, on special occasions, filled with the Spirit—
4.8, 31; 6.3,5; 7.55; 9.17; 11.24; 13.9, 52. In addition people are said
lxxxiv COMMENTARY ON ACTS

to be baptized with the Holy Spirit; the Holy Spirit comes upon them,
or falls upon them. It is only Luke’s mode of expression that is open
to logical criticism; his thought is clear and consistent. Believers
receive the Holy Spirit as the basic constituent of their believing life,
and in times of special need they receive special gifts of the Spirit
that enable them to speak or act appropriately. The Holy Spirit has
already been at work in the OT (e.g. 1.16; 4.25; 7.51; 28.25), but is
now (as prophesied by Joel—2.17, 18) given generally, so that the
possession of the Spirit by Gentiles is a proof that they are rightly
included in the church’s mission (10.44, 45, 47; 11.15, 16; 15.8). In
other ways also the Spirit directs the mission, directing Philip to meet
the Ethiopian (8.29) and subsequently conveying him to Azotus
(8.39f.), telling Peter to go with Cornelius’s messengers (10.19f.;
11.12), forbidding Paul to speak in Asia and to travel to Bithynia
(16.6, 7), and ordering him to return to Asia (19.1—si v.1.). If in
21.4 the Spirit seems to get the directions wrong this is no doubt to be
put down to a piece of careless writing by Luke. The Spirit is offered
as a gift to those who repent and believe. The offer is explicit on the
Day of Pentecost (2.38); cf. 9.17; 13.52. The gift is often but by no
means always associated with baptism, sometimes with the laying on
of hands; on this see below, pp. xcif. The Spirit, present as a guiding
principle in the Christian fellowship, is abused when Christians
practise deceit; so Ananias and Sapphira deceive the Holy Spirit and
lie to God (5.3,4); it goes too far to deduce from this text the deity of
the Holy Spirit, though it is unlikely that Luke would have wished to
deny the proposition had it been made to him. The Spirit is associated
with the Father and the Son at 2.33, a verse that is not without
ambiguity. Jesus has been exalted to (or by) the right hand of God; he
has received the gift of the Holy Spirit; on the precise meaning of this
see I.149f. Having received the promise of the Spirit (the promised
Spirit?) he pours out (the verb ἐξέχεεν points back to the quotation
from Joel in 2.17) the Spirit whose presence you can detect in
observable phenomena. The Spirit is at least a third form of divine
activity.
What was observed (heard rather than seen) was the miraculous
speaking with tongues of 2.4, 5-12; it is not clear whether the
phenomena of 2.2, 3 were observable outside the house or room in
which the disciples were assembled. Speech is in Acts the character-
istic mark of the Spirit’s presence, sometimes in glossolalia (2.4;
10.46; 19.6), sometimes in prophecy (2.17,18; 11.27; 13.1-3; 21.(4),
(9), 10, 11), sometimes in proclamation (e.g. 4.31). It is striking that
Luke (unlike Paul) does not see the work of the Spirit in the moral
renovation of human life. This does not mean that he did not believe
in the power of God to change human behaviour; it does however
mean that he was impressed by what may be regarded as the
shallower and showier aspects of Christian life—mainly no doubt
 INTRODUCTION lxxxv

because they were easy to describe and likely to impress the reader as
they had impressed him.

(c) Christology. As truly as the Third Gospel, though in a different

manner, Acts is an account of the works of Jesus the Messiah. The
opening verse refers back to the contents of the former treatise as
concerned with all that Jesus began to do and to teach. It is a probable
though less than certain implication that the new treatise is about all
that Jesus continued to do and to teach; even if he did not intend it
Luke would not have disagreed. The disciples are to be his witnesses
(1.8; for their continuing witness cf. 1.22; 2.32; 3.15; 5.22; 10.39,41;
13.31; 22.15, 20; 26. 16, for the word μάρτυς) and he is the
foundation of their testimony (2.22; 3.13; 4.2, 10, 33; 5.42; 8.12, 35;
9.20, 27; 10.38; 11.20; 13.32f.; 16.31; 17.3, 18; 18.5, 28; 20.21;
24.24; 28.23, 31). He is the agent of miracles (e.g. 3.6; 4.10; 9.34).
Sometimes he acts indirectly, through the Spirit (who may be called
the Spirit of Jesus, 16.7); angels also play their part (8.26; 12.7;
27.23); but Jesus also speaks and acts in his own person, and on
specially notable occasions, confronting Saul (9.5; 22.8; 26.15), and
encouraging him (18.9; 23.11). He is a supernatural person, able at
will to participate in historical events.
To this it is necessary to add that what he does he does as the agent
of God the Father. In the healing of the lame man at the Temple gate
it may be the intention of 3.13 to suggest that God has glorified his
servant Jesus by using him as the agent of cure; in 4.30 the disciples
pray to God that he will stretch out his hand for healing and that signs
and portents may be done through the name of his servant Jesus.
Jesus is the agent not only of healing (σωτηρία) in a physical sense
but of the salvation that means belonging to the elect people who will
be brought safely through the perils of the last days (4.12).
Jesus is described by a number of titles, some of them widely used
elsewhere in the NT. He is of course the Christ, the Messiah, the
Lord’s anointed king. Or should we say, he will be the Christ? This
question is raised with reference to 3.20; see the notes on this verse in
I.204f. It is not likely that the verse refers to an appointment lying
still in the future; Luke undoubtedly believed that Jesus was the
Christ from his birth onwards; see especially Lk. 2.11. It is not
impossible that a source used by Luke took a different view, but it is
improbable. It does however seem likely that Acts 2.36 meant
originally that Jesus became Christ at the time of his resurrection,
though Luke himself must have been able to accommodate the verse
to his own view. Lord (κύριος), also occurring in 2.36, is a word that
interprets χριστός for the Hellenistic world. It is not correct to say
that since in the LXX κύριος is the rendering of the use of the
word in Acts (and in the NT generally) in itself implies the divinity
of Jesus. At least it does not imply that Jesus is to be identified with,
lxxxvi COMMENTARY ON ACTS

or placed on the same level as, the God of the OT. Luke, quoting Ps.
110.1 (εἶπεν ὁ κύριος τῷ κυρίῳ μου) in 2.34, shows himself to be
fully aware that the word κύριος could be used in two senses, and he
took Jesus to be the second κύριος to whom the first κύριος speaks.
The word is at once an identification of Jesus with the Davidic king
and an approach to the Gentile world with its gods many and lords
many (1 Cor. 8.5). Lord undoubtedly means that Jesus was, and was
understood by the Christians to be, one who enjoyed absolute
authority, to whom they owed absolute obedience. He himself
however stands under the absolute authority of God the Father: it is
he who has made Jesus Lord and Christ. There is thus in Acts, as in
much of the NT, an element of subordinationism, but it is the
subordination of an obedient Son, active in the Father’s service.
Jesus is the Son (υιός) of God in only two passages. It may not be a
coincidence that it is Paul who proclaims Jesus as the Son of God
(9.20) and quotes Ps. 2.7 at 13.33. It would be wrong to read a
metaphysical relation into Luke’s reporting, whatever Paul himself
may have believed. Only in chs. 3 and 4 (3.13, 26; 4.27, 30) is Jesus
spoken of as the Servant (παῖς) of God; this may reflect the use by
Luke of a source. It is mistaken to suppose that there is a reference
here to the Servant of the Lord in Deutero-Isaiah. In 4.25 the word
παῖς is applied to David, and in the OT it is used of kings, prophets,
and other outstanding figures. There may however be an allusion to
Isaiah 52.13 in 3.13—to the glorification rather than the suffering of
the Servant. Luke is here using traditional material; when, in 8.32,33
he quotes Isaiah 53 the word ‘servant’ does not occur.
Other Christological terms occur infrequently. One is αρχηγός:
3.15; 5.31. The word may mean prince, leader, author, origin; see the
discussion in I.197f., 290. At 5.31 it is coupled with σωτήρ, which
occurs again at 13.23; other words of the family (σῴζειν, σωτηρία,
σωτήριον) occur also. It is uncertain whether ό δίκαιος (3.14, where
we have ό άγιος καί δίκαιος; 7.52; 22.14) is a title or a description;
see I.195f., 377. Ναζωραῖος (2.22; 3.6; 4.10; 6.14; 22.8; 26.9; cf.
24.5, where it is used in the plural, of Christians) is taken by some to
be a title, not related to the place Nazareth; but cf. 10.38, ό από
Ναζαρέθ, and see I.140. The expression Son of man, so common on
the gospels, is used once only, by Stephen, at 7.56. It is impossible to
deduce anything about the meaning of the term from this one verse,
but it was probably chosen as one suitable to denote a person now in
heaven but waiting to come with the clouds at some future time.2 At
2.22; 17.31 the word ἀνήρ is used unselfconsciously—and of course
quite properly—of Jesus; in the latter verse, where the man is to be

2In Stephen’s speech see also words that might have been used in Christology but
are in fact used of Moses: ἄρχων, δικαστής, λυτρωτής, 7.35. See I.363f.
 INTRODUCTION lxxxvii

the judge of mankind, it recalls part of the connotation of Son of

man.
There is in Acts no profound Christological thought;3 yet it is clear
that Jesus Christ of Nazareth is the person who initiated and will
conclude the whole story and directs the whole course of it. His
career, marked by portents, signs, and mighty works, was sufficient
to show that God was with him, not merely as he had been with, for
example, the prophets, but in a unique (though undefined) sense. He
was killed through an ignorant error on the part of the Jews, both
rulers and people, who handed him over to the Romans, but God
soon put the mistake right by raising him from death, thereby
confirming—or creating—his status as Messiah (a term meaningful
to Jews) and Lord (which would make sense to Gentiles). The risen
Jesus continued long enough on earth to prove that he was truly alive
and then ascended to his rightful place at the right hand of God. This
proved that he belonged essentially to the same order of being as the
Creator, the Lord of the OT, though within that order
secondary. The sense in which a second being might be said to share
the throne of God would have constituted a problem, and engendered
disputes, for some Jews; not, apparently, for Luke. Jesus had now
sent the Spirit so as to bring into being a renewed Israel, soon to be
enriched by the addition of Gentiles.4 Jesus, crucified and risen, thus
brought salvation to the mixed community: the forgiveness of sins,
the inspiring and sanctifying of the Spirit, and the pledge of safety in
the eschatological troubles that still lay ahead.
(d) The Church. The community just mentioned was an eschato-
logical entity, living within the period of fulfilment initiated by the
gift of the Spirit and awaiting its end in the portents described by Joel
(Acts 2.17-21). It was also however an historical phenomenon,
manifested in a number of groups of believers scattered through
various parts of the Mediterranean world. These groups came into
being through the preaching of apostles and others. As their hearers
accepted the word that they heard they found themselves gathered
into believing companies and shared a common life. There was little
by way of outward organization to mark this common life;5 fellow-
ship was constituted by a shared belief that Jesus in truth was what
the preachers had declared him to be, offered salvation, and claimed
obedience. The Spirit generated speech and no doubt enthusiasm;
and Jesus’ command of mutual love was expressed in the sharing of
goods (2.44, 45; 4.32, 34, 35, 36, 37; 5.1-11; 6.1). How far this
practice, which according to Luke resulted in the absence of poverty

3P. Vielhauer in his essay on the Paulinism of Acts (FS Schubert 33-50; original
version in EvTh 10 (1950-1951) 1-15) took Acts to be pre-Pauline in Christology.
4See below, sections (h) and (j).
5See however below, sections (e) and (f ).
lxxxviii COMMENTARY ON ACTS

in the community, resembled and was related to similar customs

observed at Qumran is disputed; see I.167-9, 263.
At first there was one such community, in Jerusalem. At 5.11; 8.1,
3; 11.22; 12.5; 15.4,22; and perhaps 18.22, it is denoted by the word
ἐκκλησία. As the Gospel was taken out from Jerusalem into the
world this word came to be applied to other local Christian groups: to
the Christians of Antioch at 11.26; 13.1; 14.27; 15.3; to the
Christians of Ephesus at 20.17.6 The word is used of the groups of
converts made in Paul’s first missionary journey: 14.23; 16.5; of
those in Syria and Cilicia at 15.41. An ἐκκλησία thus appears to be a
local group of Christians. There are two passages that may suggest a
wider meaning. It is often maintained that this is so at 9.31, which
refers to the church through the whole of Judaea and Galilee and
Samaria. This may be the beginning of such a use, but the geo-
graphical designation may be taken to indicate a ‘local’ church
residing in more than one town (see I.473-5). More significant is the
reference (20.28) to the church of God which he acquired by his own
blood, a strange expression in that it appears to refer to the blood of
God. However it is interpreted the verse speaks of a single body of all
Christians which God has redeemed and constituted through the
bloody, that is, sacrificial, death of Christ, ἐκκλησία, then, is here the
world-wide company of the redeemed, ecclesia cattolica. It is
important that the same word is used for both purposes. Luke may
not have seen deeply into the meaning of his own language, but it
implies (and this is borne out elsewhere in the NT) that each local
group of Christians is not merely related to the total church but in
fact is the total church in the place in which it exists. It is also
important that the word ἐκκλησία is used in the LXX as a rendering
of the people of God, living (sometimes rebelliously) under his
direction. The connection is underlined by the occurrence in 7.38 of
τῇ ἐκκλησία ἐν τή ἐρήμφ, the people of Israel between Egypt and the
Promised Land.7 Luke was familiar with the LXX and there can be
no doubt that this use of έκκλησία would be in his mind. In the OT
the Lord had acquired a people by the mighty acts that he performed
in Egypt; he had now acquired a people, racially mixed, by the
shedding of Christ’s blood. Why the death of Jesus should be
regarded as a sacrifice and how the sacrifice could have the effect of
constituting a people are questions that Luke did not address. For
questions that did concern him see below, especially (h) the Jews and
(j) Gentiles and the Gentile Mission. For the constitution and actions
of the church see (e), (f ), and (g).
(e) Apostles and Ministers. Luke’s account of the church is given
6If at 18.22 ἐκκλησία does not refer to the church of Jerusalem it will refer to that of
Caesarea.
7The use of ἐκκλησία in 19.(32), 39, 41 for the duly constituted town meeting in
Ephesus illustrates the common secular meaning of the word.
 INTRODUCTION lxxxix

for the most part in terms not of its rank and file but of its leading
members. Continuity between the earthly life of Jesus of Nazareth
and the church after the time of the resurrection is provided by
twelve men. Eleven of these (1.13) had belonged to the group of
twelve chosen by Jesus (Lk. 6.13-16); from this group Judas Iscariot
was removed by defection and death (Acts 1.16-20) and Matthias
was added to it (1.21-26), the use of the lot being taken to
demonstrate that he equally with the rest had been chosen by Jesus
(ἀνάδειξον δν ἐξελέξω, 1.24). These Twelve are referred to again
under that title at 6.2, and the title is implied at 2.14, where Peter
stood up ‘with the Eleven’ (cf. 1.26). They are thus responsible for
the initial proclamation of the Christian message and are still
regarded in ch. 6 as responsible for it, since they profess the intention
to give themselves to prayer and the ministry of the word, leaving the
administration of charity to Stephen and his colleagues (6.2-6),
whom, on the nomination of the people, they appoint.
The Eleven when they are joined by Matthias are described as
apostles, a term that accords with Lk. 6.13 (οῦς καί αποστόλους
ὠνόμασεν) and is used much more frequently than the numeral,
though not after 16.4. It is sometimes but not always suggested fairly
definitely that the Apostles are identical with the Twelve, and there
are only two verses (14.4, 14; on both see the notes, I.666f., 671f.,
678f.) where persons not belonging to the Twelve are called apostles.
Only in these verses is Paul (and Barnabas with him) called an
apostle, and it is clear that Paul (and presumably Barnabas also) did
not fulfil the conditions laid down in 1.21, 22 for appointment. The
rest of the NT makes it clear that the word απόστολος was used in
more senses than one,8 and it is probably best to suppose that the
(Antiochene?) source used by Luke in ch. 14 referred to Paul and
Barnabas as apostles, that is, envoys, or missionaries, of the church
of Antioch, and that Luke omitted to bring the reference into line
with his use elsewhere.
Luke has some notable stories about Peter: his witness on the Day
of Pentecost; his healing of the lame man at the Temple gate and the
speech that followed; his appearance before the Jewish Council; his
working of miracles, told both in general and particular terms; his
visit to Samaria; his dealings with Cornelius; his release from
Herod’s prison; his share in the Apostolic Council. In chs. 3, 4, 8
John accompanies Peter. In ch. 12 we read of the execution of James,
John’s brother. Luke has nothing else to tell us of the Twelve
Apostles and it is clear that for him their main significance as a group
is simply that they exist, and by their existence witness to and (so far
as this could be done) guarantee the continuity between Jesus and the

8See The Signs of an Apostle (London, 1970, Carlisle, 1996), with the summary on
pp. 71-3.
xc COMMENTARY ON ACTS

post-resurrection church. Only Peter (so far as Luke’s narrative goes)

makes any other contribution to the life of the church. Their function
is important (note the contrast in 13.31, 32, and cf. 10.39) but it
neither could nor needed to be repeated. There are no more apostles;
the word as used in the sense of 2 Cor. 8.23 (and perhaps Acts 14.4,
14) dropped out of use. Paul, der dreizehnte Zeuge,9 as was noted
above, does not meet with the requirements of apostleship as Luke
sets them out, yet he was, for Luke and no doubt in fact, the
outstanding missionary of the first generation. He also exemplified,
for Luke, Luke’s understanding of Judaism as the heir of the OT
people of God, who nevertheless must now understand both the OT
and their own vocation in the light of the Gospel. How is he to be
described? Luke has no category for him; he is a σκεύος εκλογής
(9.15), a tool selected by God for a special purpose, a man who
would bear Christ’s name before Gentiles and kings and the children
of Israel. He is not subordinated to the Twelve; the only passage that
suggests this is 13.32, and this is a matter not of subordination but of
distinction. His address to the Ephesian elders in 20.18-35 is
important both as showing his understanding of pastoral ministry, as
exercised by himself and expected of the elders, and as indicating the
origin of the elders’ ministry. They have been appointed neither by
the church in which they ministered nor by Paul but by the Holy
Spirit (20.28). No provision for future ministers is suggested here;
they will come from the same source. To say this is not to deny
human participation in the process; especially in the earliest days
Paul must himself have played a leading part in making appoint-
ments (14.23), but 13.1-3 pictures as 20.28 does appointment not so
much to an office as to a task by the Holy Spirit, working, one may
suppose, through the prophets of 13.1. The task of prophets and
teachers was to prophesy and to teach, to utter messages communi-
cated by the Holy Spirit and to maintain and apply the basic
traditions of the faith. Elders would like Paul himself declare the
counsel of God (20.27) and maintain discipline in the community.
Working for their own living they would out of their own as well as
out of the church’s resources aid those who were in need. This task
elsewhere is the work of deacons, but the word διάκονος does not
occur in Acts. Elders, called επίσκοποι as well as πρεσβύτεροι,
perform all the tasks of ministers; Acts 6 is not intended to be an
account of the origin of the diaconate as an order of ministers (see
I.304).
(f) Baptism and the Christian Meal. These twin topics (they seem
to us to be twins; it is by no means certain that they would have
seemed so to Luke—like other NT writers he had no word sacrament
to unite them) give rise to a number of puzzles. If one concentrates
9The title of a book by C. Burchard (Göttingen, 1970).
 INTRODUCTION xci

on certain parts of Acts (chs. 1, 2, 8, 9-11, 16, 18, 19, 22), or rather
on parts of these chapters, baptism seems to be the normal and
universal way into the Christian church. What are we to do? ask the
crowd on the Day of Pentecost. Repent, and let each of you be
baptized, Peter replies (2.37, 38). The converted Samaritans respond
in the same way; Cornelius and his friends, the Philippian gaoler and
his household, many of the Corinthians, are baptized. If however we
turn to the other chapters (and to parts of those listed above) there is
silence. The Temple crowd are urged to repent and have their sins
blotted out; they are not told to be baptized. There are no baptisms in
Luke’s account of the ‘first missionary journey’, though churches are
established (14.23). Apart from Lydia and the Philippian gaoler there
are no baptisms in Macedonia or in the main account (apart, that is,
from 19.5) of Paul’s work in Ephesus. No baptisms take place after
ch. 19, even in Malta, where Paul made so deep an impression.
Again, if we ask, Does the gift of the Spirit precede, accompany, or
follow the rite of water-baptism? we get different answers in
different parts of Acts. If we ask, Must baptism be complemented by
the laying on of hands? there is no consistent answer. What is meant
by baptism in (or with) the Spirit? Is it a consequence of water-
baptism or is it independent of water-baptism? Who are the proper
recipients of water-baptism? Adult believers only, or infants also?
None of these questions can be answered with any confidence on the
basis of Acts. There is quite enough of baptism in Acts to make it
clear that Luke was familiar with the practice; we can hardly fail to
conclude that baptism was not, as is commonly supposed, a universal
custom in the early church, or at least that some of Luke’s sources
(such as that based on Antioch) were not interested in baptism.
When baptism is mentioned, and if it is specified (and since a good
deal of baptizing was going on in the first century10 it must have been
specified—Our baptism is ...), it is said to be in the name (ἐπὶ or ἐν
τῷ ὀνόματι; εἰς τό ὅνομα) of Jesus.11 Behind this no doubt lies the
Hebrew which here would mean ‘so as to come under the
authority of. Taken into Greek the expression gains a financial
connotation: ‘so as to be added to the account of, that is, ‘so as to
become the property of. This seems to be the primary thought in
Luke’s mind. The converts (one may say, whether baptized or not)
become Christ’s men or women. From this consequences follow.
They become members of Christ’s people. Christ bestows upon them
the gift of the Spirit. Their sins are forgiven. These units form in
Luke’s mind a single whole and he is not concerned to specify an
order in which they occur. At 2.38 it is not quite stated but it is
implied that the gift of the Spirit follows upon baptism: Let each of
10See J. Thomas, Le Mouvement Baptiste en Palestine et Syrie (150 αν. J.-C.—300
ap. J.-C.), Gembloux, 1935.
11There is no hint of a Trinitarian formula such as that of Mt. 28.19.
xcii COMMENTARY ON ACTS

you be baptized ... and you will receive ... The opposite order
occurs at 10.44-48. The Spirit fell upon Cornelius and his friends
and to this Peter’s response was, ‘Can we fail to baptize these people
who have received the Holy Spirit just as we did?’12 There is a further
complication in ch. 8 (cf. 19.6) when Philip baptizes the Samaritans
and it is not said that they received the Holy Spirit; Peter and John
arrive, impose their hands and pray, and the Spirit is given. Is the
imposition of hands necessary? It is not mentioned in the other
stories we have considered. Is it a rite that only an apostle can
execute? We do not know the answers to these questions; see the
notes on the various passages. We cannot fail to conclude that Luke
gives an unclear account of baptism. He had no fixed principles about
its practice, or perhaps its meaning. For Paul baptism meant crucifix-
ion and burial with Christ, and he seems to assume that this is a
common Christian understanding (Rom. 6.3). In fact, what Luke says
about baptism matches his Christology. He knows quite well that
Jesus was crucified and that his resurrection implies his previous
death. But for him Christ crucified is not his central theme, the theme
of all themes that it was for Paul (1 Cor. 2.2); therefore the Christ to
whom the converts came to belong was not so specifically and
exclusively Christ crucified as he was for Paul. Inevitably, crucifix-
ion dropped out of baptism.
Did Luke’s church have a eucharist? He does not say so. Acts
contains several references to meals taken by Christians. The
description of the Christian fellowship that resulted from Peter’s
preaching on the Day of Pentecost includes 2.42, ‘They continued in
the teaching of the apostles and the fellowship, in the breaking of
bread and the prayers;’ 2.46, ‘continuing with one accord in the
Temple and breaking bread at home they partook of food with
gladness and simplicity of heart;’ At 20.7 it is said that Paul and his
companions with the church at Troas met to break bread; at v. 11,
after the incident of the young man who fell out of the window, Paul
went up, broke the bread and ate, and continued his address. At
27.34-36 Paul urged his fellow seafarers to partake of food; this
would be for their welfare (σωτηρία); when he had said this and
taken the loaf he broke it before them all and began to eat; and they
all cheered up and themselves partook of food. On all these passages
see the notes. It is striking that all of them refer to the breaking of
bread; the last is expressed in language that is particularly close to
that of Lk. 22.19, 20 (the Last Supper) and 9.16, 17 (the feeding
miracle); see also Lk. 24.30, 35. There is however no reference, as
there is in the account of the Last Supper (in 1 Cor. 11.23-26 as well
as in the gospels), to the drinking of wine. J. Jeremias (see the notes
on 2.42) is probably right in taking the breaking of bread to be a

12The question whether the apostles were baptized cannot be answered.

 INTRODUCTION xciii

Christian use, describing a specifically Christian meal, wrong in

taking Luke’s intention to be the concealment from non-Christian
readers of a secret Christian rite confined to believers. Luke had
already given an account of the meal including explanatory words
(‘This is my body’, etc.) not only for the bread but for the wine; and
he goes out of his way in 27.35 to say that what Paul did he did
ἐνώπιον πάντων. The four passages in Acts describe a Christian
fellowship meal which derived its special significance not from what
was eaten and drunk but from the fellowship of Christians who, as
they ate and drank, would not forget the Lord who had given himself
for them, and in giving himself for them was giving himself to them.
If others were present they would get out of the meal what they
could—and this might prove, even in the midst of the storm, to be
more than physical sustenance. Whether there was wine to drink or
not might well depend on circumstances. There is no indication that
the meals referred to in Acts were in any way connected with the
Passover, at which the drinking of four cups of wine was obligatory.
A more difficult problem is raised by the fact that Paul’s account of
the Christian meal (1 Cor. 10.16, 21 as well as 11.23-26) includes
specific reference to wine. It is natural—and correct—to say that
there was probably a great deal of variety in Christian observance in
the first century, but Luke (whatever we make of the We-passages)
was some sort of Paulinist. It is probably best to suppose that though
‘the breaking of bread’ was not a cypher designed to keep the
proceedings secret it had become a formula that pointed to a whole
meal and made no attempt to specify all the substances consumed.
It is hardly open to doubt that what was important to Luke was not
the symbolic significance of what was done, of what was eaten and
drunk, but the shared life that commensality represented. In 2.42 the
breaking of bread appears in a context determined by the teaching of
the apostles and prayers and by fellowship (κοινωνία) which in part
at least receives its definition from the fact that those taking part had
their property in common (κοινά), and used their resources for the
relief of the poor.
(g) Frühkatholizismus. It has been held that the account in Acts of
apostles and ministry, of baptism and eucharist, points to the
development of the primitive church into the stage of early Catholi-
cism. Before this position can be considered it is important to know
precisely what is meant by Frühkatholizismus. It is helpful to go back
to an early stage in the discussion, the controversy between Hamack
and Sohm on the meaning of Catholicism.13 The significance of the
13Best studied in A. Hamack, Entstehung und Entwicklung der Kirchenverfassung
und der Kirchenrecht in den zwei ersten Jahrhunderten (Leipzig 1910), especially the
criticism of R. Sohm’s ‘Wesen und Ursprung des Katholizismus’, pp. 121-86 (ET
175-258). Sohm’s article (in Abhandlungen der Phil.-Hist. Klasse der KSächs.
Gesellschaft der Wissenschaften 27.3 (1909)) is extensively quoted and summarized.
xciv COMMENTARY ON ACTS

discussion is brought out by Conzelmann14 as follows. ‘Frühkatholi-

zismus liegt noch nicht vor, wo ein Traditionsgedank da ist. Dieser
gehört zur Theologie selbst. Der entscheidende Einschnitt liegt da,
wo die Tradition institutionell, durch Bindung an ein Amt und an
Sukzession in diesem Amt, gesichert wird. Er liegt noch nicht vor,
wo eine feste Amtsordnung besteht—mag diese auch bereits eine
monarchische Spitz haben—sondern erst da, wo das Amt heilsmit-
tlerische Qualität erhält, wo die Wirkung von Geist und Sakrament
an das Amt gebunden ist. Mit Bultmann zu sprechen: Der entschei-
dende Vorgang ist die Verwandlung der regulierenden Bedeutung
des Kirchenrechts in eine konstitutive (Bultmann 449f.)’ Sohm and
Hamack treated Spirit and Office, or law, as opposite alternatives;
Bultmann15 qualifies, almost reverses the opposition. According to
Bultmann, Holl showed ‘dass das Wort des Charismatikers als
autoritatives Wort Ordnung und Tradition schafft’. The truth is better
put by von Campenhausen, because put in reciprocal form.
‘Entscheidend ist vielmehr die feste Korrelation, in der das Geist
vom Anfang an zum Begriff des Wortes oder Zeugnisses steht, die
beide auf die Person Jesu zurückfuhren. Sie sind die entscheidende
Wirklichkeit, und machen zusammen eine Verabsolutierung des
Geistes gegen die Tradition oder der Tradition gegen den Geist
gleichermassen unmöglich.’16
The theology is sharpened when it is given a specific historical
reference in relation to Acts by E. Käsemann, who says of Frühka-
tholizismus that it ‘nichts anderes als die kirchliche Abwehrbewe-
gung gegenüber der drohenden Gnostisierung ist.’17 Eschatology is
replaced by Heilsgeschichte. Luke’s story of the mission is not that
of testimony to the proclaimed Christ but the way and work of the
world-embracing Christian agency of salvation (Heilsanstalt). ‘Das
Wort ist nicht mehr das einzige Kriterium der Kirche, sondern die
Kirche ist die Legitimation des Wortes, und der apostolische
Ursprung des kirchlichen Amtes bietet die Gewähr für eine legitime
Verkündigung’ (p.21). This is a matter on which Luke is perhaps
hardly theologian enough to wish to take sides.
These authors have seized upon an important theological contrast;
this does not answer the question whether they have rightly evaluated
the historical evidence of Acts: indeed they do not all evaluate it in
the same way. The following have been alleged as features of the
narrative in Acts that point in the direction of Frühkatholizismus. (1)
14Theologie 318.
15He invokes K. Holl, unfortunately without giving a reference, also H. von Soden
in Studium Generale 4 (1951), 351ff. and H. von Campenhausen in Kirchliches Amt
und Geistliche Vollmacht (Tübingen, 1953), 324, 325. The whole of this book is in
varying degrees relevant.
16Kirchliches Amt... (as in n. 15), 325.
ZThK 54 (1957), 20; see also ‘Sätze Heiligen Rechtes im NT’, NTS 1 (1955),
17
248-60.
 INTRODUCTION xcv

The emphasis on the importance of the apostles, and (2) the notion of
a succession from the apostles as constituting the being of the church
and providing its ministry; (3) the notion of a ministry which is
essential to the being of the church, and (4) stress on the importance
of the sacraments as means by which the ministry nourishes the life
of the church.
To some extent these matters have already been discussed. It is
true that Luke lays considerable stress on the apostles.18 They are the
first believers and, with certain women, constitute the first body of
Christians. Their number is important and, after the death of Judas,
they are restored to their original strength. The twelfth apostle is
chosen in such a way as to show that the choice is made by Jesus, as
when the original Twelve were appointed. It is not clear that they
alone received the Spirit on the Day of Pentecost but it was they who,
led by Peter, stood up to speak to the assembled crowds. Converts on
that day continued steadfastly in the apostles’ teaching, and it is clear
through chs. 3, 4, and 5 that they, headed by Peter and John, are the
leaders of the Christian movement. In ch. 6, denoted by the
designation ‘the Twelve’, they solve the problem of the neglected
widows by appointing seven to serve tables while they continue to
devote themselves to prayer and to the service of the word. Accord-
ing to 8.1 they alone escaped persecution and remained in Jerusalem,
but we hear little more of them. In ch. 9, after initial distrust, they
accept Saul, and Peter makes his way to Caesarea, where Cornelius
becomes a Christian. This event is discussed in ch. 11, and Barnabas
is sent to see what has been happening in Antioch, as Peter and John
had been sent to Samaria in ch. 8. In ch. 12 James is killed and Peter
escapes, leaving Jerusalem for ‘another place’. After this we learn
that the apostles were present at the Council of ch. 15, but only Peter
speaks. We hear no more of them, and it seems probable that Luke
had no further information about them.
They were important because they had accompanied Jesus during
his ministry and thus served as a guarantee—or perhaps a symbol—of
the fact that the actions of the post-resurrection church were a valid
continuation of the work of Jesus. For this they are important, indeed
indispensable, but they execute no administrative authority, though
Peter rebukes Ananias and Sapphira and the Twelve appoint the
Seven. Their importance is unique to themselves; it cannot be trans-
mitted to a succession because they and they alone bridge the cleavage
between the time of Jesus and the time of the church, at the same time
being witnesses to the resurrection, which meant that the Jesus who
was crucified was also the continuing Lord of the church.
Succession is thus in a strict sense impossible; the primary
function of the twelve apostles was not transmissible. We have seen

18See section (e) above.

xcvi COMMENTARY ON ACTS

(p. xci) that it is impossible to infer from Acts that baptism was
universally practised, and language specifically ‘sacramental’ is not
used of the Christians’ common meal. The church in Acts has those
whom we may call ministers, but of no uniform kind. At Ephesus
there are πρεσβύτεροι, also called επίσκοποι, but at Antioch there
appear to have been no presbyters, but prophets and teachers (13.1).
All probably performed very similar functions. Philip was an
evangelist; and there were the great Pauline assistants, such as
Barnabas and Silas. The apostles could, as we have seen, have no
successors, but their task of bearing witness to the risen Jesus was
shared by others, and Luke apparently did not think that any special
authorization (beyond recognition as a Christian) was necessary for
this fundamental task; Stephen was a pre-eminent witness, but the
only commission he received (in Luke’s narrative) was to care for
Hellenist widows. Luke is undoubtedly concerned for continuity, but
this must be distinguished from succession; in 20.28 Paul does not
say that he had ordained the presbyter-bishops of Ephesus, but that
the Holy Spirit had made them what they were. Nothing here
corresponds with Clement’s account of the appointment of the post-
apostolic ministry (though that is informal enough—1 Clement
40-44) or with Ignatius’ insistence upon the indispensable centrality
of monepiscopacy and a threefold ministry. There is at least one rite,
which later became a sacrament, not (as far as we know) universally
practised. These are the main historical phenomena.
It may be said that there are some hints of what may be regarded as
elements in a developing ‘catholic’ structure of the church, but they
are not combined in such a way as themselves to constitute a catholic
structure. For this something more than a few more or less ambig-
uous elements is required, though these may point forward to the
development of further change in the future. The important word in
the quotation from Conzelmann (above, p. xciv) is institutionell. Any
group of people that persists recognizably through a number of years
persists by means of tradition: the older teach the younger what the
group stands for, and when they can no longer continue their work
find others who will take it over from them. The important question
is what is the vital force that keeps the group alive, and how is it
maintained and transmitted. In Acts (pace Käsemann), the vital
principle is the word that is committed by Jesus to those who follow
him and the Spirit of God by which this message is received and
activated. It is not bound up with a succession and the apostles are
not those who are most active in promoting it. From time to time it is
said that those who receive the word (a characteristic Acts expres-
sion) are baptized into (or, in) the name of the Lord Jesus; it is quite
possible that when this is not mentioned it is nevertheless to be
understood, but even if this is so the omission is not unimportant.
The group of Christian brothers and sisters very naturally share from
 INTRODUCTION xcvii

time to time a fellowship meal. They may (Luke does not tell us so)
have recalled and repeated the words of Jesus: ‘This is my body, This
cup is the new covenant in my blood;’ there is no indication that the
breaking of the loaf and the uttering of these words were confined to
a special sort of person, though it is not surprising that when Paul
was present he spoke for a very long time (20.7, 11).
Haenchen (105) concludes ‘dass man mit der Charakterisierung
“Frühkatholizismus” vorsichtig sein muss’. The matter might be put
more strongly.

(h) The Jews. Acts begins with the converse of Jesus, a Jew, with
his Jewish disciples in the weeks immediately after his resurrection.
His ascension follows, and soon afterwards the gift of the Holy
Spirit. The disciples preach to a large and mixed company of people,
many of whom accept their message. They have come from many
lands but they have come to Jerusalem and most if not all of them
must be thought of as Jews. The Jewish atmosphere persists through
the next chapters, so that in ch. 8 Philip’s conversion of Samaritans
and of an Ethiopian is presented as a new step into a strange world.
So is Peter’s visit to Cornelius, and Cornelius’s baptism. From this
point the movement into Gentile world (see (j) below) gathers speed;
it is led by Paul and his colleagues, and at the end of the book Paul
quotes Isa. 6.9, 10 to the Roman Jews and adds, ‘Be it known to you
that this salvation of God has been sent to the Gentiles; they will
listen’ (28.26-28). A superficial reading of this story suggests that
the mission to Jews is now over; Christians are leaving them to their
own devices and concentrating on the mission to Gentiles. This
inference from Acts is not without plausibility, and the book has been
understood to be a substantial contribution to Christian anti-Semi-
tism.
This it certainly is not. A book whose main actors—Peter,
Stephen, Paul, James, not to mention Jesus (mainly off stage)—are
all Jews cannot easily be judged anti-Jewish (still less anti-Semitic—
Acts shows no interest in race as such). Most of the leaders of the
church have to be pushed very hard, by argument, by vision, by
divine intervention through the Holy Spirit, to accept uncircumcised
Gentiles as fellow believers. Surely, salvation is for Jews, to whom it
was promised. The verse quoted above (28.28) does not stand alone.
There are parallels in 13.46 (‘Since you thrust [the word of God]
from you and consider yourselves unworthy of eternal life, see! we
are turning to the Gentiles’) and 18.6 (‘Your blood be upon your own
head, I am clean; henceforth I shall go to the Gentiles’); and in each
case Paul’s first step in the next place he visits (14.1; 18.19) is to
enter the synagogue and pursue his mission there. There is no reason
why 28.28, though it stands at the end of the book, should be read in
a completely different way.
xcvüi COMMENTARY ON ACTS

This does not mean that there are no questions to be asked. It is

clear (and in this there is no reason to doubt the accuracy of Acts)
that large numbers of Jews were rejecting the belief that Jesus was
their, Jewish, Messiah. Stephen’s accusation is hard and bitter
(7.51-53). Does it mean that the Jewish people have forfeited their
special place in the purpose of God? that they have now been
replaced by the multi-racial company of believers in Jesus? The
question what role the Jewish people may continue to have in God’s
plan, and their ultimate hope of salvation, are discussed by Paul at
great depth in Romans 9-11. Luke was not capable of this kind of
discussion and was probably less aware of the questions than most
theologians are today. We may see a pointer, which resembles those
that we have seen in 14.1 and 18.19, by returning to the end of
Stephen’s speech. ‘You constantly resist the Holy Spirit; as your
fathers did so do you’ (7.53). Stephen has described several notable
failures on the part of earlier generations of Jews. His brothers sold
Joseph into Egypt; his people rejected Moses; they made and
worshipped the golden calf; they desired and built a temple, which
God did not desire. On each of these occasions God might well have
washed his hands of them, but he did not do so. True, Jesus was in a
sense his last word, but a misjudgment by the Sanhedrin on a
particular occasion could not be regarded as the last word of Israel’s
reply to God. In comparison with the Romans the Jews come badly
out of Luke’s account of Paul’s suffering, but his first act on reaching
Rome is to send for the leading Jews (28.17), and Luke (rightly or
wrongly) asserts that his message was nothing but what the prophets
and Moses said should happen (26.22), and this he was ready to
proclaim τω τε λαῷ καί τοῖς ἔθνεσιν (26.23).
Here however was the crux of the matter. Did Paul rightly
understand what Moses and the prophets had said? Most Jews said
no, and that on two fundamental points. According to Paul the
Scriptures prophesied the death and resurrection of the Messiah, and
this was fulfilled in Jesus. According to Luke they also foretold that
God would take out of the Gentiles a people for his name
(15.14-18). On this interpretation, Jesus was the Messiah and the
church, including uncircumcised Gentiles, was the people of God,
with the nation of Israel alongside in an undefined position, waiting
till it should return by faith into the divine purpose. This is perhaps
not a very satisfactory, or a very clear, position from the point of
view of theology, but it was a practical one and probably satisfied
Luke and most of his contemporaries. The same must probably be
said of the Apostolic Decree (15.29); on this see below and the notes
on 15.20, 29. The crucial issue was the Law; see the next section.

(i) The Law. ‘By three things is the world sustained: by the Law,
by the [Temple] service, and by deeds of loving-kindness’ (Aboth
 INTRODUCTION xcix

1.2 (Danby) ). No NT writer, certainly not Luke, raises any objection

to the doing of kindnesses. The attitude to the Temple
that is revealed in Acts is not simple.19 The first Christians are
represented as continuing to use the Temple. The scene of the
Pentecost event (2.1-4) is not specified; it may have been the
Temple; see I.113f. In Acts 3 we see Peter and John on their way to
the Temple at the time of prayer; they heal a lame man at the
Beautiful Gate, the party proceeds into the Temple, and there Peter
and John address the assembled crowd. In ch. 5 the apostles are
preaching in the Temple when they are arrested. In 6.13, 14 Stephen
is accused of threatening the Temple with destruction and in ch. 7 he
launches the most violent attack on the Temple found outside the OT.
From this point in Acts the scene begins to shift from Jerusalem, but
when Paul returns he accepts the challenge to demonstrate his
faithfulness to Judaism by taking part in Temple rites (21.23, 24, 26)
and thereby affirms that he accepted its discipline (21.24). This
double attitude to the Temple reflects the attitude of Acts to Judaism
in general.20 Judaism is good if it is understood in the Christian way.
The Temple is an aid to prayer; and participation in the resolution of
a Nazirite vow included sacrifice so that Paul’s visit to the Temple in
ch. 21 must be understood to be for the purpose of sacrifice as well as
prayer. But any attempt to confine God within a dwelling of human
construction is to be rejected, so that the very existence of a Temple
was a peril to true religion, whether in Jerusalem or in Athens
(17.24). This is a different attitude from that of the Qumran sect,
which did not disapprove of the Temple on principle but only of the
way in which it was being administered and of those who controlled
it. See I.338.
It is notable that Stephen, who attacks the Temple with such
vehemence, does not attack the Law. Moses received living oracles
to give to us (7.38). The Israelites received the Law εις διαταγὰς
αγγέλων (a thought that Paul uses in a different way—Gal. 3.19).
That they had a Law was a good thing; their fault was that they did
not observe it. It cannot be said on the basis of Acts that Stephen
initiated a law-free Gospel. There is indeed no law-free Gospel in
Acts but a compromise between those who wanted to keep the whole
Law in operation and those who as a condition of salvation wanted
no law at all. This compromise is expressed in the Decree of Acts
15.29. The Decree undoubtedly inclines markedly in the law-free
direction. Circumcision is not demanded, but Gentile converts are
required (ἐπάναγκες in 15.28) to abstain from food offered to idols,
from blood, from strangled meat, and from fornication—require-
ments that combine the moral with the ceremonial.21 This is
19See FS Bammel, 345-67
20See above section (h).
21See the Commentary on ch. 15.
c COMMENTARY ON ACTS

presumably the result of years of controversy between Paul and his

allies and various groups of Judaizers. It is essentially a practical
rather than a theological compromise, and though it is set forth as
containing conditions of salvation its main practical effect, in
addition to establishing peace within the church, was probably that it
made it possible for Jewish Christians and Gentile Christians to share
together in the church’s common meal. It was, however, based22 on
those elements of Judaism that a Jew could not give up even in the
extremity of persecution—Judaism reduced to an absolute minimum
so as to impose as little strain as possible on Gentiles.
Luke’s grounds for so far dispensing with the Law are given in ch.
15. The fundamental reasons are put, probably for reasons of policy,
in Peter’s mouth. First, the pragmatic reason: God called Peter to
speak the word to Gentiles; they received it by faith (not by works of
Law), and God gave to them the same gift of the Spirit that he had
given at the beginning to (Jewish) apostles (15.7f.). The next
emphasizes that God made no distinction between the circumcised
and the uncircumcised; the latter, unclean Gentiles that they were,
were cleansed by God by faith, without any legal device. Finally
Peter asserts that Jews themselves were unable to bear the yoke of
the Law and should therefore not seek to impose it on Gentiles—an
assertion that can hardly be maintained unless observing the Law is
understood to include the practice of perfect love to one’s neigh-
bours. Barnabas and Paul support the argument by recounting the
signs and portents that God had performed in the course of their
mission to the Gentiles; these surely he would not have done had he
disapproved of what was going on. Finally James points out that
Peter’s action had been in accordance with Scripture which manifests
God’s intention to find a people among the Gentiles. James’s further,
not too clear, point (15.21) may be intended to cut both ways: Moses
is so widely read that his Law cannot be simply ignored; but he has
enough people to preach him without our joining them.
This is very different from the Pauline argument, not least in the
fact that Christ is nowhere mentioned (unless, obscurely, in the
reference to the ‘tent of David’ in 15.16). For it is Christ who is τέλος
νόμου (Rom. 10.4). The precise meaning of these words is dis-
puted,23 but it is beyond dispute that the whole treatment of the
subject turns on the figure of Christ. This is perhaps the clearest
indication that Luke did not have a profound understanding of
Christian—and especially of Pauline—theology. He was a loyal
Christian believer but he did not see all the implications of his faith.
Nor was he able to bring out of the various sources that he used a
consistent view of the Law. Stephen is accused of speaking against
22See pp. 734f.
23See Cranfield, Romans 515-20; Romans 184; Dunn, Romans 589-91, with many
references.
 INTRODUCTION ci

the Law (κατά ... τού νόμου) and of saying that Jesus would change
(αλλάξει) the Mosaic customs (ἔθη; 6.13, 14), In his speech Stephen
speaks highly of the Law and alleges that his accusers break it. The
Cornelius episode is full of difficulties. In Mk (though not in Lk.)
Jesus declares all foods clean, but Peter has to be convinced by a
vision that he must not count common what God makes clean. This
vision itself seems to have to do with clean and unclean foods, but it
is interpreted with regard to human beings and the legitimacy of
dealings with Gentiles.24 Peter further recognizes that there is no
respect of persons with God (10.34). The next verse is difficult to
interpret precisely. On the surface it seems to mean that anyone who
‘does the right thing’ is accepted by God, and that this is as possible
for Gentiles as for Jews. If however this is true there seems to be little
need for the Gospel Peter is about to preach to Cornelius. If it means
only that a good man like Cornelius has as much right as a Jew to
have the Gospel preached to him it means that the ordinary man
needs a moral conversion before he can have a Christian conversion;
and indeed the conversion of Cornelius is not the change of life of
one who has previously practised evil but the gift of the Spirit which
issues in speaking with tongues (10.44-46). The conclusion reached
after discussion (11.18), however, is that Gentiles as such may now
be admitted to salvation, hitherto understood to be confined to Jews.
From this conclusion not only those whose opinion is quoted in 15.1,
5 but the Council and the Decree appear to retreat. There are certain
necessary conditions that Gentiles must fulfil. Later we learn that
there is a report that Paul is teaching Diaspora Jews not to circumcise
their children or to follow the (Jewish) customs—in a word, is
teaching apostasy from Moses (21.21); Paul adopts James’s sugges-
tion as the means of clearing himself from this charge. Paul himself
taught that circumcision must not be forced on Gentiles and that
among Jews it was an adiaphoron: Circumcision is nothing and
uncircumcision is nothing (Gal. 6.15). According to Acts he
observed Jewish feasts (20.16), though in Gal. 4.10 he speaks
disparagingly of them. The Roman opinion is that the difference
between Paul and other Jews is a matter of the interpretation of their
Law (18.15), and though Luke, where the Romans trivialize the
matter, sees its importance, he does not wholly disagree with Gallio.
As with the OT as a whole, the Christians have understood the Law
rightly, the Jews wrongly.
(j) Gentiles and the Gentile Mission. This matter has been almost
sufficiently dealt with in the sections on the Law and the Jews; also
in that on sources (I.50-52), for to a great extent Luke’s sources may
be regarded as those accounts of the origin and conduct of the
mission to the Gentiles that he was able to collect. The chronological
24See I.494, 516.
cii COMMENTARY ON ACTS

question who first took the Gospel beyond Judaism to the Gentile
world cannot be answered for Luke himself probably did not know.
In his account of the Council (15.7) he allows priority to Peter,
though he has described Philip’s work before Peter’s, probably
because he had said (8.1) that when all other Christians were
scattered from Jerusalem the apostles remained in the city. Mission
was his theme (see above, p. xxxiii); the mission to the Gentiles was
the greatest of missions, and not its originator but its greatest leader
was Paul. It is interesting that in Acts Paul does less than Peter and
James to justify a mission to the Gentile world, invoking only (in ch.
15) the miracles that happened in the course of his missioning. He
does not invoke the figure of Abraham as one who was justified by
faith, without circumcision, and received a promise that included all
the nations (Gen. 17.5; Gal. 4.17; Gen. 12.3; Gal. 3.9).
The earliest missions that included non-Jews were not (according
to Acts) the result of planning by those who undertook them. Philip’s
mission to Samaria was the result of the persecution that scattered all
the Christians (except the apostles) from Jerusalem. His encounter
with the Ethiopian was the consequence of direct instructions given
in the first instance (8.26) by an angel, subsequently (8.29) by the
Spirit. Peter’s visit to Cornelius was the result of co-ordinated
directions given to the two men; it is made clear in his response to the
vision (10.14) that Peter was not disposed to have dealings with
Cornelius and that his resistance had to be broken down by divine
pressure. When those who like Philip were driven from Jerusalem
arrived in Antioch it is clear that their first intention was to preach
only to Jews; at a second stage unnamed men from Cyprus and
Cyrene included non-Jews in their scope. In view of this, and the
resulting mixed church in Antioch, it may be (though it is not stated)
that is was intended from the beginning that the mission to which
Barnabas and Saul were committed (13.1—3) should include Gen-
tiles. 13.5, however, refers only to preaching in the synagogues of
Cyprus, and the meeting (13.6-12) with the proconsul Sergius
Paulus seems to have been unpremeditated. It is a possible but quite
uncertain conjecture that the success of this encounter (13.12)
stimulated concern for the Gentiles. It is a further possible but quite
uncertain guess that John Mark left his senior colleagues (13.13)
because he disapproved of this unplanned step. From this point
(according to Acts) Paul’s own purpose was constant, and he usually
adopted the method, on reaching any new town, of first visiting the
synagogue, making use of hearers already collected to hear religious
discourse, and leaving it when, as regularly happened, the Jews
rejected his message, in order to concentrate on Gentiles. He never
ceased to be concerned for his fellow Jews (e.g. 28.23; also Rom.
9.1-3; 10.1), but recognized in the Gentile mission a special
vocation. This indeed is in Acts traced back to his conversion (9.15;
 INTRODUCTION ciii

22.21; 26.17); rightly so, for this is confirmed in Gal. 1.16 (... ἵνα
εὐαγγελίζωμαι αὐτόν ἐν τοίς ἔθνεσιν). The first journey (to use
Luke’s division of the material) includes a small circuit in Pamphylia
and Pisidia; after this Paul moves more widely.
It is very probable that Jesus gave no explicit command to
undertake a mission to the Gentiles. J. Jeremias25 is probably right in
the view that he foresaw after his death the eschatological pilgrimage
of the nations to Jerusalem. Their joining with the Jews in one people
of God would be one aspect of the end of history. After the
resurrection the disciples found that there was to be an unexpected
tract of history before the end. This extension of time made a mission
to the Gentiles possible (see above, pp. lxxxiif.); but the mere
extension of time was not sufficient to cause the Gentile mission. For
causes we must look further. To some extent it may have been due to
what might look like chance. A Gentile heard what the missionary
was saying, accepted it, and manifested the spiritual and moral signs
of a changed life. The preacher was faced with the question uttered
by Peter at 10.47: How can I refuse to baptize one who shows the
same marks of Christian existence as I do myself? He must be
welcomed into the saved community. There were also hints in the
story of Jesus. He undertook no mission outside Israel, but he did
devote himself to and gave his life for those who though of Jewish
race had wandered outside the religious framework of their people. If
he could eat with tax collectors and sinners it is in fact surprising that
Peter should hesitate to eat with Cornelius, and that, if Jesus declared
all foods clean,26 Peter should at first refuse to obey the order, ‘Kill
and eat’ (10.14, 28; 11.3). No doubt sporadic conversions took place,
but it was Pauline theological development that established work
among the Gentiles, especially the recognition that the significant
ancestor of Jesus was not David, the king of Israel, but Adam, the
father of the race, and that God was not the God of the Jews only but
of the Gentiles also (Rom. 3.29)—a God of half humanity would be
only half a God.
This theological development is not found in Acts. Paul is called to
go to the Gentiles, and he goes. There were other missions to the
Gentiles. There were those of whom we learn most from the Pauline
epistles but of whom we can see something in Acts: those who
declared roundly that Gentiles might be accepted but only if they
accepted circumcision and the Law (Acts 15.1, 5); perhaps there
were also those whose existence is admitted though their authoriza-
tion is denied at 15.24. Different groups insisted on various elements
in Judaism: some upon all, the Galatian Judaizers on circumcision
and the calendar, the Corinthian Judaizers on food laws, those who

25Jesus' Promise to the Nations (ET 1958), especially 55-73.

26Mk 7.19 (not in Lk.).
civ COMMENTARY ON ACTS

were responsible for the Decree on food laws, avoidance of idolatry,

and chastity. The last may possibly be connected with the Seven of
Acts 6.5. These or other Diaspora Jews may have been responsible
for Stephen’s speech (which opposes the Temple but accepts the
Law) and the Areopagus speech attributed to Paul. Both of these
speeches suggest an origin27 in Hellenistic Judaism. It would be
natural—and by no means improper—for a Hellenistic Jew who had
become a Christian to edit and reuse a synagogue sermon in which he
had combined Greek philosophy with OT religion, introducing a
reference to Jesus at the end.
At this point should be mentioned also the so-called ‘God-fearers’.
For their existence, the nomenclature applied to them, and their
possible role in Acts, see 1.499-501. Here it suffices to state without
discussion that there were Gentiles who found Jewish ethics, theol-
ogy, and worship attractive, but not to the extent of becoming
proselytes. Some of them probably had some contact with the local
synagogue. Christian preachers offered to them, as they offered to
all, a form of Judaism, stripped of its least attractive features, in
particular of the rite of circumcision. If there were no evidence at all,
it would seem probable that some of these should adopt the Christian
way of admission to the people of God, and thus come to form the
nucleus of a non-Jewish, uncircumcised element in the newly formed
church.
(k) Ethics. Acts contains hardly any direct ethical instruction. The
mission speeches include the call to repent, and this implies a change
in moral behaviour; note especially 26.20, with the demand that
hearers should show άξια τής μετανοιας έργα, works, moral acts,
that will demonstrate the sincerity of the repentance they profess.
Even here however (cf. 2.38; 3.19; 17.30) there is no attempt to
specify the works that might have this effect. The only example of
specific ethical instruction is in 20.33-35. where Paul, addressing the
Ephesian elders (it is of course Christians whom one would expect to
receive ethical teaching), speaks of the example he has given and
urges them not to depend on charity but to work as he has done so as
to help the weak. He invokes a saying attributed (somewhat improb-
ably—see the note) to Jesus: It is more blessed to give than to
receive. Cf. 1.21, 22: the twelfth apostle who is to take the place of
Judas Iscariot must have accompanied Jesus throughout his ministry;
this implies familiarity with the teaching, including the ethical
teaching, of Jesus.
It is consistent with this that the Acts narrative is written on a good
ethical level. Clearly the healing of the sick is regarded as a proper
activity in which Christians, if they have the appropriate gift, ought

27See I.338f.; pp. 331, 344-371; also 14.15-17 (1.665, 680-3).

 INTRODUCTION cv

to engage (3.1-10; 4.9—the healing is a ευεργεσία; 4.30; 5.12-16;

8.6-8; 9.33., 40f.—raising the dead is a particularly notable act;
14.10; 19.11f.; 20.10-12; 28.8, 9), though compassion for the sick is
not the only motive—miracles have evidential value. Violent attacks
on Christians are clearly not approved, but no special sympathy is
shown for Sosthenes (18.17), and the story of the sons of Sceva
(19.14-16) is told with a measure of Schadenfreude. Lying and
deceit are wrong (5.1 — 11); so are disloyalty (15.38) and cruelty (in
an OT story, e.g. 7.19). Herod Agrippa however is struck down not
because he beheaded James and imprisoned Peter but for the
theological offence of accepting glory due to God alone (12.23).
Courage and determination, in facing death (e.g. 21.13) or in a storm
at sea (ch. 27), are virtues. Fornication and the shedding of blood (if
this is what αἷμα means—see the note) are forbidden in the Decree
(15.20, 29; 21.25). The practice of magic is condemned (8.9; 13.6;
19.13-19).
The teaching of the apostles (2.42), we may suppose, would
include ethical teaching though, in a Jewish community, there would
be little need for it: high ethical principles were already taught and on
the whole practised. Paul’s teaching to Felix (24.25) was perhaps not
unnecessary. If the Western text is followed the Apostolic Decree
included the negative form of the Golden Rule, which certainly was
familiar in Judaism. The general requirement of kindness and care
for those in need is expressed in a number of ways. For the selling of
property and sharing of resources (2.44f.; 4.32, 34, 35, 36, 37;
5.1-11; cf. 6.1) see further below. The church of Antioch sent to the
needs of their fellow Christians in Jerusalem (11.29,30; 12.25). Paul,
on his last visit to Jerusalem, had come with alms (24.17). Almsgiv-
ing was among the virtues of Cornelius (10.2, 4, 31) and Dorcas was
full of good works and charities (9.36, 39). Hospitality is a notable
good work (16.15, 34; 21.8, 16). Paul took thought for his fellow
travellers who had spent too long without food (27.33-38). Ste-
phen’s prayer for the pardon of those who stoned him is modelled on
that of Jesus (7.60).
It is interesting to observe that virtue and charity are not confined
in Luke’s narrative to Christians. Cornelius’s practice of charity and
righteousness has already been noticed (10.2, 4, 31, 35). Paul’s
friends among the Asiarchs are not said to be Christians but they took
thought for him and tried to protect him (19.31). Julius the centurion
took steps to save Paul (27.43—as other Roman officials had done),
and the barbarians of Malta ‘showed us no common kindness’
(28.2).
Paul twice claims to keep, or to have done his best to keep, a good
conscience (23.1; 24.16). If Paul himself uttered these words he
probably added—aloud or under his breath—the words he uses in 1
Cor. 4.4, άλλ’ οὐκ ἐν τούτῳ δεδικαίωμαι. For Luke they mean quite
cvi COMMENTARY ON ACTS

simply that Paul regularly tries to do what he believes to be the right

thing.28
At 21.9 Philip is said to have four daughters who were virgins and
prophesied. One can only wonder what, if anything, lies behind this.
Did their virginity permit them to prophesy? Was it a qualification
for mention in the book? Or was Luke simply stating a set of facts:
there were four of them; they had remained unmarried; they uttered
prophecies?
The special interest in poverty and wealth, in the danger of riches
and the importance of the care of the poor, which are frequently
noted as characteristic of the Third Gospel, are present but less
emphasized in Acts. They most often take the form of the organizing
of charity for those in need. Paul’s collection (Rom. 15.25-28, and
elsewhere), which may be alluded to at Acts 24.17, is not mentioned
in Acts 15 (but cf. Gal. 2.10). It is possible that the Antiochene
collection (see above) refers to the same gift. Most relevant here, but
discussed in 1.167-70, 251-60, 310, 312f., are the sale of properties
mentioned in 2.44f.; 4.32-37; 5.1-11 and the διακονία of 6.1. We
should note also the use of the word κοινωνία and the statement that
the Christians had άπαντα κοινά (2.44; 4.32). The ministry of 6.1 is
comparable with familiar Jewish charities; the common possession
of all goods calls to mind Greek proverbs, which no doubt were
sometimes expressed in concrete arrangements, but within Judaism it
finds a parallel only in the practices of the Qumran sect; see I.167-9.
It is not surprising that arrangements of this kind, natural in minority
groups, should be found both at Qumran and among the Christians; it
does not prove any close relation between them.

Bibliography
General
F. Bovon, RScR 69 (1981), 279-300.
F. Bovon, Luke the Theologian. Thirty three years of research, ET Allison
Park 1987.
H. von Campenhausen, ZNW 63 (1972), 210-53.
C. H. Cosgrove, NovT 26 (1984), 168-90.
E. Franklin, Christ the Lord, London 1975.
B. R. Gaventa, Interpretation 42 (1988), 146-57.
J. L. Houlden, JSNT21 (1984), 53-65.
W. G. Kümmel, EThL 46 (1970), 265-281 = ZNW 63 (1972), 149-65.
B. F. Meyer, FS Farmer, 243-63.
J. C. O’Neill, The Theology of Acts, London 21970.

28For moral ἄσκησις in Deutero-Pauline literature see FS Furnish 161-72.

 INTRODUCTION cvii

J. Panagopoulos, NovT 14 (1972), 137-59.

C. H. Talbert (ed.), Luke and the Gnostics, Nashville and New York 1966.
U. Wilckens, Die Missionsreden der Apostelgeschichte, WMANT 5, Neu-
kirchen-Vluyn 1963.

L Eschatology
F. F. Bruce, FS Beasley-Murray, 51-63.
J. T. Carroll, Response to the End of History, SBLDiss 92, Atlanta 1988.
E. Grösser, Kremer, Actes 99-127.
K. Haacker, NTS 31 (1985), 437-51.
R. H. Hiers, NTS 20 (1974), 145-55.
J. D. Kaestli, L'eschatologie dans l'oeuvre de Luc, Geneva 1969.
A. J. Mattili, CBQ 34 (1972), 276-93.
D. P. Moessner, NTS 34 (1988), 96-104.
G. Schneider, Lukas, Theologe der Heilsgeschichte, BBB 59, Bonn 1985.
S. G. Wilson, Gentiles 67-80.

2. The Holy Spirit

A. George, RB 85 (1978), 500-42.
R. P. Menzies, JSNT 49 (1993), 11-20.
Μ. Turner, NTS 38 (1992), 66-88.

3. Christology
G. Delling, NTS 19 (1973), 373-89.
L. Hartman, see Baptism.
E. Kränkl, Jesus der Knecht Gottes, Biblische Untersuchungen 8, Regens-
burg 1972.
D. P. Moessner, NovT 28 (1986), 220-56.
R. F. O’Toole, Bib 62 (1981), 471-98.
E. Schweizer, FS Schubert 186-93.
P. Vielhauer, FS Schubert 33-50 = EvTh 10 (1950-1), 1-15.
G. Voss, Die Christologie der lukanischen Schriften in Grundzügen, Paris
and Brügge 1965.
J. A. Ziesler, JSNT4 (1979), 28-41.

4. The Church
R. E. Brown, Proceedings of the Catholic Theological Society ofAmerica 36
(1981), 1-14.
J. Pathrapankal, Zeitschrift für Missionswissenschaft und Religionswis-
senschaft 70 (1986), 275-87.
B. Reicke, Glaube und Lehre der Urgemeinde, AbThANT 32, Zürich 1957.
J. B. Tyson, Interpretation 42 (1988), 132-45.

5. Apostles and Ministers

F. H. Agnew, JBL 105 (1986), 75-96.
cviii COMMENTARY ON ACTS

E. Bammel, Augustinianum 30 (1990), 63-72.

C. K. Barrett, The Signs of an Apostle, London 1970; Carlisle 1997.
C. Burchard, Der dreizehnte Zeuge, Göttingen 1970.
A. Campbell, JTS 44 (1993), 511-28.
R. A. Campbell, The Elders: Seniority Within Earliest Christianity, Edin-
burgh 1994.
H. von Campenhausen, Kirchliches Amt und geistliche Vollmacht in den
ersten drei Jahrhunderten, Tübingen 1953.
L. Cerfaux, Rec. 2.157-74.
E. Ferguson, JTS 26 (1975), 1-12.
B. Gerhardsson, SEÅ 27 (1962), 89-131.
K. Haacker, NovT 30 (1988), 9-38.
A. E. Harvey, JTS 25 (1974), 318-32.
Μ. Karrer, NovT 32 (1990), 152-88.
G. Klein, Die zwölf Aposteln, FRLANT 59, Göttingen 1961.
D. Powell, JTS 26 (1975), 290-328.
K. H. Rengstorf, Apostolat und Predigtamt, Stuttgart 1934 (1954).
K. H. Rengstorf, TWNT 1.397-448.
C. H. Roberts, JTS 26 (1975), 403-5.
W. Schmithals, Das kirchliche Apostelamt, FRLANT 61, Göttingen 1961.
H. Schürmann, FS Erfurt (1977), 1-44.
B. E. Thiering, JBL 100 (1981), 59-74.
F. Μ. Young, JTS 45 (1994), 142-8.
6. Baptism and the Christian Meal
G. Barth, ZThK 70 (1973), 137-61.
S. Brown, AThR 59 (1977), 135-51.
E. J. Christiansen, StTh 40 (1986). 55-79.
N. A. Dahl, FS Mowinckel 36-52.
L. Hartman, StTh 28 (1979), 21-48.
T. W. Manson, JTS 48 (1947), 25-33.
P. H. Menoud, RevHPhR 33 (1953), 21-36.
Μ. Quesnel, Baptisés dans l'Esprit, Lectio Divina 120, Paris 1985.
B. E. Thiering, NTS 27 (1981), 615-31.
J. Thomas, Le Mouvement baptiste en Palestine et Syrie, Gembloux 1935.

7. Frühkatholizismus
E. Bammel, Hervormde teologie Studies 11 (1993), 690-707.
R. Bultmann, Theologie 446-70; 446, 464, 621-3, 718-25 for extensive
bibliography.
H. von Campenhausen, Kirchliches Amt und geistliche Vollmacht in den
ersten drei Jahrhunderten, BHTh, Tübingen 1953.
H. Conzelmann, Theologie 58-77, 333-7.
A. Harnack, Entstehung und Entwicklung der Kirchenverfassung und des
 INTRODUCTION cix

Kirchenrechtes in den zwei ersten Jahrhunderten, Leipzig 1910; ET The

Constitution and Law of Church in the first two Centuries, London
1910.
E. Käsemann, EVB 1. 109-34 (especially 130-3); cf. 135-57; 2. 239-52,
262-67.
E. Käsemann, in Das Neue Testament als Kanon (ed. E. Käsemann),
Göttingen 1970, 371-78, 399-410.
H. Küng, in Das NT als Kanon (see above), 175-204.
J. B. Tyson, Interpretation 42 (1988), 132-45.
8. The Jews
R. L. Brawley, Luke-Acts and the Jews, SBLMono 33, Atlanta 1987.
V. Fusco, NovT 38 (1996), 1-17.
A. George, RB 75 (1968), 481-525.
K. Haacker, NTS 31 (1985), 437-51.
W. Horbury, JTS 33 (1982), 19-61.
L. T. Johnson, JBL 108 (1989), 419-41.
G. Lohfink, Die Sammlung Israels, StANT 33, Munich 1975.
H. Merkel, NTS 40 (1994), 371-98.
Μ. Rese, FS Schneider, 61-79.
J. T. Sanders, The Jews in Luke-Acts, London 1987.
R. C. Tannehill, JBL 104 (1985), 69-85.
J. B. Tyson, NTS 41 (1995), 19-38.
A. Vanhoye, Bib 72 (1991), 70-89.
L. Μ. Wills, JBL 110 (1991), 631-54.
9. The Law
C. K. Barrett, FS Bammel 345-67.
C. L. Blomberg, JSNT 22 (1984), 53-80.
D. R. Catchpole, NTS 23 (1977), 428-44.
F. G. Downing, JSNT 26 (1986), 49-52.
N. J. McEleney, NTS 20 (1974), 319-41.
Μ. A. Seifrid, JSNT 30 (1987), 39-57.
A. J. Μ. Wedderburn, NovT 35 (1993), 362-89.
10. The Gentiles and the Gentile Mission
J. Μ. G. Barclay, JSNT 60 (1995), 89-120.
E. E. Ellis, StEv IV (= TU 102), 390-99.
P. Frederikson, JTS 42 (1991), 532-64.
A. T. Kraabel, Numen 28 (1981), 113-26.
A. T. Kraabel, FS Stendahl, 147-57.
J. Μ. Lieu, JTS 46 (1995), 483-501.
J. Μ. Lieu, FS Goulder, 329-45.
B. F. Meyer, see General section.
J. Murphy-O’Connor, RB 99 (1992), 418-24.
J. A. Overman, JSNT 32 (1988), 17-26.
cx COMMENTARY ON ACTS

J. Reynolds and R. Tannenbaum, Jews and Godfearers at Aphrodisias,

Cambridge Philological Society Supplementary Volume 12, Cambridge
1987.
E. Richard, SBL Seminar Papers 1980, 267-82.
J. T. Sanders, NTS 37 (1991), 434-55.
C. H. Talbert, FS Schneider 111-26.
J. B. Tyson, NTS 33 (1987), 619-31.
Μ. Wilcox, JSNT 13 (1981), 102-22.
S. G. Wilson, Gentiles.

11. Ethics
C. K. Barrett, FS Furnish 161-72.
P. Borgen, FS Kee 126-41.
R. J. Cassidy, Society and Politics in the Acts of the Apostles, Maryknoll,
New York 11987 21988.
F. G. Downing, NTS 27 (1981), 544-63.
L. T. Johnson, The Literary Function of Possessions in Luke-Acts, SBLDiss
39, Missoula 1977.
A. J. Malherbe, Social Aspects of Early Christianity, Philadelphia21983.
D. L. Mealand, ThZ 31 (1975), 129-39.
D. L. Mealand, JTS 28 (1977), 96-9
 INTRODUCTION cxi

CONCLUSION

In the preface to his gospel (Lk. 1.1-4) Luke claims to have

associated with persons1 who may or may not have been able to
supply him with accurate historical information about the life and
teaching of Jesus but must have been involved in some way in the
life of the early church. They are described as eye-witnesses and
ministers of the word (αὐτόπται καὶ ὑπηρέται τοῦ λόγου). Such
contacts formed the basis of his affirmation, which was evidently
important to him, of the continuity between the pre-crucifixion Jesus
and the post-resurrection church. They will have been sources for
Acts as well as (in a different way) sources for the gospel. They must
also have been sources for Luke’s own theological and religious
thinking. What is to be made of the author and his book? In the
following pages, which will necessitate a small amount of repetition
of matters already dealt with, this question will be considered.
It will not be wrong to begin with the observation that Luke was a
man who liked telling stories and was good at telling them. He was
less good at the connections between the episodes he narrated;
perhaps he was less interested in them. He had before him a similarly
episodic model, Mark. For this he had no little respect; otherwise he
would not have used so much of it. But it was capable of improve-
ment, and he set about improving it. He improved the Greek style,
shortened passages that were unnecessarily long, and used his
economies in space to add a good deal of fresh material. His
revisions and additions had the effect of producing a story less
starkly theological, more ‘human’ in interest and feeling. When he
had absorbed Mark that work was finished with; it was superseded
and could be dispensed with. Fortunately there were Christians who
did not agree, and retained it.
One can hardly suppose that pleasure in telling stories would
suffice to produce a written work of some size. Interest in story-
telling leads to an interest in history; and there were additional
reasons for such an interest. Christianity was a religion, an institu-
tion, a system of thought—none of these terms is satisfactory, but
they may suffice—that could maintain its identity only by recalling
its origin, for when it was truly itself it was determined by its origin.
Luke wrote at a time when the old sense of an imminent consumma-
tion of history had waned. In the early days there had been no need to

1If πᾶσιν is taken as masculine παρηκολουθηκότι will imply so much.

cxii COMMENTARY ON ACTS

remember because the future was short and the end immediate.
Memories however were now in danger of fading. Someone had to
take steps to secure the church’s memory not only of Jesus but of the
way in which the transition from Jesus to the church had been
effected. Mark, and any other gospel there may have been, had not
recorded this; Paul, whose letters (which Luke seems not to have
known) contained valuable pieces of history, had not recorded it.
Luke may not have seen the need as clearly as it has been stated here,
but he, and so far as we know no one else, did something to meet it.
In addition to the danger of forgetting there was the danger that
moral and doctrinal standards might slip. Most societies tend to think
of their origins as heroic days, in which members of the society stood
firm in faith and morals and stood by one another. A picture of the
past, perhaps an idealized past, will inspire and instruct the present.
Luke’s picture of first generation Christians was intended to do this
for his generation. He was, moreover, aware of the dangers that he
puts on Paul’s lips in 20.29, 30. What was future to Paul in Miletus
was still in the main future to Luke, but it was a good deal nearer and
no doubt had already appeared on the horizon. There were those who
spoke perverse things and drew away disciples into their own
schismatic coteries. Let the warning stand and an appropriate
example be provided. Example, both in preaching and in morals, is
there in Acts, but it has little positive content. There is little ethical
teaching in Acts, though it is clear that the Christians are expected to
be, and on the whole are, ‘good’ people. And Luke has no theological
doctrines that he wishes to commend beyond basic Christian convic-
tion. He believes in God, conceived on the lines of the OT, though he
is aware of a convergence between the OT and the best Greek
thought. Jesus Christ is central in his thought, but there is no attempt
to think through the problems of the incarnation. He is Son of God
(9.20), as all Christians knew, but the term itself carries with it no
implication of ‘being of one substance with the Father’; he is Lord
and Christ, but there is at least one hint that he became Lord and
Christ only at his exaltation (2.36); his death was the result of sin and
ignorance, speedily put right by God (3.13-18), but only at 20.28 is
there a suggestion that it was by his death that he redeemed mankind.
Luke undoubtedly believed in the Holy Spirit, and at 5.3f. there is a
hint that he is divine, but it is the phenomenology rather than the
personality of the Spirit that interests Luke. He accepted the OT as
the word of God, and to this extent stood with the Jewish people, but
he would have taken the Christian line that the OT must be
interpreted in terms of Christ, not Christ in terms of the OT.
Luke does not argue a special theological position but, it seems,
takes the line of the majority of Christians in the 80s and 90s of the
first century. This however is a proposition that will call for careful
examination. It is by no means clear what group of Christians can be
 INTRODUCTION cxiii

reckoned as constituting a majority in the 80s. In the view of J.

Jervell Jewish Christians were numerically a minority but exercised
an influence out of proportion to their numbers; it was for example
their concern for the theological legitimacy of the Gentile mission
that is reflected in Acts. This is in many respects an important
observation, but it calls for some modification.23That the earlier
decades of the century were marked by conflict, often bitter and
unrelenting, appears without question from the Pauline letters. The
church had now emerged into a period of relative calm; this had
happened earlier than Baur (see pp. lxxiiif.) thought, and not quite in
the way that Goulder (see p. xli, n. 18; p. lxiii, n. 1) maintains. It was
not Paul who won, though it may go too far to say that he lost.3 Paul
and James, the extremists of right and left, were both defeated by the
centre party, whom we may call if we wish the Hellenists. It is their
compromise that appears in the Apostolic Decree. This Decree is
clearly accepted by Luke as the basis of the Gentile mission and the
guide line on which the mission was to be conducted. It also provided
its theology, which appears in a different form in the Areopagus
speech (17.22-31)—the doctrine of God on which Jews and pagan
theists could unite, with a reference to Jesus attached to it. This is the
theology that prevailed in this period of the church’s life, taking
inevitably variant forms in different places. It originated with
Hellenistic Jews and was accepted by very many Gentiles. The
reference to Jesus is minimal in Acts 7.52 and 17.31, and Luke
himself, as is clear from many passages, when writing on his own
would greatly increase it; so did e.g. 1 and 2 Clement and Ignatius;
hardly Hermas. The theological substructure lasted till it was
attacked by Marcion, who failed in the end to establish his exag-
gerated reaction. Acts made a substantial contribution to his defeat,
and when this was grasped the book emerged from the obscurity into
which it had fallen, and served the purpose of showing that the
Creator God and the Redeemer God were one and the same, and that
Paul was in agreement with the Twelve. Irenaeus and Tertullian were
not wrong in seeing that Luke made these points, though making
them was not his main intention.
Historically, then, Luke is right when he celebrates a Hellenist
victory; Hellenists (if we may use the word) were a centre party and
their Decree guided the church through an obscure period and
continued to direct it for much longer. Luke was wrong in represent-
ing Paul as one of the victorious Hellenists; he describes himself (2
Cor. 11.22; Phil. 3.5) as a Hebrew. This is the central (though by no
means the only) point on which any judgement of Luke as a historian
and a theologian must be based.

2See my ‘What Minorities?’ in FS Jervell 1-10.

3See FS Jervell 6-9.
cxiv COMMENTARY ON ACTS

There are many features of Acts that must win a favourable verdict
on the author as a historian. There are fairly frequent references to
contemporary events and institutions and they are on the whole
satisfactory. Many examples are given in the commentary; here it
may be sufficient to mention the two kings, Agrippa I and Agrippa II
(Luke describes the death of the former in substantial agreement with
Josephus and knows that the latter’s sister was Bernice); the two
Roman governors, Felix and Festus (there is behind 24.27 a problem
in the date of accession, but it is non-biblical evidence that lacks
conclusiveness); the two proconsuls, Sergius Paulus and Gallio (the
latter’s governorship in Achaea can be dated with some precision and
there is probable epigraphic evidence for the former); the Emperor
Claudius (there is evidence for food shortages during his reign and
Luke cannot be blamed for some uncertainty with regard to the date
of his expulsion of the Jews from Rome). To persons may be added
places and institutions associated with them: Luke knows Philippi
was a colonia and that its magistrates were known as στρατηγοί; he
knows of the Areopagus court in Athens and of the world-famous
goddess of Ephesus, Artemis; of the connection of Ephesus with
magic, of its Asiarchs, and of its γραμματεύς; he knows that one
might find a centurion of the Cohors Italica in Caesarea; that Appii
Forum and Tres Tabernae, not in themselves important towns, were
stages on the road to Rome; that to lay hands on a Roman citizen was
a dangerous thing to do, but that to resist or oppose δόγματα
Καίσαρος was equally dangerous, though a citizen might escape at
least immediate punishment by an appeal to Caesar. Luke is less
successful in his references to Jews and to Jewish affairs. His
reference to Annas may perhaps with some difficulty be defended;
Gamaliel is rightly named but Luke puts on his lips a historical
howler in the mention of Judas and Theudas; it is difficult to
accommodate the details of the events that led to the riot in the
Temple with the regulations for vows, though Luke is right in
representing the introduction of Gentiles into forbidden areas of the
Temple as a very serious and provocative offence.
Luke, then, was in general well informed about persons, events,
and institutions in the Graeco-Roman world of the first century,
probably better informed than most of his contemporaries, than most
of his readers. And he was no fool. Where he agrees with other
historical sources, his evidence is confirmed; where he disagrees, or
where other evidence is lacking, he must at least be taken seriously.
These matters however are incidental to his purpose, for he was not
writing political, social, economic, military, or institutional history,
but Christian history, and it is in this field that he must be judged.
Little can be said about the first twelve chapters of Acts except that
in outline they must represent, with many omissions, the sort of thing
that must have happened in the early years of Christianity. This is not
 INTRODUCTION cxv

the place to discuss the historicity of the resurrection; 1 Cor. 15.4-8

is sufficient to prove that the earliest (pre-Pauline) Christians
believed that the crucified Jesus had appeared alive to a number of
their leaders. This belief is represented in Acts. The rest of the first
five chapters is made up of some traditional narratives and of Lucan
constructions which are not without historical foundation. Twelve
was a significant number for the inner group of disciples; the
Christians did believe themselves to be inspired by the Spirit of God;
they must have made speeches to communicate their beliefs; there
was a traditional group of miracle stories; there was trouble between
the Christians and the Jewish authorities. With ch. 6 a new stage of
historiography begins. The sources used are discussed at I.54f. Into
Antiochene traditions Luke incorporates more Petrine stories, Jerusa-
lem traditions, stories going back to Philip, and the tradition about
Paul’s earliest contacts with Christianity. This is serious historical
material; what it lacks is continuity and chronological coherence.
Luke has set down different accounts of how the mission to the
Gentiles began. Only one of these, it seems, has a future. This is
given at first in the Antioch-based narrative of chs. 13 and 14. It
continues in the Pauline missions, of which traditions, some drawn
from members of the Pauline circle, some collected locally in Pauline
cities, are set out first in a sequence of missionary journeys. These
journeys of Paul as a missionary free to travel wherever he was led
by the Spirit end in Jerusalem; from ch. 22 onwards Paul is not a free
man, and the story of his encounters with Jewish and Roman
authorities is very difficult to evaluate historically. Paul was not held
incommunicado. Information about the various legal—and illegal—
proceedings could have been got out; but in the narrative itself there
is no indication that this happened, no mention of a friend who kept
in touch with him except at 24.23. Luke makes nothing of this, and in
any case it does not cover the events of chs. 25 and 26. From 27.2
Paul is again accompanied by one or more companions and it is
possible to guess with some plausibility at passages inserted in
traditions derived from them.
So far nothing has been said about ch. 15, and it is here that
questionings arise that cannot but affect our judgement of the rest of
Acts. The composition of the chapter is discussed in some detail
below (pp. 696f., 710f.). It is Luke’s work, though it is not without
historical foundation. It contains however difficulties in itself, and
more appear when it is compared with the direct Pauline evidence,
especially that of Galatians. Did Paul help to produce, assent to, and
disseminate a Decree to which, even when he deals with its subject-
matter, he makes no reference? a Decree which tells his Gentile
Christians of Jewish conditions (not indeed including circumcision)
which they must fulfil if they are to be members of the people of
God? The only possible answer is No.
cxvi COMMENTARY ON ACTS

At this point the question about Luke the historian runs into the
question about Luke the theologian. For that Luke deliberately
intended to calumniate Paul, ascribing to him views Luke knew that
he did not hold, is inconceivable; he admired him far too greatly for
this. The only alternative is that he did not truly understand him. That
‘a man is justified by faith and not by works of the Law’ (Rom. 3.28)
is a proposition not covered by the one reference to justification in
Acts (13.38, 39). The explanation of the misunderstanding is latent in
the sketch of the last few pages. As Luke himself shows, the mission
to the Gentile world did not have a single official beginning. It started
independently in different places, through different persons, and on
different lines. Among those who began it, in their own way, were
Hellenistic Jews who had become Christians. They were the founders
of the mixed church in Antioch; their way of theological thinking and
of preaching is seen in Acts 7 and 17. Their mission continued and
prospered; it became (see above, pp. xlff.) the main official line. But
the line of development that Luke knew, admired, and described was
Paul’s; and he confused the two. Paul was connected with Stephen
and Antioch (7.58; 8.1; 11.25f.; 12.25; 13.1-3; 14.26-28; 15.2;
22.20); he must have been Stephen’s successor and continuator—
this was the volte-face of his conversion. He began as an Antioch
envoy; he must have continued to be one. So the Hellenistic Jewish
compromise Decree was Paul’s decree; and Paul not only lost the
battle for a radically law-free Gospel, he lost his integrity at the same
time: a strange fate for Luke’s hero to suffer at Luke’s hands.
It was the only point of serious difference between them. Else-
where it is enough to say that Luke lacks Paul’s profundity. Luke
believes in Christ, crucified and risen; but does not think about pre-
existence or, in any profound sense, deity, and on the whole (except
at 20.28) thinks of the cross as an unfortunate error put right by
resurrection, which happily sets the cross aside. He believes in the
Spirit, who causes ecstatic speech, rather than love, joy, and peace.
He believes in the church, but scarcely sees it as the body of Christ.
He knows baptism and a Christian meal, but they are not focused on
crucifixion, on dying with Christ and proclaiming the Lord’s death
till he come. All this does not mean that Luke disagreed with Paul;
only that he was not so good at theology. And this takes us back to
history; how close to Paul and his ministry did Luke stand?
Theology however must have another word. Is Luke’s theology
true? Or have we disfranchised a book from the NT?
It must not be assumed that omission, especially in Christological
matters, implies contradiction; and Luke either tolerates, or perhaps
fails to observe, a measure of contradiction. A superficial reader of
Acts 2.36 will infer that Luke supposed that Jesus became Messiah
and Lord at his resurrection. Luke’s source may have meant this, and
Luke may have accepted it as he transcribed the source. But Lk. 2.11
 INTRODUCTION cxvii

states with equal clarity that Jesus was born as Messiah and Lord.
The reformulation in Acts reflects the impression made by the
resurrection on those who experienced it.4 No Christology is com-
plete without the notion of pre-existence, but Christology has often
lacked it. Luke’s Christology lacks it, but this does not mean that
Luke would have denied it, or needed to deny it. The same
observation can be made with reference to the doctrine of atonement
through the death of Christ. That Luke does not assert it (except at
20.28) does not mean that he would have denied it, still less than his
readers need deny it.
Comparison of Acts 17.22-31 with Rom. 1.18-25 raises sharply
the question of the place in Paul’s thought, and thus in Christian
thought generally, of natural theology.5 The speech must be read in
the context that Luke has provided for it and as determined less by
Stoic speculation than by the OT prophetic denunciation of idolatry.
This denunciation Paul shared (so one learns—credibly—from
Luke’s own story); he was provoked (παρωξύνετο, 17.16) by the
sight of a city overgrown with idols. What the Athenians had made of
nature was not natural theology (in a proper sense) but natural
idolatry. This is not inconsistent with Romans 1, but it lacks the
distinctive Pauline analysis and we cannot think that the speech was
delivered by Paul; it comes rather from Hellenistic Judaism, adapted
for Christian purposes (see pp. 825f.).
It is in regard to the Law that the greatest problem arises. Luke
never says anything more positive about the Law than Paul’s ‘holy,
righteous, good, spiritual’ (Rom. 7.12, 14); only once does he
condemn it as a burden that neither we nor our ancestors have been
able to bear (Acts 15.10), and he never describes it as the origin of
sinful passions (Rom. 7.5). The Pauline dialectic is missing, and so is
the radical rejection of the Law (legalistically conceived) as an agent
of salvation. If Gentiles are to be saved they must accept certain legal
conditions—a sharply reduced list of conditions but conditions
nonetheless. It is not surprising that Paul was obliged to reject the
Decree of Acts 15.29, or at least to ignore it. Nor is it surprising that,
as the textual phenomena show,6 the Decree was given different
interpretations; it may sometimes have amounted to little more than a
request for consideration and courtesy on the part of Gentile
Christians taking meals with Jewish Christians.
It is at this point, as have seen, that Luke’s lack of penetrating and
radical thinking reacts upon the historical worth of his book. Those
whose thought was like his own he can represent successfully even
4See Romans 20-22, on Rom. 1.3,4.
5See my lecture ‘Paulus als Missionar und Theologe' in ZThK 86 (1989) 18-32;
also Paulus und das antike Judentum, ed. Μ. Hengel and U. Heckel, Tübingen, 1991,
1-15; ET in Jesus and the Word, 149-62.
6See on 15.20, 29 (pp. 735f., 746).
cxviii COMMENTARY ON ACTS

when he lacks precise contemporary sources. The preaching of Peter

in the early chapters has, as has often been remarked, a primitive
appearance, but it is the kind of primitiveness that belongs to the 80s
as well as the 30s, and indeed to countless excellent but unreflecting
Christians in every age. It manifests an absolute loyalty to Jesus
Christ as a person but little attempt to evaluate him as very man and
very God—or even in less orthodox but equally reflective terms.
With Paul a new dimension entered Christian life; and it is probably
no more than truth to say that Luke (like many others) was unable to
see the difference between an approach to the Gentiles that rested
upon the discovery that Gentiles might be as good as Jews, and an
approach that sprang from the shattering discovery that Jews were
sinners just as much as Gentiles, though perhaps in a different way. It
is certain that the author of Acts regarded Paul as the outstanding
missionary to Jews and Gentiles, and that to find fault with him was
no part of his plan. To identify him with the kind of Hellenistic-
Jewish-Christian message accepted in his own day was for Luke a
natural error; but it was an error, which needs correction from Paul’s
own letters and must not be allowed to determine our picture of the
first Christian century. Both parts of the quotation from F. C. Baur
given on pp. lxxiiif. are true. Acts is a most valuable historical source
for the history of early Christianity; but it attains its full value only
when used with the strictest—historical and theological—criticism.
 X
THE COUNCIL IN JERUSALEM
(15.1-35)

38. DISPUTE IN ANTIOCH 15.1-5

(1) Certain people came down from Judaea and began to teach the brothers,
‘Unless you are circumcised in accordance with the Mosaic practice you
cannot be saved.’ (2) Between them and Paul and Barnabas no small
measure of contention and discussion arose, and they appointed Paul and
Barnabas, and certain others of their number, to go up to the apostles and
elders in Jerusalem about this subject of dispute. (3) So, having been sent on
their way by the church, they travelled through both Phoenicia and Samaria,
recounting the conversion of the Gentiles, and they gave great joy to all the
brothers. (4) When they reached Jerusalem they were welcomed by the
church and the apostles and the elders and reported the things that God had
done by means of them. (5) But there stood up some believers who belonged
to the party of the Pharisees, saying, ‘It is necessary to circumcise them and
command them to observe the Law of Moses.’

Bibliography
E. Bammel, Kremer, Actes, 439-46.
C. K. Barrett, FS Black (1969), 1-14.
C. K. Barrett, FS Black (1979), 15-27.
R. Bauckham, JSNT 2 (1979), 61-70.
P. Benoit, RB 64 (1957), 35-47.
J. D. G. Dunn, JSNT 18 (1983), 3-57.
T. Holtz, NovT 16 (1974), 110-48.
J. L. Houlden, JSNT 18 (1983), 58-67.
B. Orchard, CBQ 7 (1945), 377-97.
B. Reicke, FS de Zwaan, 172-87.
E. Richard, in C. H. Talbert, Luke-Acts 188-209.
C. H. Talbert, NovT9 (1967), 26-40.

Commentary
Chapters 13 and 14 contain an account of the commissioning of Paul
and Barnabas to act as missionary envoys—called apostles, perhaps
(14.4, 14)—of the church at Antioch and of their journey through
695
696 COMMENTARY ON ACTS

Cyprus and parts of Asia Minor. At the end of ch. 14 they report to
those who sent them out, and with the new paragraph a new stage in
the narrative of Acts begins. In the course of the missionary tour
Gentiles have been converted to Christianity. This had already
happened in Antioch itself (11.20f.) and had been, apparently,
accepted without question; it had also been approved by a representa-
tive of the church of Jerusalem (11.22-24). Now however the
objection is raised by travellers from Judaea that since circumcision
and legal observance are the marks of Judaism those who would be
members of the people of God must accept these as necessary
conditions. Those who maintain this view are opposed by Paul and
Barnabas; it is tacitly accepted that the question cannot be settled in
Antioch; it must be taken to the apostles and elders in Jerusalem,
where Paul and Barnabas, and some others, are to represent the
Antiochene point of view. They set out for Jerusalem, and passing
through Phoenicia and Samaria report what has been happening, to
the approval and joy of all the Christian brothers. In Jerusalem
however the demand for legal observance is repeated by Christian
Pharisees. The renewed demand sets in motion a debate, the
conclusion of which sets the scene for further missionary work and
thus opens the way to the rest of the book.
The debate itself (15.6-29) is rightly described as the centre of
Acts (‘ihr Herzstück’—Beyer 91); see pp. 709f. The present small
paragraph is important in that it introduces the debate, setting out
clearly the question at issue: Are circumcision and the keeping of the
Law necessary for salvation or not? One view is that Luke allows the
theme to shift in the course of ch. 15 from this fundamental problem
of theology to the practical question of the terms on which Jewish
Christians and Gentile Christians might have fellowship, especially
at the common Christian meal, within one society; see pp. 700, 705,
717, 720f., 745f. This however is not what Luke says the debate is
about (though it may represent one way in which the decree of 15.29
was used).
Luke’s hand is clearly to be seen throughout this introductory
paragraph. The pair κατέρχεσθαι, ἀναβαίνειν, is Lucan; so is the use
of ἔθη for Mosaic commands, σωθήναι for entering upon the
Christian faith, the litotes of οὐκ ολίγης, the description of Christian
ministers as πρεσβύτεροι, the use of ζήτησις, διέρχεσθαι,
έκδιηγεῖσθαι, not ἐπιστροφή but ἐπιστρέφειν, παραγίνεσθαι,
παραδέχεσθαι, the use of μετά in v. 4, αἵρεσις, and the use of
πεπιστευκότες for those who have become Christians. Evidence is
given in the notes. It does not seem possible to distinguish sources,
though μέν οὖν might have marked the beginning of a paragraph.
Luke however has not simply invented his material; see e.g.
Lüdemann (176-8), with the conclusion that there is in the tradition
behind ch. 15 ‘eine hohe historische Zuverlässigkeit’. This may be
 38. DISPUTE IN ANTIOCH. 15.1-5. 697

too favourable, as Pesch (2.72-74) and Weiser (376) are too precise
in distinguishing Antiochene and Jerusalem (or Decree) traditions;
see below. But we have the first-rate evidence of Galatians that there
was a meeting in Jerusalem, attended on the one side by James, Peter,
and John, and on the other by Paul, Barnabas, and at least one
other—Titus (Gal. 2.1-10). We know too that certain (unnamed)
people went from Judaea (in fact from James in Jerusalem) to
Antioch and there caused a separation between Jewish and Gentile
Christians (Gal. 2.11-14). The same epistle makes clear that there
was an attempt to persuade all Gentile Christians that circumcision
was necessary for their salvation. The correspondence is not exact for
in Acts the Jerusalem travellers to Antioch appear first, the Council
follows; in Galatians the order is reversed. Luke’s order has the
effect—an intended effect?—of representing the disagreement as
only temporary, whereas Galatians shows that it was not ended by
the Council, and was intense at the time the letter was written. We
may say that Luke knew that there was a meeting in Jerusalem; knew
that Jerusalem made trouble for Antioch; he knew that there was a
compromise Decree (see § 39); he knew that when he wrote
controversy had subsided and the church was at peace. He did not
know precisely how these data were to be fitted together, and in
making his own composition he had no intention of going out of his
way to present a picture of a disorderly and inharmonious church.
For further analysis and consideration of the chronology see the
introduction to § 39 and the general Introduction, pp. xxxvi-xl, lvi,
lxf.

1. Καί τινες κατελθόντες ἀπό τής Ίουδαίας. In the absence of

any indication to the contrary it must be assumed that the scene is
unchanged; that is, the persons in question came to Antioch (see
14.26-28). There is no immediate reference (but see 15.3, 12) to the
missionary activity of chs. 13 and 14; all that is necessarily
presupposed is the existence in Antioch (see 11.20) of a mixed
church containing (whether this be regarded as proper or improper)
uncircumcised Gentiles as well as Jews. This means that the new
material could follow directly upon 11.27-30—a fact to be borne in
mind when the chronology of Acts, and especially whether the ‘First
Missionary Journey’ should precede or follow the Council, is
considered.
The τινες are defined only by their actions (which appear imme-
diately) and by the place from which they come. They were από τής
Ίουδαίας, from Judaea. Luke does not say at this point that they
came from Jerusalem, though in fact they did so (15.24); Con-
zelmann (82) and Roloff (228) think that Luke omitted any reference
to Jerusalem at this point because he did not wish to suggest that the
trouble-makers had had any support from the leaders of the mother
698 COMMENTARY ON ACTS

church. This may very well be true; but the travellers from Jerusalem
are immediately disowned at 15.24 (οἷς ού διεστειλάμεθα), and this
could have been done equally well at the present point. Judaea may
be intended in an ethnic rather than a strictly geographical sense;
they came from Jewish territory and may therefore be expected to
represent a Jewish point of view. They are defined more explicitly by
Ψ 614 pc syhmg, which, after Ίουδαίας, add τῶν πεπιστευκότων ἀπό
τής αἱρέσεως των Φαρισαίων; this secondary reading is due to
assimilation to v. 5.
κατελθόντες is geographically correct, since Antioch was on, or
near, the coast, but ἀνέρχεσθαι, κατέρχεσθαι, and similar com-
pounds were used of pilgrimages to and departures from the capital;
see on 11.2. This may hint at an authorized visitation; cf. 8.15
(καταβάντες). Cf. Gal. 2.12, πρὸ τού γάρ έλθεῖν τινας ἀπὸ
Ιακώβου—from James, and therefore presumably from Jerusalem.
The two groups can be identified only if the Jerusalem meeting of
Gal. 2.1-10 preceded that of Acts 15, or if we suppose that Galatians
2 does not follow but inverts chronological order. Neither of these is
probable; it remains possible however that either Paul or Luke has
misplaced the Judaizing visitation. The whole question of the
reconstruction of the events behind Galatians 2 and Acts 15 is thus
already raised. It cannot be settled or even sensibly discussed on the
basis of the present verse alone, but it (and Gal. 2.12) must be kept in
mind.
ἐδίδασκον is presumably an inceptive imperfect: when they
arrived they set about teaching. The visitants taught τούς αδελφούς.
For the use of αδελφός see on 1.15; it implies that the persons in
question are Christians (since by definition they are not fellow Jews).
It thus grants the question under discussion. This means of course
only that to Luke they are Christians; it was quite clear to him that
circumcision was not necessary. The visitants took the opposite
view. Without circumcision there is no salvation. Circumcision is a
Mosaic requirement: ἐάν μὴ περιτμηθήτε τφ ἔθει τῷ Μωυσέως. The
Western text (D (syp) sa mae) makes it clear that more than an
initiatory rite is required: ἐὰν μὴ περιτμηθῆτε καὶ τω ἔθει Μ.
περιπατῆτε (note the present—continuous—tense). This is implied
though not mentioned by the Old Uncial text; there would be no
point in being circumcised and then neglecting to keep the Law.
Characteristically the Western text leaves nothing to imagination—
or to common sense, έθος is not adequately rendered by custom: it
refers to the practice originated by Moses (though in fact circumci-
sion goes back to Abraham; Gen. 17.10-14), and this has the force of
law. Cf. 6.14; 16.21; 21.21; 26.3; 28.17; also 2 Macc. 11.25 (τὰ ἐπί
των προγόνων αυτών έθη); 4 Macc. 18.5; Josephus (e.g. Ant.
20.100, τοῖς γάρ πατρίοις οὐκ ἐνέμεινεν ουτος ἔθεσιν); Philo (e.g.
Spec. Leg. 2.148, πάτριον έθος). This use of the word seems to have
 38. DISPUTE IN ANTIOCH. 15.1-5. 699

been a Jewish development (not noted in LS); see however Ditten-

berger, Syll. 2.1073.20f., κατά τό πάτριον τῶν αγώνων ἔθος. This
use (with έθος) of κατά might have been expected rather than Luke’s
dative. This is described by Μ. 3.242 as a dative of cause (...
'because of the law'). BDR § 196.1, n.1 also classifies the use as
Dativus Causae, but—surprisingly—translates ‘gemäss’ (in con-
formity with?), and compares PHolm 2.18, τῆτε (legd. τῇδε) τάξει,
‘nach diesem Rezept’. This seems to be Luke’s meaning: Gentiles
must be circumcised in accordance with the Mosaic practice.
The Judaeans do not say: Gentiles cannot be saved at all. They say:
You cannot be saved unless you are circumcised. This almost all
Jews would have allowed, for though some Jews were more, others
less, enthusiastic about making proselytes, it was generally recog-
nized that Gentiles, if they complied with the necessary conditions,
might enter the Jewish fold. The Judaeans simply affirm the familiar
proposition: the Jews are the elect people of God, and male Jews are
circumcised—as infants if born into a Jewish family, otherwise upon
conversion. Exceptions to this requirement are hard to find. Josephus,
Ant, 20.38-48 is only a partial exception. Izates, king of Adiabene,
wished to become a Jew, and supposed that circumcision (though it
would be for the king of a non-Jewish people a perilous act) was
essential. The Jew Ananias said that it was not, δυνάμενον δ’αὐτὸν
καὶ χωρίς τῆς περιτομής τὸ θειον σέβειν, εἴγε πάντως κέκρικε
ζηλοῦν τά πάτρια των ’Ιουδαίων τοῦτ’ εἶναι κυριώτερον τοῦ
περιτέμνεσθαι συγγνώμην δ’ ἕξειν αὐτω καί τόν θεόν ... (41).
Another Jew, however, Eleazar from Galilee, maintained the con-
trary view, and the rite was carried out. Thus Ananias was overruled;
moreover he seems to have argued on the principle that a command
might be omitted if it was dangerous to life. At Ant. 20.139 Azizus,
king of Emesa, and at 145 Polemo, king of Cilicia, are circumcised,
in each case in order to marry a Jewish woman. At Yebamoth 46ab
the dispute is not whether or not circumcision is necessary but
whether baptism or circumcision marks the precise moment of
conversion.
σωθῆναι. Certainly not here, perhaps nowhere, does Luke define
what he means by being saved; see on 4.12. He probably thought the
matter too obvious to warrant discussion. We may, in the context,
paraphrase it as ‘to receive in full the benefits provided by God for
his people’, without specific reference to particular benefits. It
appears that Luke tells us in this opening verse what the argument of
the chapter is to be about: it will be about being saved, about being a
Christian at all, not about the regulation of relations between Jewish
Christians, who wish to retain their Jewishness, and Gentile Chris-
tians, who do not wish to become Jews. This appears to be the issue
raised in Galatians 2, and answered by implication in 2.3, ουδέ Τίτος
... ήναγκάσθη περιτμηθῆναι. ‘The issue in chapter 15 is thus not
700 COMMENTARY ON ACTS

merely post-conversion behaviour but what constitutes true conver-

sion in the first place’ (Wilson, Law 72). Whether Luke adheres to
this issue throughout the chapter is a question that must be kept in
mind. It must also be remembered that the matter seemed to have
been settled at 11.18; see the notes. If that verse does not mean that
for Gentiles repentance, without circumcision, suffices for life—
salvation—it is meaningless. On these questions see further p. 696
and the references there.

2. The variant οὖν (P74 A E <a> d 1 vg syh) for δέ ( B C D L Ψ

36 81 453 945 1175 al gig p) emphasizes, what is in any case clearly
implied, that it was in consequence of the arrival of the Judaeans, and
of their stand, that there arose no small (the litotes is characteristic of
the later part of Acts—12.18; but 15.2; 19.11, 23, 24; 20.12; 21.39;
26.19, 26; 27.20; 28.2) στάσεως καὶ ζητήσεως (καί ζ. is omitted by
P74 E vg bo). ζήτησις is the word that would be expected; it never
quite loses its normal sense of inquiry (see LS 756); BA’s Wortge-
fecht (686) perhaps suggests controversy too strongly, though this is
supported by the context and not least by the companion word στάσις
(‘... the well-known classical word for an outbreak between the
democratic and oligarchical parties in a state’—Page 175). Cf. Mk
15.7, but also Acts 23.7, 10; Luke does not mean to suggest that the
conflict ended in murder, but sharp contention (not merely discus-
sion) is clearly indicated. Josephus, Ant. 18.374, ἐν στάσει καί
διχονοία, resembles Luke’s combination, and here the conflict
(between Greeks and Syrians at Seleucia—not far from Antioch)
was serious enough.
Paul and Barnabas, leaders in the church of Antioch (11.26) and
pioneer missionaries on its behalf (13.1-3), are named as leading
contenders on the non-circumcision side; cf. Gal. 2.4f.
The opening clause in the verse (γενομένης ... πρὸς αυτούς) is a
genuine genitive absolute; this means that the subject of έταξαν
cannot be inferred from the context. Hanson (159) thinks it possible
that the subject was intentionally left vague, but adds that syntactical
usage demands as subject those who came from Jerusalem. This is not
certain; προπεμφθέντες in v. 3 suggests rather that it was the mem-
bers (or possibly the prophets and teachers—Preuschen 93) of the
church at Antioch who were the subject of έταξαν. τάσσειν with
accusative and infinitive means, according to LS (1760; s.v. II 2), to
'appoint or order one to do or be’. In the present passage appoint
seems suitable, but in a number (not all) of the passages cited by LS
the verb is used in parallel with κελεύειν. Nevertheless, Barnabas and
Paul were leaders in the church at Antioch, and it is best to render,
‘They (the Christians at Antioch) appointed Paul and Barnabas and
certain others to go up’—for ἀναβαίνειν cf. κατελθόντες in v. 1.
The companions of Paul and Barnabas (τινας ἄλλους) are not
 38. DISPUTE IN ANTIOCH. 15.1-5. 701

named; at Gal. 2.1 Titus accompanies them. Galatians does not

exclude the possibility that others went too, and the silence of Acts
does not mean that Titus was not one of the άλλοι—nor does it mean
that Titus was Luke’s relative, whose name Luke omitted as he
omitted his own (see FS Black (1969), 2f.). Mission by appointment
is however different, at least in emphasis, from Gal. 2.2, ἀνέβην δέ
κατά ἀποκάλυψιν. But the two are not irreconcilable, The human
arrangement may have been the result of divine guidance; cf. 13.2,
which, written from a different angle, could have been expressed,
‘The prophets and teachers appointed Paul and Barnabas to ...’.
It was argued above that the subject of έταξαν is the Christians of
Antioch, but the nearest antecedent is αὐτούς, the Judaeans, and
these are taken to be the authors of the arrangement by the Western
text, which instead of έταξαν ... αυτών has ἔλεγεν γάρ ό Παύλος
μένειν ούτως καθώς ἐπίστευσαν διισχυριζόμενος. οἱ δέ
ἐληλυθότες από ’Ιερουσαλὴμ παρήγγειλαν τω Παύλῳ καί
Βαρναβᾷ καί τισὶν ἄλλοις ἀναβάινειν (D (gig w syhmg mae)) and
adds after a second ’Ιερουσαλήμ, ὅπως κριθῶσιν ἐπ’ αὐτοῖς (D(c);
and with varying order 614 pc syh**). See also on 15.7. This is one of
those passages in Acts (see Introduction, pp. xxif.) where the Old
Uncial text and the Western text are said to give fundamentally
different views of an event. In the Old Uncial text arrangements are
made for a debate on equal terms; in the Western text Paul and
Barnabas and other members of the erring church are peremptorily
summoned to Jerusalem to stand trial. The difference is less great
than is sometimes supposed. Each side of the debate expresses its
views more forcefully. If the Judaeans παρήγγειλαν, Paul is repre-
sented as διισχυριζόμενος (for the word see 12.15; Lk. 22.59; also
PMich 659.14, cited in ND 2.81). The Western characteristic that
appears here is that of sharpening the picture, making the story more
vivid and exciting; see 1.22; FS Black (1979), 15-27.
Black, AA 104 (cf. Wilcox 132) notes that the Western text has
αὐτοῖς τῷ Π. καί Β., and suggests that αὐτοῖς represents an Aramaic
ethic dative; cf. Metzger 428 (‘... a clear example of the Semitic
proleptic pronoun’). If there were strong evidence for a continuous
Aramaic source at this point the suggestion might be convincing, but it
is more probable that the Western editor wrote at first ordered them,
and then thought that he had better specify who they were.
The Antiochian representatives were to consult with—or be
judged by—the apostles and elders, cf. 15.4, 6, 22, 23. For the
apostles see I.94f., and frequently ; for the elders see on 11.30; 14.23.
It is not at this stage clear in what category James (see I.586f.),
who plays a notable part in Luke’s account of the Council
(15.13-21), is to be placed. Cf. 12.17; 21.18. In terms of the
requirement of 1.22, 23, he was no more an apostle than Paul, but he
certainly stands out beyond the unspecified elders. See Gal. 2.9; also
702 COMMENTARY ON ACTS

1.19, which may or may not refer to him as an apostle. Lack of a

specific designation may correspond to historical fact: he was
undoubtedly an important and influential person, but one who owed
his influence to a special relation with Jesus (Gal. 1.19) and to the
strength of his character and convictions rather than to any definable
office.
The parties are to consult about τοῦ ζητήματος τούτου. The
ζήτημα is the specific matter upon which the ζήτησις (see above)
centres. Cf. Plato, Laws 1.630e-631a: ἡμεῖς δέ φαμεν είναι τό περί
νόμους ζήτημα τῶν ευ ζητούντων.
Bultmann (Exegetica 417) notes how easy it would be, here and at
15.3-5, 12, 22f., 25f., to drop the named references to Paul and
Barnabas, regarding them as redactional supplements to a source. For
this theory see on the verses in question, especially 15.12.

3. μὲν οὖν often marks the beginning of a story (see on 1.6). Begs.
4.171 accepts this here. ‘What has gone before is structurally rather
the end of the previous narrative, though it is surely editorial and is
intended to lead up to the following narrative.’ If this were so one
would expect Luke to use his μὲν οὖν in his introductory editorial
note, unless he allowed it to stand where he found it in a source. It is
better to compare with 13.4, where μὲν οὖν marks the change from a
preparatory situation to the movement of a narrative.
The representatives were seen off, sent on their way (for
προπέμπειν cf. 20.38) by the ἐκκλησία (see I.271), and then
journeyed by stages to Jerusalem. The farewell is expressed in the
aorist tense, the journey by the imperfect (διήρχοντο, ἐποίουν), and
arrival again by the aorist (παραγενόμενοι , v. 4); see BDR § 327.1,
n. 1, who bring out the force of ἐποίουν by ‘überall, jedesmal’. The
present participle ἐκδιηγούμενοι (narrating) matches the imperfect
indicatives.
At least sometimes in Acts (e.g. 13.8) διέρχεσθαι has almost the
technical sense of going on a preaching tour. Perhaps Luke means to
suggest that so far from being inhibited Paul and Barnabas took the
opportunity of the journey to Jerusalem to continue their mission.
Luke’s main interest, however, if not his only interest, is in the report
given by Paul and Barnabas and its reception in the districts through
which they passed.
In Acts as it stands the ἐπιστροφή των ἐθνῶν will refer mainly to
the events of chs. 13 and 14. The reference could however be
satisfied by the conversion of the Antiochene Gentiles (11.21; note
the verb (ἐπέστρεψεν)) if the chs. 13 and 14 are thought to be
misplaced; see Introduction, pp. lxf. This is the only occurrence of
ἐπιστροφή in the NT; ἐπιστρέφειν however occurs at 3.19 (see the
note); 9, 35, (40); 11.21; 14.15; 15.19, (36); (16.18); 26.18, 20;
28.27. ND 2. 72, noting papyrus use, comments, ‘Whether the word
 38. DISPUTE IN ANTIOCH. 15.1-5. 703

[επιστροφή] in its only NT occurrence ... need mean as much as

“conversion” is at least worth querying.’ The use of the cognate
verb suggests that the query is unrewarding.
Zerwick (§ 227) probably presses too hard the distinction between
the active ποιεῖν and the middle, though causabant gaudium is
undoubtedly correct for ἐποίουν χαράν. All the Christians (πᾶσιν
τοῖς ἀδελφοῖς) encountered by Paul and Barnabas on their journey
approved of and rejoiced in the mission to the Gentiles. Public
opinion was on their side, and the battle was over before it was
fought. In fact the Council turns out to be a sham fight; no dissenting
voice is heard. We know however from the Pauline epistles that there
was some very serious fighting indeed—another fact that must be
borne in mind when we discuss the tradition available to Luke and
his handling of it. Apparently he knew that there was a dispute, but in
order to represent it as speedily dealt with had to invent a few ill-
disposed and easily beaten trouble-makers in order to account for it.
But he is correct in speaking of a mission from Judaea to Antioch,
even if he misplaces it.
Phoenicia (see 11.19; 21.2) is not a precise geographical term; it
denotes the coastal area of Palestine stretching northwards from
Carmel and including Tyre and Sidon. It borders in the south on
Samaria (see 8.5, 14, 25). Stählin (201) thinks that the churches in
Phoenicia had been founded by Hellenist refugees.
At this point the Coptic MS G67 ends; see I.13f. The Gentile
mission is now accepted by all the (Christian) brothers. The book
may finish here; the author’s goal is already reached.

4. For the tense of παραγενόμενοι see on v. 3. In this verse the

textual authorities are fairly evenly divided between the two spellings
of Jerusalem; ’Ιερουσαλήμ in C D E SR, Ίεροσόλυμα in P45 P74 A B
Ψ 81 614 1175 2495 pc gig vg. There is little doubt that
παρεδέχθησαν is what Luke wrote, though ὑπεδέχ. and ἀπεδέχ. both
have some slight support; the adverb μεγάλως is added by C D(*)
6 614 1704 pc syh** sa—the Western editor, notwithstanding his
revision of v. 2, wished to underline the warmth of the welcome
accorded to Paul, Barnabas, and their colleagues when they reached
Jerusalem. This confirms the view that he is emphasizing what he
finds rather than introducing new points of view.
They were welcomed by (Metzger 428 takes the more Semitic ἀπό
(B C 36 453 1175 pc) to be original rather than υπό (P74 ADE
Ψ <a>)) the church and the apostles and elders. Church (ἐκκλησία)
here probably refers to those Christians who did not hold office,
whether as apostles or elders. For the extent to which (in Luke’s
view) these members participated in the Council see on 15.6, 12,22,
23. For the apostles and elders see on v. 2. Wilson (Gentiles 182),
following Wikenhauser and Stählin, suggests that in the present verse
704 COMMENTARY ON ACTS

we have a preliminary open meeting, whereas in 15.6 there is a

meeting of the apostles and elders alone, at which ‘the real business
was decided’.
ανήγγειλαν (the imperfect of the same verb is used at 14.27) is
parallel to ἐκδιηγούμενοι in v. 3; reported, perhaps, rather than
recounted, ὅσα ἐποίησεν ὁ θεός μετ’αύτῶν means ή ἐπιστροφὴ τῶν
ἐθνῶν. This too reproduces the language of 14.27. This use of μετά
occurs only in Lk. and Acts (Lk. 1.72; 10.37; Acts 14.27; 15.4);
Wilcox 84 considers whether it is taken from the LXX or direct
Semitic influence. The former is the likelier alternative. The conver-
sion of the Gentiles had been the work of God; Paul and Barnabas
had acted as his agents, but it was his doing. This in itself proved that
it was right; it needed no further defence.
The present verse reproduces the statements and themes of vv. 2
and 3; Luke wishes to represent the journey to Jerusalem as
something like a triumphal progress. The Gentile mission has begun
and it is already clear that nothing is going to stop it. This intended
emphasis provides a better explanation of the repetition than a theory
of parallel sources; but see below.

5. What had been done could not be undone; but it could be

maintained that it was right only so far as it went. It was necessary
now to complete what had been done by circumcising the converts
and instructing them henceforth to obey the Law of Moses. Cf. Gal.
5.2, where Paul’s words seem to imply the proposition (inverted by
Paul), If you are not circumcised Christ will do you no good. Those
who make this demand are described by Luke as believers
(πεπιστευκότες) and as coming from the party (αἵρεσις; for the word
see on 5.17) of the Pharisees. According to Kosmala 110f., ‘Dies
braucht noch nicht zu heissen, dass sie an den Messias Jesus
glaubten, denn der Sprachgebrauch des Lukas erfordert das kei-
neswegs. Danach könnten sie ebensogut nur Essener oder Johannes-
jünger geworden sein, und sie brauchen darum auch gar nicht zu der
in xv. 4 genannten jesusgläubigen Gemeinde gehört zu haben.’ This
is hardly correct. Luke’s usage does in fact connect πιστεύειν with
acceptance of the Christian message about Jesus; so e.g. 2.44 (oἱ
πιστεύοντες); 4.4 (πολλοὶ ... ἐπίστευσαν), 32 (των πιστευσάντων);
et al. Earlier references to the Sadducees (4.1; 5.17) have carried the
possible implication that the Pharisees (see on 5.34) were more
favourably disposed to the new faith; Luke’s statement here confirms
the implication (at least, as regards his opinion, if not as regards fact).
These Pharisees had presumably come to believe that Jesus was the
Messiah without changing their views of the Law. It is however
probably correct that ‘Welche Authorität die Gruppe in Jerusalem
hatte, fragt Lukas nicht, da ihm nicht die historischen Zusammen-
hänge, sondern die systematische Frage am Herzen lag’ (Schille
 38. DISPUTE IN ANTIOCH. 15.1-5. 705

319). But Luke saw theological questions in historical contexts, and

no doubt believed that the position described was one that Pharisees
would adopt.
These Pharisaic believers stood up—and perhaps stood out—for
their opinion, ἐξανέστησαν (cf. e.g. Xenophon, Anabasis 6.1.30,
ἐξαναστάς εἶπε Ξενοφῶν ...); perhaps in order to take part in a
discussion, but the word may be metaphorical and simply represent
their readiness to put their point of view. If there is any special
significance in the compounded ἐξ it will be that they stood out from
among their fellow believers, who adopted a more accommodating
attitude.
ὅτι introduces direct speech. Now that the scene has shifted to
Jerusalem it cannot be certain whether αυτούς refers to the Gentile
Christians in Antioch or to Gentiles (including Titus, Gal. 2.1, 3)
whom Paul and Barnabas had brought with them to Jerusalem. Verse
5 repeats v. 1 (with the addition that after circumcision observance of
the Law would be required), as v. 4 to some extent repeats v. 3. If
Luke is here writing freely in order to introduce the Council he
probably adds v. 5 in order to make clear that there really was an
issue to debate; if there were those who rejoiced in the success of the
mission there were also those who took a hard line, asserting ‘... sine
circumcisione salutem obtineri non posse’ (Bengel 448). This is
perhaps the best view; alternatively, there may have been two
sources, vv. 1, 2 and vv. 3-5, which Luke put together, or one
source, vv. 3-5, to which Luke added vv. 1,2 as an introduction. It is
worth noting that v. 3 could be attached directly to 14.28, or even
11.30, though not as satisfactorily as v. 1 (see above). If we set this
passage beside Galatians 2 we may note Gal. 2.4: the ψευδάδελφοι
creep in to spy out our liberty, ἵνα ἡμᾶς καταδουλώσουσιν. To be
saved, men must become Jews. There is no question that at this point
this proposition is given as the theme of the Council. It is often held
that in the course of the chapter the theme changes, so that the Decree
(15.29) answers the question, not On what terms may Gentiles be
saved?’ but 'On what terms may Jewish Christians and Gentile
Christians have fellowship, including table-fellowship, in one
body?’. Whether this is so will be considered below (p. 745). There
appears to be a comparable shift in ch. 11, where the discussion starts
from the complaint that Peter had had dealings and had eaten with
Gentiles, but ends with the conviction that uncircumcised Gentiles
may as such be saved. This may well mean that Luke himself did not
make a clear distinction between the two questions.
 39. COUNCIL IN JERUSALEM 15.6-29

(6) The apostles and elders gathered together to look into this matter. (7)
When much discussion had taken place, Peter stood up and said to them
‘Brothers, you know that in days of old God made his choice among you that
through my voice the Gentiles should hear the word of the Gospel and
believe; (8) and God, who knows the human heart, bore testimony to them
when he gave them the Holy Spirit, as he had done to us, (9) and he made no
distinction between us and them when he cleansed their hearts by faith. (10)
So now why do you put God to the test by laying on the neck of these
disciples a yoke which neither our fathers nor we were able to bear? (11) On
the contrary, we believe that it is through the grace of the Lord Jesus that’ we
shall be saved, in the same way as they.’ (12) The whole company fell silent,
and they listened to Barnabas and Paul as they narrated the signs and portents
that God had done among the Gentiles through them. (13) When they had
ceased speaking, James joined in,2 saying, ‘Brothers, listen to me. (14)
Symeon has reported how3 God at the first took action so as to take out of
the Gentiles a people for his name; (15) and with this the words of the
prophets agree—as it is written: (16) “Afterwards I will return, and I will
build up again the tent of David that has fallen down, and I will build up
again its ruins, and I will raise it up, (17) in order that the rest of mankind
may seek the Lord, even all the Gentiles4 upon whom my name has been
named.’’ So says the Lord,5 making these things (18) known from of old.
(19) Therefore for my part I give my judgement that we should not make
trouble for those who from the Gentiles turn to God, (20) but write to them
that they should abstain from the defilements caused by idols, from
fornication,6 from that which has been strangled, and from blood. (21) For
Moses from generations of old has in every city those who proclaim him,
since he is read out every Sabbath in the synagogues.’
(22) Then the apostles and the elders, along with the whole church,
decided to choose men from among themselves to send to Antioch with Paul
and Barnabas; [they chose] Judas called Barsabbas and Silas, leading men
among the brothers, (23) and wrote for them to deliver [as follows:] ‘The
apostles and the elders, your brothers, to the brothers in Antioch, Syria, and
Cilicia, who come from among the Gentiles: Greeting. (24) Since we had
heard that some of us had disturbed you by what they said, unsettling your

1NEB, We are saved, and so are they; NJB, We are saved in the same way as they
are.
2NEB, summed up.
3NEB, it first happened that God took notice of the Gentiles. NJB, God first arranged
to enlist a people for his name out of the Gentiles.
4NEB, whom I have claimed for my own.
5NEB, whose work it is, made known long ago.
6NJB, illicit marriages.
706
 39. COUNCIL IN JERUSALEM. 15.6-29. 707

souls,—men to whom we had given no such instructions—(25) we,7 when

we met together, decided to choose and send men to you along with our
beloved Barnabas and Paul, (26) men8 who have devoted their lives to the
name9 of our Lord Jesus Christ. (27) So we have sent Judas and Silas, who
will report the same things to you by word of mouth. (28) For the Holy Spirit
and ourselves have reached the decision that we should lay upon you no
other burden except these necessary things: (29) that you should abstain from
things sacrificed to idols, from blood, from things that have been strangled,
and from fornication.10 If you keep yourselves from these things you will be
doing right. Farewell.’

Bibliography
E. Bammel, as in (38).
C. K. Barrett, FS Osborn, 50-59.
C. K. Barrett, FS Hofius, 323-39.
J. B. Bauer, BZ 32 (1988), 114-17.
P. Benoit, 3.285-99 (= Bib 40 (1959), 778-92).
C. Blomberg, JSNT 22 (1984), 53-80.
Μ. E. Boismard, EThL 64 (1988), 433-40.
T. Boman, NovT 7 (1964-5), 26-36.
P. Borgen, FS Kee, 125-41 [= Early Christianity and Hellenistic Judaism
(1996), 233-51].
U. Borse, FS Zimmermann, 195-212.
F. F. Bruce, FS Greeven, 115-24.
R. Bultmann, FS T. W. Manson, 68-80 (= Exegetica 412-23).
A. Campbell, JTS 44 (1993), 511-28.
D. R. Catchpole, NTS 23 (1976-7), 428-44.
L. Cerfaux, Rec. 2.105-23.
O. Cullmann, Petrus.
N. A. Dahl, NTS 4 (1958), 319-27.
F. W. Danker, Cassidy and Sharpes, Political Issues, 49-58.
W. Dieterich, Petrusbild.
J. Dupont, NTS 3 (1957), 47-50 (= Études 361-5).
J. Dupont, NTS 31 (1985), 321-35.
J. Eckert, in J. Ernst (ed.), Schriftauslegung (1972), 281-311.
J. Fitzmyer, Essays 271-303 (= FS Schubert, 233-57).
P. Gaechter, ZKTh 85 (1963), 339-54.
B. Gerhardsson, Memory and Manuscript (1961), 249-61.
Μ. D. Goulder, Two Missions.
7NEB, We have resolved unanimously; NJB, We have decided unanimously.
8RSV, men who have risked.
9NEB, cause.
10NJB, illicit marriages.
708 COMMENTARY ON ACTS

R. Μ. Grant, HThR 73 (1980), 299-310.

E. Haenchen, FS Jeremias, 153-64.
T. Holtz, as in (38).
P. W. van der Horst, NovT 16 (1974), 309.
J. Jensen, NovT 20 (1978), 161-84.
G. D. Kilpatrick, Kremer, Actes 84-6.
A. F. J. Klijn, NovT 10 (1968), 305-12.
Μ. Klinghardt, Gesetz und Volk Gottes, 158-224.
W. G. Kümmel, Heilsgeschehen und Geschichte (1965), 278-88.
H. Lietzmann, FS Harris, 203-11.
J. L H. McDonald, TU 126 (= StEv VII) 327-32.
J. Manek, Comm. Viatorum 15 (1972), 151-60.
P. H. Menoud, SNTS Bulletin 2 (1951), 22-8.
E. Molland, StTh 9 (1955), 1-39.
J. Murphy-O’Connor, RB 99 (1982), 71-91.
J. Nolland, NTS 27 (1981), 105-15.
J. L. North, NTS 29 (1983), 264-6.
C. Perrot, RechScR 69 (1981), 195-208.
R. Pesch, FS Zimmermann, 105-22.
W. Pratscher, Jakobus (1987).
B. Reicke, as in (38).
B. Reicke, in Stendahl, Scrolls 146.
E. Richard, as in (38).
R. Richard, NovT 24 (1982), 37-53.
H. Sahlin, NovT 24 (1982), 187.
H. van de Sandt, JSNT 46 (1992), 73-97.
A. Schmidt, ZNW 81 (1990), 122-31.
D. R. Schwartz, Bib 67 (1986), 276-81.
Μ. Simon, BJRL 52 (1969-70), 437-60.
E. R. Smothers, HThR 46 (1953), 203-15.
A. Strobel, FS Zimmermann, 81-104.
C. H. Talbert, NovT 9 (1967), 26-40.
J. Taylor, RB 99 (1992), 373-8.
Y. Tissot, RB 77 (1970), 321-46.
C. H. Turner, Theology 20 (1930), 4-14.
A. J. Μ. Wedderburn, NovT 35 (1993), 362-89.
A. Weiser, BZ 28 (1984), 145-67.
P. Winter, EvTh 17 (1957), 399-406.
 39. COUNCIL IN JERUSALEM. 15.6-29. 709

E. Zuckschwerdt, ZNW 68 (1977), 122-31.

G. Zuntz, Opuscula Selecta (1972), 216-51.

Commentary
The meeting planned in the preceding paragraph takes place. It does
not take the form of a trial (so that the Western reading in 15.2 does
not correspond to Luke’s intention) but rather of a general discussion
of a question of practice. B. Gerhardsson (Memory and Manuscript
(1961), 249-61) compares it to a rabbinic discussion of a piece of
halakah, and thinks it to be an example of the διακονία τοῦ λόγου to
which the apostles propose to devote themselves in 6.4. It is the
apostles and elders who gather for discussion and evidently assume
the authority to make a decision; they write the letter in which the
decision is promulgated, though the whole church concurs (v. 22). At
the outset there is much debate, which Luke does not report. After
this he assigns speeches to Peter, to Barnabas and Paul, who are not
separated as speakers, and to James. Peter is firmly in favour of a
liberal attitude; the Law, he says, is an intolerable burden even to
Jews, and God has clearly shown that he does not require observance
from Gentiles. Barnabas and Paul report the miracles they have
witnessed among the Gentiles, a demonstration of God’s favour to
the Gentiles and of his approval of the way in which the evangelists
have conducted their mission. James’s attitude is less clear, though
he agrees with Peter and finds support for his position in the OT;
some requirements, however, must be made. The whole company
agrees with his conclusion, and a letter is written, disavowing those
who have caused confusion in the Antiochene church and stating the
decree proposed by James; if this is observed all will be well in the
Gentile churches.
This paragraph is rightly described as the centre of Acts. It is the
best example of a pattern that occurs several times in Acts and
represents the way in which Luke conceived the progress of Christi-
anity. In this pattern a difficulty is encountered; steps are taken to
deal with it; not only is the problem solved but a notable advance
takes place as a result. Thus for example at 6.1 the problem of the
Hellenist widows (which could have ended in schism) arises; steps
are taken to deal with it (6.2-6); the result (6.7) is a great increase in
the number of disciples. Similarly in 19.9 there is such vehement
opposition to Paul’s work in Ephesus that he is obliged to leave the
synagogue for another building; but the result (19.10) is that all the
Jews and Greeks in the province of Asia hear the Lord’s word. In ch.
15 the extension of the Gospel to the Gentiles (which is Luke’s
primary concern) is threatened by those who would compel all
Gentile converts to become Jewish proselytes; steps are taken to deal
with this problem; the result is not simply an answer to the problem
710 COMMENTARY ON ACTS

but the further expansion of the church. This appears immediately in

15.35, but is in fact the theme of the rest of the book. Thus Acts 15
supplies the key to the arrangement and movement of the book as a
whole. ‘Luc place le récit, tel qu’il l’a recomposé, au centre de
l’activité missionaire de Paul’ (Taylor 5.222).
Haenchen saw the account of the Council as a Lucan composition.
It was not indeed pure fiction; Luke wrote on the basis of traditions,
which are not further defined. With this view Bultmann disagreed
(Exegetica 415f. = FS T. W. Manson 71f.); Luke made use of written
sources. The sharpest point of disagreement was the Decree, together
with the letter of which it formed part. Bultmann agrees with
Dibelius that the address to Antioch, Syria, and Cilicia is a clear
indication of a written source. Luke believed the letter and Decree to
apply to all Gentile Christian churches; he would never have made
up so limited a statement of its destination. Certainly (in Bultmann’s
view) Luke edited the written source, or sources, that he possessed;
in particular he introduced the references to Barnabas and Paul.
Subsequent writers have taken different views of the composition
of the paragraph. According to Hanson (155) ‘The “Apostolic
Council” is an imaginative reconstruction by Luke’; but from this
Hanson excepts the letter and the Decree. Lüdemann 176 thinks that
ch. 15 rests on tradition, but is undecided whether this was written or
oral. He sees the marks of tradition in the concreteness of the
narrative and in its agreements with Galatians 2. Many think that
Luke has combined two traditions, based respectively on Jerusalem
and Antioch. Sometimes the combination is set out in literary form.
Thus Pesch (2.72) believes that Luke has combined a tradition of an
Apostolic Council with an account of the Apostolic Decree; more
precisely (2.74), he distinguishes Antiochene tradition (of the found-
ing of the church, 11.19-26, and of the delegation to Jerusalem,
11.27-30; 12.25; 15.1-4, 12b) and the origin of the Decree (10.1-
11.18; 15.5-12a, 13-33). Weiser (376) finds the story of a Jerusalem
meeting in 15.1,2,4, 5,6,12,7,13,10,11,19, and an account of the
Antiochene problem and its solution in 15.5, (1), 23, 30, 20, 29, 23,
22, 27, 30, 31, 32. ‘Lukas hat die Elemente beider Traditionen
miteinander verbunden und so die eindrucksvolle Gesamtszene
“Apostelkonzil” gestaltet.’ It may be questioned whether such
precise delineation of sources is possible; Luke gave himself too free
a hand, and believed that the Decree, formulated as the result of a
Council in Jerusalem, originated as the solution of the Antiochene
problem. That Luke used all the sources he could find, whether
written or oral, was argued already in I.49-56; he did so here. Of the
Decree Haenchen (454) writes, ‘Lukas hat also nicht ... diese vier
Forderungen einem alten Dokument entnommen, das er irgendwo
gefunden hat, sondern er hat eine lebendige Tradition beschrieben,
die man wahrscheinlich schon damals auf die Apostel zurückgeführt
 39. COUNCIL IN JERUSALEM. 15.6-29. 711

hat.’ That Luke found the Decree in use, observed by churches, is

true; but were his contemporaries so illiterate that they never wrote
down basic rules of their community? The paragraph reproduces
written material, folk memory, and Luke’s own story-telling.
The story invites comparison with Galatians 2. There are such
close parallels between the two passages that it is hard to doubt that
somewhere behind both lies a single event. In both Paul accompanied
by Barnabas goes up to Jerusalem. In Acts ‘certain others’ go with
them; in Galatians Titus goes, and it is not said that he is the only
additional traveller. In Galatians James, Cephas, and John take part;
in Acts James and Peter take part and there is nothing to suggest that
John was not there as one of the unnamed apostles. In Galatians Paul
and Barnabas are accepted by the Jerusalem apostles as colleagues;
in Acts they take part in the discussion and are described in the letter
as ‘our beloved Barnabas and Paul’. In Acts the theme discussed in
the Council is the proposition that all converts to Christianity must be
circumcised; if the Galatians Council did not deal with this theme it
was irrelevant to the Galatian situation to which Paul applies it. This
is a weighty list of parallels. On the other hand, the Acts Council
ends with the issue of a Decree, binding on Christians; in Galatians
there is no sign of this Decree, nor is there any sign of it anywhere
else in Paul’s letters. In Galatians there is some sort of division of
apostolic labour (Gal. 2.9); there is no trace of this in Acts. It would
be difficult to maintain (though not a few have done so) that Acts and
Paul are describing different events; equally difficult to think that
Luke had read Galatians.
We may add, and take as a pointer to the unravelling of the
historical problem, that both Acts and Galatians speak of the arrival
in Antioch of representatives from Jerusalem. In Acts they take a
hard line on circumcision and the Law: all must be circumcised and
observe the Law. This demand is represented as the immediate
occasion of the Council. In Galatians (assuming, as without indica-
tion to the contrary one should, that Paul sets out events in
chronological order) the visit is of envoys from James, and it follows
the Council. As a result of it the church in Antioch is divided: Peter,
followed by Barnabas and all the Jewish Christians, withdrew into
isolation, refusing to eat with uncircumcised Gentile Christians. Paul
maintained his position and stood by the Gentiles, rebuking Peter to
his face. Assuming Paul to have been neither stupid nor dishonest,
though doubtless like all men fallible in memory, his account, which
is first-hand, must be accepted where it differs from Luke’s; and the
likeliest explanation of the course of events is as follows.
A mission based on Jerusalem was circulating among the Pauline
churches (cf. Μ. D. Goulder, A Tale of Two Missions, London,
1994). Circumcision of all was demanded. This, if unchecked, could
have ruined Paul’s work, and he went up to Jerusalem that he might
712 COMMENTARY ON ACTS

not run in vain (Gal. 2.2). Apart from an encounter with false
brothers (who are distinguished from the apostles) the meeting was
not inharmonious. There was agreement that Paul should go to the
Gentiles (to evangelize them), the Jerusalem apostles to the Jews.
This sounded well but was inadequately thought out and the terms
remained undefined. It probably was related to the question whether
an attempt should first be made to win the Jews for the Gospel,
leaving the Gentile mission as a second-stage operation. It did not
take into account what might be done in churches with mixed Jewish
and Gentile membership. In Antioch it was at first assumed that since
all were Christians all might have table fellowship together. The
messengers from James (Gal. 2.12) must have said something like,
We agreed that Gentiles might be accepted as Christians without
circumcision; we did not say that Jews might so far cease to be Jews
as to have unrestricted dealings with Gentiles. The two leaders,
James and Paul, dug in their heels, and it was the Hellenistic Jewish
Christians who found a way out by proposing the Decree—which
Paul did not accept.
The theme of Luke’s narrative does not change (see p. 745); the
point of contention between Paul and Jerusalem changed. It was
accepted that Gentile Christians did not have to be circumcised; it
was not accepted by Jerusalem that there might be table fellowship
between Jewish Christians and Gentile Christians. It is clear that the
mission that required the circumcision, that is, the full proselytiza-
tion, of Gentiles continued; it was certainly active in Galatia, but it
no longer had the full support (though it may have had the implicit
encouragement) of Jerusalem.
For the dating of the Council see Introduction, pp. Ivi-lxi. Taylor
(5.208) holds that the Council took place towards the end of the reign
of Agrippa I, since it is clear from both Acts and Galatians that Peter
(cf. 12.17) was still in Jerusalem. The argument is fallacious; 12.17
does not necessarily mean that Peter moved out of Jerusalem, and if
he did he might easily have returned, especially after the death of
Agrippa.

6. The ἐκκλησία of 15.4 now disappears (to reappear in v. 22 and

possibly, as το πλήθος, in v. 12); the serious business of the Council
is to be done by apostles and elders. 614 pc syh add σὺν τῷ πλήθει; it
may be that the Western text did not lay quite so much stress on the
importance of the apostles as is sometimes supposed. Of the apostles
mentioned in ch. 1 only Peter speaks; Barnabas and Paul, apostles
only at 14.4, 14, report on their work; James, who seems to be
merged with the apostolic group, proposes a solution which is
accepted. It is very probable that Luke is here constructing a scene
out of a small quantity of material. Elders (presbyters) probably
represent the church officers with whom he was himself familiar. At
 39. COUNCIL IN JERUSALEM. 15.6-29. 713

13.1 the church at Antioch has prophets and teachers; at that point
Luke was probably using an Antiochene source. Here he assumes the
church order that he knows, adding in the apostles as, while they
lived, the highest authority. It may be said that Luke, though he
quotes only Peter, regarded them as the real decision makers (cf.
Peter, James, and John in Galatians 2); the elders listen and agree.
ἰδεῖν περί is an unusual expression, though it must mean some-
thing like look into. J. L. North (NTS 29 (1983), 264-6) points out
that the Latin videre de suggests a legal, judicial sense. Is it intended
to suggest (cf. 15.2 and especially the Western text) that Paul and
Barnabas are on trial? More probably we should think of something
like a rabbinic, or Qumran, court settling a matter of halakah. Cf. B.
Gerhardsson (above, p. 709). It is doubtful whether Luke knew much
about such courts, and it must be borne in mind that groups of
people, assembled to discuss courses of action, are all likely to
behave in similar ways, whatever their backgrounds.
For λόγου, E 614 pc gig syh have ζητήματος, by assimilation to
15.2.

7. πολλῆς δέ ζητήσεως γενομένης. There was much discussion.

For ζήτησις see 15.2; it means search (ζητεῖν) for truth through
public inquiry and debate, and it will shortly appear that the leading
figures are agreed. Such dispute as is implied must (in Luke’s view)
come from those mentioned in 15.5.
At this point p45(v,d) makes substantially the same addition to the
text as is made by D in 15.2. Of 15.2, P45 contains only ζητήματος
τούτου, so that it is impossible to tell whether it contained the words
there also. They are better restored in the apparatus of NA26 than in
F. G. Kenyon’s editto princeps (1933).
Peter rose up to speak, ἀναστάς; cf. ἐξανέστησεν in 15.5: biblical
style rather than a linguistic Semitism. D* (614 syhmg) have
ἀνέστησεν τῷ Πνεύματι Πέτρος καί: Peter is inspired; his pro-
nouncement comes from God. Black (AA 69) notes the introduction
of parataxis (ἀνέστησεν καί εἶπεν). Peter appeals to common
knowledge: ὑμεῖς έπίστασθε—you know for yourselves, without
my telling you. This applies to the reader, who will recall the
contents of ch. 10, but there is no reason why some of Peter’s
activities should not have been generally known.
Peter’s first sentence contains several obscurities, ἀφ’ ἡμερών
ἀρχαίων should introduce the assertion that something has been
going on continuously since an early time; one would expect an
imperfect, but we have in fact the aorist ἐξελέξατο, which calls for
the definition of a point of time. We must understand the preposi-
tional phrase to mean not from but in ancient days, in days of old.
Hebrew can occasionally be taken in this way (Gen. 6.4; Josh.
24.2; Ps. 77.6), but this is of little help. Preuschen (94) takes the
714 COMMENTARY ON ACTS

expression to be a stronger equivalent to πρότερον. But what are the

days of old? Cf. v. 14 (πρῶτον); also 11.15 (ἐν ἀρχῇ), and v. 21,
though there ἐκ γενεῶν αρχαίων introduces a continuous state (έχει).
The meaning seems to be that from the start (of the Christian
movement—in v. 15 the intention will be traced back to the
prophets) God’s plan included Gentiles. Whether those who were
involved in the movement at the beginning were at the time aware of
this is doubtful. There can hardly, in view of the dispute that arose,
have been a specific instruction on the matter from Jesus, and in the
absence of such an instruction those who questioned the step can
hardly be blamed for doing so. Peter claims that he was the first to
understand what should be done.
ἐν ὑμῖν ἐξελέξατο ό θεός (P74 A B C 33 36 81 945 1175 1739 al
bo Irlat) gal
riations in order) ὁ θεός ἐν ήμῖν ἐξελέξατο; D* 614 pc have
ἡμῖν ό θεός ἐξελέξατο; 189 pc syp sa have ό θεός ἐξελέξατο. The
text of Β is almost certainly correct, but it is difficult in that
ἐξελέξατο has no direct object and that if the reference is to the
Cornelius story (as it must be) Peter might have been expected to say
that God made his choice (not among you but) among us (as in the
reading of Dc), since he himself was the one chosen. The difficulty is
not to be dealt with by the observation that the Greek ἐξελέξατο
could render the Hebrew whose direct object is often introduced
by the preposition sometimes translated in die LXX by ἐν (Bengel
ad loc. compares 1 Chron. 28.4, 5). If this were accepted we should
have to translate God chose you, that is, God chose the early Jewish
Christian body (in the event represented by Peter) as his means of
reaching the Gentiles. Wilcox (92f.) takes the Greek in this way and
remarks on the Semitism (probably Hebrew, possibly Aramaic).
There is however no reason to think that Luke is here translating with
verbal literalness a Hebrew source; see below on vv. 16-18 where
James quotes the LXX in a passage where it differs from the Hebrew.
Peter is speaking of the divine choice that gave him a special task,
and it is better to take the object of ἐξελέξατο to be the accusative
and infinitive clause ἀκοῦσαι τά έθνη ... και πιστεῦσαι; ἐν ὑμῖν
will mean that God chose among you (Christians, or perhaps
apostles), that is, to the exclusion of the rest of you, that the Gentiles
should hear and believe the Gospel through my mouth. Johnson
(261) takes τά έθνη to be the direct object of ἐξελέξατο: God chose
the Gentiles to hear the message ... This is not impossible, and gives
an interesting oxymoron: God chose the un-chosen people... But the
order of words makes this seem improbable.
In this sentence the first phrase (in Greek) is emphatic. Peter has a
special right to be heard in the debate because, as all are aware, it was
through him that the word of the Gospel (for this expression there is
no parallel in Acts, or elsewhere in the NT, but cf. Col. 1.5; in Acts
 39. COUNCIL IN JERUSALEM. 15.6-29. 715

εὐαγγέλιον occurs only here and at 20.24) was first communicated to

the Gentiles. The reference is evidently to the story of Cornelius,
which is regarded as preceding not only the preaching to the
Έλληνισταί in 11.20 but also Philip’s preaching to the Samaritans
and to the Ethiopian in ch. 8. It is hard to know whether this claim
rests upon Luke’s own understanding of the history or is based upon
a memory or tradition of what Peter was believed to have said,
whether at the Council or on some other occasion. For the editorial
processes that underlie Luke’s narrative see pp. 710-12.
What had been required of the Gentiles was that they should hear
and believe the Gospel, not that they should be circumcised or
commit themselves to observance of the Law.
The verse affirms the absolute priority if not the primacy of Peter
in the Gentile mission. One might be inclined to suspect an anti-
Pauline interest here, were it not that Luke goes on to fill the rest of
his book with material that depicts Paul as the leader in this work,
whereas Peter, after this assertion of God’s choice of him, dis-
appears. It may be that the motive is ‘not only but also’ ; Paul may
have been the greatest but Peter was the first. The two must not be set
over against each other. This would accord with the eirenic interest
of Luke (displayed not least in this chapter), as also perhaps would
his silence regarding the rest of Peter’s career, of which some
aspects, from a Pauline point of view, were scarcely creditable (Gal.
2.11-18). Luke, through Peter, makes clear that the law-free Gentile
mission originated within the apostolic tradition (Schmithals 137).
Peter speaks first; it is not so clear as some (e.g. Roloff 230) think
that James was now head of the Jerusalem church.

8. For καρδιογνώστης see on 1.24; there the term is applied

(probably) to Christ, here to God. Luke would find no difficulty in
the double use of the term. God knew what was in the hearts of
Cornelius and his friends and bore witness to them. μαρτυρεῖν is
used, especially in the passive, as a word describing a person’s
reputation; cf. e.g. 16.2, Timothy had a good reputation with the
brothers. This, in view of the context and the participial clause that
follows, though on the right lines will hardly suffice here. If however
we ask in respect of what God bore witness to Cornelius the answer
is not clear. It might be to the fact that he feared God and practised
righteousness, and was thus acceptable to God (10.35); it might be to
the fact that he was, though a Gentile, one of the elect; it might be
(looking at the matter from the other side) to the fact that he was
about to exercise, was already beginning to exercise, faith. The last is
probably nearest to Luke’s thought. God bore witness to the fact that
Cornelius had fulfilled all the conditions (hearing and believing the
Word) necessary for being a Christian, and was thus qualified to
receive the Holy Spirit.
716 COMMENTARY ON ACTS

The participle δούς (C E (Ψ) <a> 1 Irlat add αὐτοῖς, D adds

ἐπ’αὐτούς) is adverbial; God bore witness to them in that he gave...
The gift was the visible sign of his approval, or rather the audible
sign (10.46): Luke regards inspired speech as the sure sign of the
work of the Spirit. It was a sign that placed Cornelius and his
company on the same level as Peter and his apostolic colleagues. It
was the gift of the Spirit that constituted the apostolic body at the
beginning (2.4), and it was manifested in speaking with tongues. The
Holy Spirit was given to the Gentiles in Caesarea (10.44) and was
described as τὴν ἴσην δωρεάν (11.17). God thus treated them exactly
as he had treated Peter and the Eleven (καθώς καί ήμῖν).
Speaking with tongues is not the best sign of the Spirit’s presence;
so Calvin (2.33): ‘It seems that the evidence of election, that the Holy
Spirit fell on them, is not altogether convincing (minus firmum).’

9. God gave them the same gift that he gave us (v. 8); that is,
οὐθέν διέκρινεν μεταξύ ἡμών τε καί αυτών, he made no distinction
between us and them, treating both groups, circumcised Jews and
uncircumcised Gentiles, alike. Since evidently God did not see fit to
require circumcision before bestowing his gifts it was not for men to
demand it.
For διακρίνειν see 11.12 (for διακρίνεσθαι, 10.20). With God as
the subject and the verb in the active there can be no doubt about the
meaning. P74, διεκρίναμεν, would have something to commend it
were it not that the finite verb is followed by a singular participle,
καθαρίσας, which makes the plural impossible.
For οὐθέν (over against ούδέν) see Thackeray (58-62) and Μ.
2.111f. (also BDR § 33.2, n. 2 and LS 1269, s.v. ούθείς). ‘First found
in an inscription of 378 BC, it [οὐθείς] is practically the only form in
use throughout the Greek-speaking world during iii/B.C. and the first
half of ii/B.C. In 132 B.C. the δ forms begin again to reassert
themselves, and the period from that date to about 100 B.C. appears
to have been one of transition, when the δ and θ forms are found side
by side in the same documents. For i/B.C. we are in the dark, but in
i/A.D: we find that οὐδείς has completely regained its ascendancy’
(Thackeray 58). In the NT forms with θ occur at Lk 22.35; 23.14;
Acts 15.9; 19.27; 26.26; 27.33 (μηθέν); 1 Cor. 13.2; thus almost
always in Luke-Acts, and never without textual variation. According
to Moulton (loc. cit.), ‘There has been a re-formation οὐδ’ εἷς, with δ
+ h producing θ.’ There is an inconclusive note in ND 4.16f. There
seems to be no specific reason why Luke should have chosen to use
this form of the word at this place, though it may mean that he drew
Peter’s words from a source. It is however impossible to regard all
the passages cited above as derived from a single source; they are not
connected with one another.
God’s decision to make no distinction between Jew and Gentile
 39. COUNCIL IN JERUSALEM. 15.6-29. 717

was expressed in the act (καθαρίσας, aorist) of cleansing their

hearts. There is no parallel to this in Acts. The adjective καθαρός
occurs at 18.6; 20.26, each time with reference to Paul’s innocence in
his relations with others—Jews in Corinth, Christians in Ephesus.
καθαρίζειν occurs in the Cornelius story (10.15) and its repetition
(11.9), in the injunction with which Peter’s vision closes: What God
has cleansed (ἐκαθάρισεν), do not you count common. Here, though
the reference is to the Cornelius story, the meaning appears to be
different. In the story (as Luke tells it—for the question whether the
vision and the visit to Cornelius originally belonged together see
I.493f., 496f.) the cleansing is presumably anterior to Peter’s entry
into Cornelius’s house: ‘You must not consider Cornelius and his
house unclean; God has cleansed them.’ Here in ch. 15 the cleansing
takes place in view of the faith generated in Cornelius as he hears the
word: τῇ πίστει (though Knowling (319) takes it to refer to the
(Christian) faith) in v. 9 must be interpreted in terms of πιστεῦοαι in
v. 7. It is by faith (not by circumcision or other legal works) that
God’s gift of the Spirit and of cleansing is received. The distinction is
real but it is doubtful whether it would occur to Luke. Cornelius had
been admitted to the cleansed people of God, and without legal
requirement; that was all that mattered.
In the story, the mark of Cornelius’s acceptance was baptism
(10.47f.; cf. 11.17). Baptism was a bath and could therefore be
associated with cleansing; cf. 22.16. This is not however an image
that Luke regularly uses (for him baptism is primarily a rite of
initiation), and the cleansing of the heart probably means for him the
forgiveness of sins (cf. 13.38f.) and inward renewal with a view to
future obedience. Baptism is not viewed as the Christian replacement
of circumcision.
The notion of inward cleansing is known in antiquity, also in
circles that are not notably religious; thus Lucretius 6.23, Veridicis
igitur purgavit pectora dictis; Xenophon, Symposium 1.4, ...
ἀνδράσιν ἐκκεκαθαρμένοις τάς ψυχάς.

10. νυν ούν begins to bring the brief speech to its conclusion with
a rhetorical question that leads to a reductio ad absurdum. Why do
you ... ? means You ought not to... It is absurd to expect Gentiles to
put up with what we Jews cannot endure.
τί πειράζετε τον θεόν; The thought appears to be that God has
already, by his action, made clear that it is his intention to incorpo-
rate the Gentiles as they are—without circumcision or any other
legal observance—in his people, and that to hinder their reception by
legal stipulation is wantonly to provoke him. For this use of
πειράζειν cf. e.g. Exod. 17.2, τί λοιδορεῖσθέ μοι, καί τί πειράζετε
κύριον; Men are seeing ‘how far they can go’ with God, insisting on
that which is plainly contrary to his will. In this case their action
718 COMMENTARY ON ACTS

would be ‘eine Herausforderung Gottes, der durch die in Jerusalem

bekannte Offenbarung im Hause des Cornelius die Gesetzesfreiheit
der Heidenchristen als seinem Willen entsprechend erwiesen hatte’
(H. Seesemann in TWNT 6.33). This use of πειράζειν is almost
exclusively biblical, but BA (1291), who understand it in a slightly
different way, quote Herodotus 6.86.3 (6.86.γ.2 in OCT), ή δέ Πυθίη
ἔφη τό πειρηθῆναι του θεοῦ καί τό ποιήσαι ἴσον δύνασθαι. Cf.
Chrysostom (Homily 32.1): τί πειράζετε τόν θεόν ὡς οὐκ ἰσχύοντα
σῶσαι τῇ πίστει; ἄρα απιστίας ἐστὶ τὸ τὸν νόμον εἰσφέρειν.
The construction continues with the infinitive, ἐπιθεῖναι, which
Bruce (1.293) describes as epexegetic, and Zerwick (§ 392) explains
as equivalent to a gerundive: ‘Quid tentatis Deum ἐπιθεῖναι ζυγόν =
imponendo iugum super cervices...’. Μ. 3.136 follows this; cf. BDR
§ 391.4, n. 8. The Vulgate, translating literally, has imponere. τῶν
here retains something of its demonstrative force; these (Gentile)
disciples.
Rabbinic literature uses the word yoke, but not as it is used
here. See e.g. Aboth 3.5, He that takes upon himself the yoke of the
Law from him shall be taken away the yoke of the
kingdom (that is, oppression by worldly authority). Thus
the yoke of the Law, the obligation to obey it, is a blessing and a
privilege. Peter takes it otherwise. In other passages of the NT ζυγός
has a bad sense. At 1 Tim. 6.1 slaves are described as being υπό
ζυγόν, and the image is used theologically at Gal. 5.1 (μή πάλιν
ζυγῷ δουλείας ἐνέχεσθε); cf. Barnabas 2.6: the new law (ὁ καινός
νόμος) of Christ is ἄνευ ζυγού ανάγκης; but cf. also 1 Clement
16.17: We have come υπό τόν ζυγόν τής χάριτος αυτού, which
recalls Mt. 11.29f.
Peter claims that to require circumcision and legal observance
from Gentile converts would be to lay upon their necks a yoke ὄν
οὔτε οἱ πατέρες ἡμών ούτε ἡμεῖς ἰσχύσαμεν βαστάσαι, to demand
of them something that Jews themselves could neither endure nor
achieve. Cf. Mt. 23.4. This does not agree with the attitude to the
yoke of Torah quoted above—it was a privilege and joy; and Paul
claims to have fulfilled the Law completely (Phil. 3.6). He was also
well aware of Jewish Christians who continued to feel and to respond
to its attraction. Conzelmann (83), noting these facts, argues that
Luke was writing about something that he did not himself know at
first hand. The controversy, when he wrote, lay in the past, and he did
not fully understand the point at issue. Only a minimum of legal
requirement now remained for any Christian, and Luke was unable to
believe that there ever were Christians who loved the Law, and some
who wished to impose it on Gentiles. If Peter really felt like this
about the Law—that it was an intolerable burden—why did not he
and his fellow Jewish Christians themselves give it up? Indeed, if
Jews in general (oἱ πατέρες ἡμῶν, not only the Christians but earlier
 39. COUNCIL IN JERUSALEM. 15.6-29. 719

generations) found the Law insupportable how did it survive, and

why was it painstakingly developed and interpreted in halakah and
haggadah? Haenchen (429) similarly thinks that Peter is made to give
the views not of his own but of Luke’s time and environment; so also
e.g. Weiser (381). This conclusion may be correct, but it is certainly
an oversimplification. Paul, who had observed the Law, was a
Pharisee (Phil. 3.5; cf. Acts 23.6; also 22.3), as were those who have
given us the rabbinic literature. We have no reason to think that Peter
was a Pharisee; he may well have been an 'am ha- 'ares, and was a
Galilean into the bargain. Perhaps he was one of those Jews who
made no serious attempt to keep the Law. His opinion may reflect
also the views of liberal Hellenistic Jews, who, though they are to be
distinguished from Paul, contributed to the Gentile mission. Accord-
ing to Schille (320), Peter represents ‘die Anschauung der hellen-
istischen Kirche’. This again does not say all that is to be said.
According to 10.14 and 11.8 Peter claimed that he had never touched
unclean food (with the probability that one who was scrupulous in
this respect would not be negligent in others). It is in agreement with
this, or at least not in disagreement, that according to Gal. 2.12 he
withdrew himself from table-fellowship with Gentiles. It is impos-
sible to save the historicity of this speech (or at any rate this part of
it) and Peter’s reaction to the animal vision in ch. 10, even if some
variability in Peter’s character (attested by Gal. 2.12-15 as well as in
the gospels) is taken into account. He is described here as something
of a Paulinist, though the Paulinism is not accurately portrayed. That
Luke should describe him in this way must cast some doubt on the
view that towards the end of the century there was an influential
minority of Jewish Christians.
This is the only place in Luke-Acts where βαστάζειν has the
meaning endure, put up with (see J. L. Nolland in NTS 27 (1980-1),
105-115), but ισχύειν is a Lucan word: Lk. 8 times; Acts 6, rest of
the NT 14.

11. ἀλλά implies, ‘No, indeed, we could not do this, but on the
contrary ...’.
The language of the verse is superficially Pauline (see on v. 10) but
lacks Paul’s precision. ‘Paulinische Begriffe werden von Lukas
übernommen, aber nicht die paulinische Theologie, welche sich
dieser Begriffe bedient’ (Haenchen 429). The grace of the Lord Jesus
(C D Ψ 33 36 453 945 1175 1739 pc it syp bopt Irlat add Χριστού)
recalls 2 Cor. 8.9, but receives no definition. The general sense
however is clear. Whatever precisely Luke may have understood by
χάρις it did not mean works of the Law. And διά τής χάριτος stands
emphatically at the beginning of the sentence, though it is probably
to be connected with σωθῆναι rather than with πιστεύομεν (the third
time a word of this root appears in this short speech); the latter
720 COMMENTARY ON ACTS

connection however is not impossible. Cf. 13.48; faith itself is of the

Lord’s gracious appointment.
It is to be observed (see the note on 15.1 and p. 696) that in this
verse the theme of the Council is still (not rules for table fellowship
but) salvation and the conditions on which salvation may be
received. ‘De Salute agebatur’ (Bengel 449). The tense of οωθῆναι
does not determine the time at which salvation is thought to take
place. Past, present, and future are all possible: We believe that we
have been saved, that we shall be saved, or even as a general
proposition, We believe that we are saved. It is however probable
that Luke here uses a construction whereby ‘Verbs and expressions
signifying to hope, to expect, to promise, and the like, after which the
Infinitive in indirect discourse would naturally be in the Future... as
representing a Future Indicative of the direct discourse, sometimes
take the Aorist (as well as the Present) Infinitive’ (Goodwin, Moods
and Tenses § 23.2). This however, though probable, is not certain,
because the construction normally applies only in cases where the
main verb itself points clearly to the fixture, and πιστεύειν does not
necessarily do this. See Nolland, as on v. 10.
In καθ’ ον τρόπον κἀκεῖνοι, ἐκεῖνοι is most naturally taken to
refer to those (the Gentile believers) to whom the Holy Spirit has
already been said to have been given καθώς καί ἡμῖν. The parallel-
ism between them and us would then be expressed in two ways: the
Holy Spirit was given to them as to us; salvation comes to us as to
them, by grace through faith. This is probably what Luke intends, but
the grammatically nearest antecedent to εκείνοι is οἱ πατέρες ημών;
if this is regarded as determinative, the point would be that the saints
of the OT, prophets and the like, were saved by faith, and we are, or
shall be, saved in the same way, apart from legal works. This was the
view of Augustine (Contra duas Epistolas Pelagianorum 1.21(39),
... ut eadem gratia Jesu Christi salvi facti credantur antiqui), Calvin
(2.42), and Bengel (449). On either view the parallel is to be found
neither in believed nor in saved alone but in the combination of the
two.
According to Pesch 2.78 this passage shows the fundamental
agreement between Peter and Paul (cf. Gal. 2.15f.). It does so only if
the opinions expressed here are correctly ascribed to Peter; it is
certainly true that Luke makes Peter speak—more or less—like
Paul. Bruce (1.295) quotes the opinion of Ramsay (Paul the Travel-
ler 164) who looks at the matter from a different angle and uses it to
prove that Gal. 2.11f. cannot have happened after the Council
because it would make Peter guilty of a ‘meaningless tergiversation’.
This is quite unconvincing, for (a) tergiversation is precisely the
accusation Paul levels against Peter, and (b) whatever the relation
between Gal. 2.11f. and the Council the event in Antioch must have
happened after the Cornelius episode.
 39. COUNCIL IN JERUSALEM. 15.6-29. 721

Peter (as represented by Luke) concludes his argument that by his

actions God himself has made clear that Gentiles who believe are
acceptable to him, and that he is prepared to bestow his gifts upon
them without any condition in addition to their faith. If this is so, who
are even apostles to impose conditions when God requires none?
This is as far as Peter goes; he, and Luke, do not see that it is implied
(Roloff 231) that Jewish Christians also have given up the Law as a
means of salvation. It is fair to add, with Marshall (250), ‘What Peter
disputed was thus the need to obey the law in order to be saved;
whether Jews kept it for other reasons was a secondary matter.’

12. Έσίγησεν δέ πᾶν τό πλήθος. Luke presumably means that

Peter’s speech silenced the ζήτησις of v. 7. The πλήθος is probably
the mass of the people (cf. the use of ἐκκλησία in 15.4,22) as distinct
from the apostles and elders, rather than the whole company of
apostles and elders. The distinction is made clear in the reading of D
(1) syh**, συγκατατιθεμένων δέ τῶν πρεσβυτέρων τοῖς ύπό τοῦ
Πέτρου εἰρημένοις ἐσίγησεν πᾶν τό πλήθος. Luke could mean
simply that the company were silenced, but probably intends to
suggest that they were convinced and agreed. See the long quotation
from Musonius Rufus given by P. W. van der Horst in NovT 16
(1974), 309.
With ἤκουον Luke forgets that he has used the singular ἐσίγησεν
with a noun of multitude, πλήθος; or rather, picturing the occasion,
thinks of a general silence, and of listening individuals, each intent
on the story that Barnabas and Paul had to tell.
With ἐξηγουμένων cf. ἐκδιηγούμενοι in 15.3. There is no sig-
nificant difference in meaning, ἐξηγ. is more frequently used in Acts
(10.8; 15.12, 14; 21.19, always in the sense of narrating events) than
ἐκδιηγ. (13.41, quoting Hab. 1.5; 15.3), and, it seems, in Greek
literature generally. In 15.3 Paul and Barnabas recount the conver-
sion of the Gentiles, in the present verse the signs and portents that
God had done among the Gentiles. Luke is probably thinking of
essentially the same account (cf. 14.27), but the occurrence of signs
and portents gives it greater probative value since it shows God’s
approval. No details are given; Luke expects the narrative of chs. 13
and 14 to be in his readers’ minds, though these chapters do not
contain many signs and portents, σημεία καί τέρατα is one of
Luke’s standard phrases (2.19, 22, 43; 4.30; 5.12; 6.8; 7.36; 14.3)
and it slips in; he is responsible for the verse in its present form. The
point here is that the signs and portents were done ἐν τοῖς ἕθνεσιν.
Haenchen (121) makes the point that since Luke lacks Paul’s
theological justification of the Gentile mission he is obliged to fall
back on miracles. In fact, Luke has several ways of justifying the
mission—Peter’s argument that God has treated Jews and Gentiles
alike, and James’s argument (see below) from Scripture; it is
722 COMMENTARY ON ACTS

paradoxical that Paul should be left with miracles. Cf. however Rom.
15.19; 2 Cor. 12.12; Gal. 3.5. Stählin (203) thinks that the miracles
were ‘eine wichtige Form der Verkündigung’; Luke might have
agreed, but it is doubtful whether Paul would have done so. The most
that can be said is ‘dass Gott zu dem Missionswerke des Paulus sein
Ja gesprochen hat' (Bauemfeind 191).
Bultmann (Exegetica 417 = FS T. W. Manson 72f. + n. 6) thinks
that v. 12, like all the other references to Paul in 15.1-35, was
inserted into a source that did not originally contain it; Paul and
Barnabas were not present at the Jerusalem meeting that issued the
Decree of 15.29. On this question see p. 710 and below; there is no
difficulty in conjecturing a source that read ἐσίγησεν δέ πᾶν τὸ
πλήθος, καί ἀπεκρίθη Ιάκωβος. ‘Die Sätze, in denen von Paulus
und Barnabas die Rede ist, lassen sich leicht herausheben: V. 2 (es
würde genügen: ἔταξαν ἀναβαίνειν τινάς ἐξ αυτῶν πρὸς κτλ.),
3-5, 12; das σὺν τῷ Παύλῳ καί Βαρναβᾷ in V. 22; V. 25f.—Vgl.
W. Bousset, ZNW 14 (1913), S. 156-162’ (Bultmann). On the other
hand Pesch (2.79) thinks that the order Barnabas and Paul does not
suggest the hand of an interpolator. If Luke did insert the references
to Paul and Barnabas this could have been on one or both of two
grounds. (a) He knew a tradition which (correctly) recorded that the
two had taken part in a conference at Jerusalem (cf. Gal 2.1-10), and
(b) he needed in his record of the Council representatives of those
whom he describes as Hellenists (6.1; see the note). He seems to
have regarded Paul and Barnabas as representatives of the movement
that began with, or was notably represented by, Stephen.

13. μετά δέ τό σιγῆσαι αυτούς may be part of an addition made by

Luke. For this possibility—it is no more—see on v. 12. ἐσίγησεν
had been used in the introduction of Barnabas and Paul; James had
now to be introduced, and the verb σιγάν was in Luke’s mind and
reused.
The introduction of James is slightly more formal in D syp:
ἀναστάς (cf. v. 7) Ιάκωβος εἶπεν. Luke does not tell us who is
intended by the name James. Presumably he is the James of
12.17—evidently a person of such consequence that he needs no
description; Luke’s readers are sure to know who is meant. It is no
doubt he who appears again at 21.18, again as a leading figure among
the Jerusalem Christians. A James is mentioned among the brothers
of Jesus (Mk 6.3), and as a witness of the resurrection (1 Cor 15.7). It
is very probably the same James who plays an active role in
Galatians: he was the only notable figure, in addition to Peter, seen
by Paul on his first visit to Jerusalem (Gal 1.19); he was one of those
who bore the title στύλοι (2.9—with Cephas and John) but made no
contribution to Paul’s Gospel or his authority (2.6), rather recogniz-
ing his commission (2.7, 8). It was persons who came from James
 39. COUNCIL IN JERUSALEM. 15.6-29. 723

who induced Peter to change his attitude in Antioch (2.12). The only
other material worth taking seriously is provided by the accounts of
the death of the Lord’s brother in Josephus, Ant. 20.200, and in
Hegesippus, quoted by Eusebius, HE 2.23.4-18. The latter especially
suggests that the martyr had a great reputation for Jewish piety.
Whether, and to what extent, we can also use the Epistle of James is a
disputed question. On all this material see W. Pratscher, Der
Herrenbruder Jakobus und die Jakobustradition (Göttingen, 1987);
also Benoit (2.289, cf. 254). James seems at this point to be in
agreement with Peter, Paul, and Barnabas. Whether the evidence of
21.18 and of Galatians confirms or contradicts this is a question that
cannot be avoided.
ἄνδρες αδελφοί. For this mode of address cf. 1.16. Sometimes it
means Fellow Christians, sometimes Fellow Jews; here it could mean
either—or both.
ἀκούσατέ μου. According to Roloff (231) this address is a mark
of authority; cf. Stählin 203. This is not necessarily so; cf. 2.22; 7.2;
13.16; 22.1. According to W. Dieterich (Das Petrusbild der luka-
nischen Schriften (Stuttgart, 1972), 317), James’s is an Erwider-
ungsrede: it was not Peter who initiated the mission but God himself.
This seems however to be what Peter himself says (ἐξελέξατο ὁ
θεός, v. 7).

14. Συμεών. ‘Jacobus, Hebraeorum apostolus, Hebraico nomine

Petrum appellat’ (Bengel 449). Cf. 2 Pet. 1.1; this form, which recalls
the Hebrew was probably intended to give the passage a
Semitic air, regarded as suitable for James. Simon Peter, who had
just spoken, is intended; notwithstanding Fitzmyer (Essays 108), it is
mistaken to see here a reference to Symeon Niger (13.1). Cf.
Cullmann (Petrus 18). James refers to Peter, not to Barnabas and
Paul; this does not prove but is at least consistent with the suggestion
that the original source did not contain v. 12; see on that verse.
ἐξηγήσατο. Cf. ἐκδιηγούμενοι (15.3), ἐξηγουμένων (v. 12). The
repetition is striking. The word must surely have the same meaning
here as in v. 12 and will therefore mean that Peter recorded an event;
ἐπεσκέψατο will therefore refer to the conversion of Cornelius
(which Peter clearly has in mind in vv. 7-9).
καθώς has here the very unusual (but see Ep. Aristeas 263; 3 Jn 3)
sense, how: Symeon has narrated how God ...
With πρώτον cf. ἀφ’ ἡμερῶν αρχαίων (v. 7); the event in mind
belongs to the beginnings of the Christian story, and was the first of
its kind. It is not easy to find a good rendering of ἐπεσκέψατο. Page
(177) says that its meaning is not visit but showed regard or
consideration in taking. Begs. 4.175 says that the meaning is to make
provision for. There is in fact much to be said for the meaning
not uncommon in the LXX and Lk. (see 1.68, 78; 7.16), visit;
724 COMMENTARY ON ACTS

unfortunately it is not easy in English to combine this with the

following infinitive. One may suggest the English take action; cf.
Exod. 3.16, where God takes decisive, liberating, action on behalf of
the persons concerned. In this case he so acted in order to take
(understanding λαβεῖν as an infinitive of purpose) out of the Gentiles
a people for his name. In ἐξ ἐθνῶν λαόν the two nouns stand in sharp
contrast with each other. Out of those who are by definition not Jews,
who are not the people of God, God takes men (in the first instance,
Cornelius and his friends—but on this see below) to be a λαός, to be
his own people, even though they lack what has hitherto been the
necessary qualification for this: they do not belong to the right race
and have not been circumcised as proselytes. J. Dupont (NTS 3 (1957),
47-50 = Études 361-5, here with an additional note referring to the
article by N. A. Dahl, mentioned below) draws attention to the fact
that James’s words recall the language of the LXX, especially of Deut.
14.2 (σε ἐξελέξατο κύριος ὁ θεός σου γενέσθαι σε αὐτῷ λαόν
περιούσιον ἀπό πάντων των ἐθνῶν; cf. Deut. 7.6; Exod. 19.5;
23.22). In a different context, James could simply be referring to the
distant past when God chose Israel to be his special people; as we
have seen, he is taking up Peter’s reference to the Cornelius story,
which his language interprets as parallel to the original call made on
an ethnic basis. A further point is that the LXX distinguishes the two
nouns λαός and ἔθνη whereas in the Hebrew there is no distinction
Thus the dependence on the LXX that
will be noted in vv. 16-18 occurs already in v. 14; James speaks
throughout as a Hellenistic Jew dependent on the LXX.
The article by Dupont was taken up by N. A. Dahl (NTS 4 (1958),
319-27), who pointed out that λαός τω ὀνόματι αυτού does not
occur in the LXX, but that ‘the phrase “people for His (My, the
Lord’s) name” is a standard idiom in the old Palestinian Targum,
where it is regularly used to render the Hebrew
(321). He considers that the most interesting parallel to Acts 15.14 is
Zech. 2.15(11), ‘where it is stated that even Gentiles shall become a
people of the Lord’ (323). It would be unwise to build much on τω
ὀνόματι αυτού; it recalls the baptismal phrase εις τό όνομα αυτού
(see 8.16), and like it probably means ‘to his account’, ‘so as to
become his property’. Conzelmann (83) may well be right to
question the specific biblical allusions; the words may be due
to ‘lukanischer Bibelstil’. See also Wilson (Gentiles 224).

15. This verse shows that though, as Dupont correctly pointed out,
the language of v. 14 echoes that of the LXX, it would be mistaken to
lay much stress on the OT background of the verse. Verse 14 states a
historical fact, in words that hint at a much earlier historical fact: God
acted so as to take to himself a people from among the Gentiles; he
did this in the conversion and acceptance of Cornelius. Verse 15
 39. COUNCIL IN JERUSALEM. 15.6-29. 725

makes the further point that what God then did fulfilled what he had
through the prophets promised to do. With this (τούτῳ will refer to
the whole action described in the preceding verse) the words of the
prophets are in agreement. For συμφωνεῖν cf. 5.9. James refers to
the words of the prophets but quotes only Amos 9.11f. (the slight
verbal echoes of Jer. 12.15; Isa. 45.21 can hardly count as prophetic
sayings that agree or disagree with anything). Prophets (plural) may
be a reference to the book of the Twelve (minor) Prophets, but this
will hardly account for λόγοι unless λόγος is taken as word in the
most literal sense. It is better to suppose that καθώς γέγραπται (cf.
7.42 and very frequently in the rest of the NT) is intended to
introduce a specimen quotation. G. D. Kilpatrick (Kremer, Actes
84-6) punctuates thus: ... τῶν προφητών, καθώς γέγραπται μετά
ταῦτα. αναστρέψω κτλ.

16. The greater part of James’s quotation is undoubtedly taken

from Amos 9.11f., and it will be well to begin by setting out the LXX
text of Amos 9.11, which is reproduced in this verse.
ἐν τῇ ήμερα εκείνη ἀναστήσω την σκηνήν Δαυίδ τήν
πεπτωκυῖαν καί ανοικοδομήσω τὰ πεπτωκότα αυτής καί τά
κατεσκαμμένα αυτής ἀναστήσω καί ανοικοδομήσω αυτήν
καθώς αἱ ἡμέραι τού αἰώνος.
It appears at once that Luke (James) (1) has changed that day to μετά
ταῦτα (contrast the quotation from Joel in 2.17), (2) may in doing so
have recalled, possibly subconsciously, Jer. 12.15 (καί ἔσται μετά τό
ἐκβαλεῖν με αυτούς ἐπιστρέψω; in Acts D has ἐπιστρέψω for
αναστρέψω), (3) has simplified and abbreviated Amos’s cumber-
some sentence, (4) has introduced the verb ανορθώσω (which may
reflect 2 Kdms 7.13, 16, 26; 1 Chron. 17.12, 14, 24; 22.10), and (5)
has omitted Amos’s reference to the days of old ( )—Acts is
concerned with something new, not a renewal of the past though it
was foretold in the past.
In Amos, the fallen tent of David is presumably the Davidic royal
house, which came to an end with the fall of Jerusalem in 587/6 BC.
The last paragraph of the book (whether written by Amos himself or
a subsequent editor) promises its restoration. This however would
involve also the restoration of the kingdom, that is, of the tribe of
Judah (and Benjamin). Luke (James) may have taken this to refer to
the Messiah, whether to his appearance in history or specifically to
his having been raised up from death (cf. the ambiguity of 13.33); it
is also however possible that he understood the prophecy to mean the
restoration in the sense of the conversion of Israel. This divergence
of interpretation will have an important bearing on our understanding
of the passage as a whole. Is James made to say, Now that Christ has
been raised from the dead the way is open for the Gentiles to enter at
726 COMMENTARY ON ACTS

once into the people of God? Or, The conversion of the Gentiles must
wait upon the conversion of the Jews; that is, in practical terms,
There must be no such wholesale, indiscriminate, and unregulated
admission of the Gentiles as would prejudice or hinder the mission to
the Jews? See the brief but clear discussion in Wilson (Gentiles
224f.). That the latter view was held is supported by Paul’s argument
to the contrary in Romans 9-11 (see Romans 196-210) as well as by
other controversial material in the NT (e.g. Mt. 10.5f.). It is however
unlikely that Luke, who is clearly an enthusiastic supporter of the
Gentile mission, and understood the Council as furthering the
mission, took James’s quotation to bear this meaning. It is not
unlikely that the OT passage was in early years understood in one
way but was used by Luke and his contemporaries in another. It is
also possible that at first only Amos 9.11 was used, in support of the
view that the conversion of the Jews must come first; there is no
question here of a difference between the Hebrew and Greek texts (as
in v. 17). Each predicts a restitution of Israel under a Davidic king,
who could be understood to be the Messiah.
There is some support for this in the fact that Amos 9.11 is quoted
in the Qumran MSS (see especially Fitzmyer Essays 25, 50f., 88). In
CD 7.15, 16 (in the course of an interpretation of Amos 5.26; cf. Acts
7.42f.) the words Ί will raise up the fallen tent of David’ are quoted,
and it is explained that the books of the Law are
the king’s tent; the king is the community (7.17: ). In 4QFlor.
1.11—13 the fallen tent, which will be raised up, is the shoot
of David, who will stand up to save Israel
that is, the Messiah. Fitzmyer (51) says, ‘There is, however, no
similarity in the use of the text in the three places [two in the Qumran
MSS, one in Acts],’ and adds that ‘without any reference to a scion
of David he [James] asserts the fulfilment of the verse in the
conversion of file Gentiles to the Gospel.’ It seems better to see the
conversion of the Gentiles in the next verse (not quoted at Qumran;
see below); one interpretation (see above) of Amos 9.11 as
quoted here does see a reference to the Davidic Messiah. There
remains however a marked difference between Acts and the Qumran
Florilegium; in the latter it is said that the Shoot of David will be
the one who stands with the interpreter of the
Law. Acts knows no such second figure beside the Messiah.
This verse is discussed in the light of the Qumran passages by
Wilcox (49); it is hard to see why the ἀναστήσω of Acts should
be particularly associated with the Massoretic the LXX’s
ανοικοδομήσω with of CD 7.16. The Messianic inter-
pretation of Amos 9.11 reappears in Sanhedrin 96b-97a, where R.
Nahman explains to R. Yitzhaq that Bar Naphli is a name of the
Messiah, and bases this interpretation (there are others) on Amos
9.11 (where ‘that has fallen’ is han-nopheleth).
 39. COUNCIL IN JERUSALEM. 15.6-29. 727

For κατεσκαμμένα (P74 A C D <a>), (B) Ψ 33 326 pc have

κατεστραμμένα, and E has ανεσκαμμένα. κατεστραμμένα appears
also in some MSS of the LXX.

17,18. Here we may begin with the LXX text of Amos 9.12.
ὅπως ἐκζητήσωσιν οἱ κατάλοιποι τῶν ανθρώπων [A adds, τόν
κύριον], καί πάντα τά έθνη, ἐφ’ οὓς ἐπικέκληται τό όνομά
μου ἐπ’ αυτούς, λέγει κύριος ό θεός ό ποιών ταῦτα.
James (Luke) adds ἄν after δπως; this is common in Attic prose
and verse but was dropping out of use in the Hellenistic period
(though not uncommon in the papyri). It is used also at 3.20; the MS
A has it at Amos 9.12, but this is probably a reading back from
the NT, as also was τόν κύριον which provides an object for
ἐκζητήσωσιν. James omits ό θεός, probably because he understands
κύριος to refer to Christ.
ἐπ’ αυτούς, taking up the relative ἐφ’ οὕς, crudely represents the
construction of the indeclinable Hebrew relative: upon whom my
name has been named, that is, with a view to marking them out as my
property. The masculine οὕς, αυτούς, follows the neuter πάντα τά
έθνη ad sensum; ‘the Gentiles’ signifies a multiplicity of individual
human beings. πάντα τά έθνη is explanatory of οἱ κατάλοιποι των
ανθρώπων, ‘the rest of men’, those who are not Jews. What is not
clear is whether the relative clause is a limiting or an exclusive
definition, that is, whether it means, All the Gentiles upon whom my
name has been named, to the exclusion of those over whom it has not
been named, or All the Gentiles, upon all of whom my name has
been named. It is probable that Luke was prepared to let questions of
this kind answer themselves. Cf. 13.48; it became clear who were
appointed to eternal life when some believed and others did not. In
the same way it would become clear which Gentiles bore God’s
name and which did not.
Luke’s (James’s) use of his text is thus clear; but his text is based
on the LXX, not on the MT, which differs in several respects. It is
translated in RSV: That they may possess the remnant of
Edom and all the nations who are called by my name
(literally, upon whom my name has been called,
says the Lord who does this. In this Hebrew sentence Amos predicts
the dominion of Israel over all other nations, including what is left of
Edom, whom God is claiming as his possession. This is not what
James and the LXX say. In place of they shall possess, they
read they shall seek. Instead of nx (the sign of the direct
object), they read the Lord. Instead of the remnant
of Edom, they read, the remnant of men, of humanity. In
this way the meaning of the verse is almost reversed.
It is clear that neither James nor Luke invented this convenient
728 COMMENTARY ON ACTS

form of the text; it already existed in the LXX. But is it likely that
James, meeting in Jerusalem with fellow Jews (who happened also to
be Christians) would use a form of the text that differed from the
Hebrew? The Targum, incidentally, makes it even clearer that the
house of Israel ( is inserted in the text) will possess (porr;
v.1., ) the remnant of Edom and all the nations. Gerhardsson
(Memory and Manuscript 260) holds that James might have taken up,
as the rabbis did, any known textual variant that would serve his
purpose, but this would need a good deal more discussion and
justification than is provided. One would expect at least a trace of the
form of argument, e.g. Read not but and so on.
It must be concluded, not with certainty but with high probability,
that the quotation, and probably therefore the whole speech, cannot
be attributed to James. This does not mean that Luke did not find the
quotation among OT passages traditionally used in arguments about
the Gentile mission, but it was probably used by Christian Jews who
habitually used the OT in Greek. This observation has an important
bearing on the historicity of the Acts council as a whole; see
Introduction, pp. xxxvi-xli. The different ways (see above pp. 725f.)
in which the Amos quotation may have been applied to the legit-
imacy of the Gentile mission are particularly important.
The LXX of Amos 9.12 ends, λέγει κύριος ὁ θεός ό ποιων ταῦτα,
which is a reasonable representation of the Hebrew
Acts 15.17 reproduces this with the omission of ό θεός and of ὁ
(though A C D2 E <a> syh Cyril have ὁ). In NA26 v. 18 continues
with the words γνωστά ἀπ’ αἰῶνος. In this reading the editors follow
B C Ψ 33 81 323 (1175) 1739 2495 al w. The words have been
held to be based on Isa. 45.21, ... ἵνα γνῶσιν ἅμα τίς ακουστά
ἐποίησεν ταῦτα απ’ αρχής, though the resemblance is far from
close. James’s version makes what appears to be the point more
strongly than the LXX (of Isaiah); by the prophetic oracle God made
these truths about the inclusion of the Gentiles not only heard but
known; and ἀπ’ αἰῶνος says more forcefully than ἀπ’ αρχής that they
were part of his eternal plan. But γνωστά ἀπ’αἰῶνος is probably a
simple gloss rather than an additional quotation.
There are variants for γνωστά απ’ αἰῶνος. P74 A D E 3ft lat (sy) Ir
have ταῦτα (+ πάντα Ε <a>) γνωστόν (γνωστά Ε 3ft) ἀπ’ αἰῶνος
ἐστιν (om. P74 A) τῷ κυρίῳ (θεῷ Ε <a>) τό ἔργον (πάντα τά ἔργα Ε
3ft) αυτού. 945 pc have ταῦτα πάντα ἅ έστιν γνωστά αὐτῷ απ’
ἀιῶνος. 2127 has πάντα τά ἔργα αυτού. All these variants except
the last make the same point: God has not suddenly thought of the
inclusion of the Gentiles; it has always been his intention, and he has
long made his intention known. They are wordier variations on the
short text, and this should be preferred.
H. Sahlin (NovT 24 (1982), 187) conjectures ὡς τὰ ἀπ’ αἰῶνος.
G. D. Kilpatrick (Kremer, Actes 84-6) punctuates:... λέγει κύριος ό
 39. COUNCIL IN JERUSALEM. 15.6-29. 729

ποιῶν ταῦτα. γνωστόν ἀπ’ αἰῶνος έστιν τῷ κυρίῳ τὸ ἔργον

αυτού.

19. διό presumably looks back to both the historical fact adduced
in v. 14 and to the prophetic word of vv. 16, 17. ἐγώ is pointless if
not emphatic: This is what I say. And not merely say. κρίνω may
express no more than a private judgement: This is my opinion. But it
may very naturally be taken as the judgement of a judge, who has the
right to declare a final verdict: I give judgement that... ; for example,
cf. Homer, Odyssey 12.440, κρίνων νείκεα πολλά; Thucydides
1.87.1, κρίνουσι γὰρ βοῇ καί οὐ ψήφῳ. Weiser (383) thinks the
word no stronger than erachten, and at Thucydides 4.60.1 the
scholiast paraphrases ὡς εγὡ κρίνω as ὡς εγώ νομίζω. But there is
no further discussion after James’s speech, and though Ί decree’
(Begs. 4.177) is too strong and perhaps does not catch the context
correctly James is at least acting as a chairman and expressing in his
own words the sense of the meeting. If Luke is right in the picture at
which he hints James occupies in the assembly a position if not of
pre-eminence at least of great prominence. Gerhardsson (op. cit. 252)
refers to Bacher (Terminologie 1.20f.) for the similar expression

The impression of James’s prominence gains some support from

the absence of an accusative subject with the infinitive παρενοχλεῖν.
It is not wrong to supply the first person plural, but James does not
express it, and without a pronoun the singular could well be
understood: I am deciding not to trouble ... (with the implied
assumption that the speaker has it in his power to trouble or to
abstain from troubling). There is little difference between ὁχλεῖν,
ἐνοχλεῖν, and παρενοχλεῖν, though the effect of composition is
usually to reinforce the sense of the simple verb, which means to
cause trouble or annoyance. This may be material (e.g. Plutarch,
Timoleon 3.1 (237)) or mental, spiritual (e.g. Epictetus 1.9.23, of the
annoyance caused by Socrates in his persistent questioning). In
Hellenistic Greek the verb usually takes not, as here, the dative but
the accusative (MM 493); the accusative may be avoided here
because it could be taken to be the subject of the infinitive. Further
examples of παρενοχλεῖν from early inscriptions are given in ND
4.166f.; in them the verb takes the accusative. Begs. 4.177 takes the
present infinitive (with μή) to mean stop annoying.
τοῖς ἀπό τῶν ἐθνών ἐπιστρέφουσιν ἐπί τόν θεόν could be said
by a Jew of Gentile converts to Judaism; to a Jewish Christian the
Christian conversion of Gentiles must have had to a considerable
extent the same appearance. Gentiles were turning from whatever
heathen gods they had previously worshipped to the God of the OT,
the God of the Jews. It was this fact that gave strength to the
requirement that they should behave like converts to Judaism, that is,
730 COMMENTARY ON ACTS

should be circumcised and thereafter keep the Law. ἐπιστρέφειν

(though without ἐπί τον θεόν) could be used to describe what was
required of Jews (3.19).
The Gentile converts must not be pestered; the context makes clear
that this means that demands of full legal observance must not be
made.

20. This verse, substantially repeated as adopted by the Council in

v. 29 (cf. 21.25), calls for particularly careful examination; it
contains noteworthy problems both in text (see, in addition to the
note below, Metzger 429-34, with bibliography) and interpretation.
James proposes not to cause trouble for the converted Gentiles but
ἐπιστεῖλαι αὐτοίς... The precise force of ἐπιστέλλειν is not easy to
assess. It may mean no more than to send a written message; so Heb.
13.22, where it appears that the content of the message is
παράκλησις. Cf. Thucydides 8.50.2 (κρύφα ἐπιστείλας, having sent
a secret message); also 8.38.3; 8.99.1. More frequently however the
word conveys the further meaning of send written instructions; not
simply a message or exhortation is conveyed but orders (usually in
writing). ‘From the usage of the word in official documents the
meaning readily passed over into “instruct”, “enjoin” ’ (MM 245),
and passages show that the instructions were normally communi-
cated in writing. That in the present passage ἐπιστέλλειν points to a
required course of action is made clear by the infinitive (τού
ἀπέχεσθαι) that follows. What cannot be certainly inferred is where
the injunction stands on the scale between We write with the
suggestion that you might consider abstaining ... and These are our
written instructions: You shall abstain ... This question must be
considered again when we reach (vv. 23-29) the letter that Luke
ascribes to the Council. The impression however at this stage is that
James (as represented by Luke) wishes to proceed suaviter in modo,
fortiter in re. He wishes to make things as easy as possible for the
Gentiles, but there are limits ... ‘Not to make trouble for them but
write to them that they should ...’ retains the ambiguous feeling in
English.
The content of the request, suggestion, or command is expressed
by τού with the infinitive. The τού is pleonastic, but characteristic of
Luke; cf. 3.12; (5.31); 10.25; 13.47; 21.12; 23.15, 20; 27.1. The
middle ἀπέχεσθαι is used with the meaning to abstain from a thing
or from a course of action at all stages in Greek literature; there is a
good example in a second century AD inscription given in ND 1.21
(προστάσσουσι δὲ Δορατῇ τω νεωκόρω τούτων των μυστηρίων
ἀπέχεσθαι). In this inscription and in older Greek generally the verb
is used without a preposition (as here, in P45 B D 81 1175 pc e p*;
ἀπό is supplied by P74 A C E Ψ <a>lat).
Four abstentions, described by nouns in the genitive case, are
 39. COUNCIL IN JERUSALEM. 15.6-29. 731

given in the text of NA26. These will be considered first before

variant texts are discussed. A number of important Jewish parallels
will be found in full in StrB 2.729-39.
(1) ἀλισγημάτων τῶν ειδώλων. ἀλίσγημα is hapax legomenon in
the Bible and very rare elsewhere. It is derived from άλισγεῖν, to
pollute, defile. This verb occurs four times in the LXX. In Dan. 1.8 it
is used in describing Daniel’s determination not to allow himself to
be defiled by the food and drink served at the table of the king of
Babylon. This was food which for a variety of reasons a Jew would
be unwilling to touch; it would not have been prepared in the ways
required by the Law, it might well consist of the flesh of unclean
animals, it was certain that tithe would not have been paid on it. It is
not noted in the context that it might have been offered in sacrifice to
a pagan god, or idol (and thus an εισωλόθυτον, cf. v. 29). In Mal. 1.7
the verb is used twice (three times in the MSS B A) of the priests
who have profaned (φαυλίζειν; (1.6)) God’s name; they have
done this by offering on the altar άρτους ἠλισγημένους. The loaves
are defiled because the priests defile the altar (τράπεζα κυρίου
ἐξουδενωμενη ἐστίν; and the offering. At Mal.
1.12 the word is used of the altar itself. There is no suggestion that
offerings have been made to other gods. Sirach 40.29 is difficult: A
man who looks (with envy) on another man’s table, his life is not to
be reckoned life; he will defile his soul (ἀλισγήσει ψυχήν αυτού)
with another man’s dainties. Again, the context is concerned with
food, but not with idolatry; defilement arises because food is the
cause of envy and is obtained only by begging (40.28: ‘Son, do not
live a life of begging; it is better to die than to beg’). That the eating
of food sacrificed to idols was a source of defilement was of course a
Jewish belief, which no doubt many Jewish Christians continued to
hold.
ειδώλων must be a subjective genitive; the idols cause defilement,
but in the word ἀλισγήματα there is nothing that would restrict this
defilement to that caused by the eating of food that had been
sacrificed to the idols. This however is what ἀλισγήματα becomes in
v. 29 and 21.25; it is a possible but not a necessary meaning of the
word. Thus at The Rest of the Words of Baruch 7.32(37) Jeremiah
έμεινε διδάσκων αυτούς τού ἀπέχεσθαι ἐκ των ἀλισγημάτων των
ἐθνών τῆς Βαβυλῶνος. It seems possible that there was an original
prohibition of idolatry which (under the influence of such tensions as
are apparent in 1 Corinthians 8 and 10) developed into a specific
prohibition of εἰδωλόθυτα. Idolatry is forbidden in the OT (e.g.
Exod. 20.4, οὐ ποιήσεις σεαυτῷ εἴδωλον), and at Lev. 17.7 it is
forbidden to offer sacrifice to idols (οὐ θύσουσιν ἔτι τὰς θυσίας
αὐτῶν τοῖς ματαίοις; in view of what follows in Acts it is
interesting that the verse continues, οἷς αυτοί ἐκπορνεύουσιν ὀπίσω
αὐτῶν).
732 COMMENTARY ON ACTS

(2) τής πορνείας (omitted by P45; see below). The common

meaning of the word is prostitution, fornication, uncleanness; this is
borne out by the use of related and compound words, such as πόρνη,
πόρνος, πορνοβοσκεῖν, πορνοκόπος and the like (see LS 1450). A
prohibition of πορνεία in this sense is intelligible and appropriate. It
is however to be noted that in 1 Cor. 5.1 Paul uses the word πορνεία
to describe the act of one who has taken (ἔχειν) his father’s wife.
This Paul describes as an exceptionally gross case of sexual irregu-
larity. It is expressly forbidden in Lev. 18.8 (ἀσχημοσύνην γυναικὸς
πατρός σου οὐκ αποκαλύψεις), which occurs in a passage that deals
with forbidden degrees of marriage (Lev. 18.6-18). This passage
does not use the word πορνεία but it (together with some other
regulations for sexual behaviour) is made to apply not only to the
native Israelite but also to ό προσγενόμενος προσήλυτος ἐν ὑμῖν
(Lev. 18.26, It is sometimes supposed (e.g. Schneider
2.183; Weiser 383) that we have here the proposed extension to
Gentile Christians of the marriage laws that were (according to
Leviticus) to be observed by aliens resident in Palestine. There is
however no convincing reason for thinking that πορνεία was ever
understood to refer simply to marriage within the forbidden degrees.
In 1 Cor. 5.1 (see above) the word signifies a relation that was not
adultery (assuming that the father was not alive) and not in the
physical sense incest, if the woman was, as is usually supposed, the
fornicator’s stepmother.
(3) τοῦ πνικτοῦ (so P45 C E <a> sy; the omission of τού, by P74
A B Ψ 33 81 pc, is not important; for the omission of the whole by D
gig Irlat see below), derived from the verb πνίγειν, to choke, strangle,
suffocate, is, literally rendered, that which is strangled. H. J. Schoeps
(Paulus 60) turns this into a more definite picture with ‘Tieren, die auf
der Jagd mit Schlingen gefangen und getötet worden sind.’ It is
however extremely difficult to find evidence in Jewish sources for the
specific classification by this or any related word of food that must be
rejected. The natural Hebrew and Aramaic equivalent is the root
this does not however seem to be much used of animals, though there
is an important passage in Philo, Spec. Leg. 4.122, where Philo attacks
men of the type of Sardanapolus who desire novel kinds of pleasure
and prepare meat unfit for die altar (ἄθυτα) by strangling and throt-
tling (ἄγχοντες καί ἀποπνίγοντες) the animals. The use of the com-
pound verb may perhaps be taken as an indication that the simple
πνίγειν was sometimes used in the same way, but we lack direct
evidence of this. Moreover, as the context in De Specialibus Legibus
shows, and as would in any case be deduced from the picture of a
snared and strangled animal, the objection to the use of such an animal
as food lay in the fact that its blood had not been drained away. That is
to say, if this is what πνικτόν means, along with αἶμα (as αίμα is
often understood; see below) it is superfluous.
 39. COUNCIL IN JERUSALEM. 15.6-29. 733

According to StrB 2.730, πνιχτόν includes both, ‘sowohl was das

AT als auch was es nono nennt’. These two Hebrew words are
conveniently defined in terms of Josephus, Ant. 3.260 (κρέως τοῦ
τεθνηκότος αυτομάτως ζφου τήν βρῶσιν διεκώλυσεν) and of
Exod. 22.30 This interpretation of
πνικτόν has been widely accepted together with the further points
involved in the claim (StrB 2.730-4) that in rabbinic use came
to mean any beast that had not been subjected to the ritually correct
slaughtering process and a beast which, dead or alive,
had a disqualifying blemish. The Greek word πνιχτόν may have
been regularly used and understood in this way, but evidence to
prove this is lacking.
(4) τοῦ αἵματος. In the form of the text that we are at present
considering this is usually understood in relation to the OT prohibi-
tion of the consumption of blood. This appears in the primeval
commandment given to Noah (Gen. 9.4) and is repeated in later
codes (Lev. 7.26, 27; 17.10-14; Deut. 12.16, 23). But, since loss of
blood means loss of life, αἷμα came also to be used of bloodshed, in
particular, of murder. This appears in general narrative and in
idioms, such as ἐφ’ αἵματι φεύγειν, ‘to avoid trial for murder by
going into exile’ (LS 38; Demosthenes 21.105; Dittenberger, Syll.
1.58.2). See also Sophocles, Oedipus Tyrannus 101; Euripides,
Orestes 285. Since αἷμα can thus mean bloodshed it is equivalent to
the rabbinic It must therefore not be too readily assumed
that avoidance of blood means avoidance of blood in food, especially
since avoidance of πνιχτόν (if this means meat improperly slaugh-
tered) would include avoidance of blood, making the specific τού
αἵματος (in this sense) redundant.
The textual variations, which have the effect of producing a
somewhat different ‘draft Decree’, will be considered shortly; it will
be convenient first to consider several suggestions that have been
made regarding the background and origin of James’s proposals (and
of the Decree as formulated in v. 29). These suggestions are not
mutually exclusive.
(a) It may be that the terms of the Decree were thought of as
practical considerations bearing on the situation as it could be seen to
be arising out of the Christian mission to the Gentile world. The
fundamental requirement of the Gentile convert was that he should
abandon the religion that he had previously practised; that is, he must
abandon the gods he had worshipped, turning his back on idolatry.
He must abstain from the spiritual defilement that comes from
idolatry. Verse 20 states this in absolute terms; v. 29 and 21.25 make
the assumption that to eat εἰδωλόθυτα is to commit idolatry and is
probably for most Gentile Christians the way in which they would be
most likely to commit it. Jews had long known that the temptation to
idolatry came most often through the butcher’s shop and the brothel.
734 COMMENTARY ON ACTS

Hence what is in effect the command to use only Jewish butchers,

where one could be confident that no εἰδωλόθυτα, πνικτά, or αἷμα
would be sold, and the prohibition of πορνεία. It should be noted that
such commands, especially the prohibition of idolatry, would be
necessary for salvation, and not merely in order to facilitate fellow-
ship between Jewish Christians and Gentile Christians.
(b) The requirements may have been based upon the so-called
Noachian precepts, precepts which it was believed were incumbent
not only upon Jews but upon all, of whatever race. In their earliest
form (Jubilees 7.20) these are ‘to observe righteousness, and to cover
the shame of their flesh, and to bless their Creator, and honour father
and mother, and love their neighbour, and guard their souls from
fornication and uncleanness and all iniquity’. These were subse-
quently developed (evidence is given in full in StrB 3.37f.) into the
prohibition of idolatry, blasphemy, murder, incest, stealing, pervert-
ing justice, and eating flesh containing blood. Of these the first, third,
fourth, and seventh appear (more or less) in James’s decree. The
parallel is not close, and there is nothing in the text of Acts to call
Noah to mind.
(c) A better, and today widely accepted, parallel is to be found in
the regulations given in Leviticus 17 and 18 for Gentiles living
among Jews. These are directed to ‘Any man of the house of Israel,
or of the strangers that sojourn among them
Lev. 17.8)’. They require that sacrifices must be brought to the
door of the tent of meeting (17.8f.); that blood must not be eaten
(17.10—14); that what dies of itself or what is tom by beasts
must not be eaten (17.15f.); the forbidden degrees of
marriage must be observed (18.6-30). These commands may be said
to cover, in Acts, the prohibition of πορνεία (if that is understood to
mean not fornication in the ordinary sense but forbidden family
relationships), πνικτόν (if this is interpreted to mean and
and αίμα (if this means blood as a food and not bloodshed). It cannot
be said that the connection is close; see however Stählin (205)
Schmithals (139); also Wilson (Law 85-102).
(d) A still better background, which fits readily with (a) above, is
provided by a group of passages (p. Shebiith 35a. 49f.; p. Sanhedrin
21b. 10f.; b. Pesahim25ab; b.Sanhedrin 74a) in which it is urged
that, though in persecution a Jew was not expected to give his life
on any minor issue, there were three matters on which compromise
was impossible. These were idolatry; the
shedding of blood; and incest. Here are what Luke calls
the defilements of idols, blood, and πορνεία (which includes incest).
There is nothing to correspond with the prohibition of πνικτόν; it is
hard to supply any motivation for the inclusion of this except the
desire to make it easy for Jewish and Gentile Christians to eat
together; such food, with εἰδωλόθυτα (readily understood—though
 39. COUNCIL IN JERUSALEM. 15.6-29. 735

not by Paul—as included in the ἀλισγήματα), would be objec-

tionable to Jews and no doubt also to many Jewish Christians. See
further Wilson (Law 68-102). Johnson (273) combines (c) and (d).
The important textual variants in the verse must now be noted.
(a) καί τής πορνείας is omitted by P45 (not extant at 15.29; 21.25);
also (though not noted in the apparatus of NA26) by some Ethiopic
MSS, and (we may infer) by Origen in c. Celsum 8.29, though here
Origen is referring to the letter sent by the Council, that is, to v. 29. He
speaks of the decision ‘to write a letter to the Gentile believers,
demanding that they should abstain only from what they called “the
essentials” (μόνα (ὡς ὠνόμασαν) ἐπάναγκες). These are things
sacrificed to idols, or things strangled, or blood’ (Chadwick 472).
Little can be built on this passage, for Origen is dealing with food (and
might therefore omit πορνεία as irrelevant), and is writing with less
than his usual care, for he says that the Council was held at Antioch. It
is probably correct to regard the reading of P45 not as one important
witness to an otherwise unknown (Caesarean) text of Acts but as a
simple error. See however G. Zuntz, Opuscula Selecta (1972), 227-9.
It is unwise to use the omission as an argument for an original two-
clause decree (P. H. Menoud, SNTS Bulletin 2 (1951), 22-8).
(b) καί τοῦ πνικτοῦ is omitted by D gig Irlat. The same omission is
found in v. 29 and 21.25; there can be no doubt that it represents
another form of the text of the Decree. The omission of an element
that is purely ritual, with no ethical bearing whatever, undoubtedly
alters the balance of the Decree and has often been held to affect the
interpretation that must be given to the reference to blood. With
πνικτοῦ, blood suggests a ritual food law; without it, it may more
readily be understood as a reference to bloodshed, murder. This is
however a double-edged argument. Knowling (325) thought that
πνικτόν was omitted because it was thought to be covered by αἷμα,
Clark (360) that it was added ‘to make it clear that αἵματος referred
to diet as in Leviticus xvii.14, not to homicide’.
(c) After αἵματος the Western text makes a substantial addition:
καί ὅσα ἂν μὴ θέλωσιν (ὅσα μή θέλουσιν, D) αὐτοῖς (ἑαυτοῖς, D)
γίνεσθαι ἐτέροις μή ποιεῖν (ποιείτε, D), D 323 945 1739 1891 pc sa
Irlat. This addition of the negative form of the Golden Rule strongly
reinforces the ethical effect of the omission (also by the Western
text) of πνικτού.
It has often been maintained that when the Old Uncial text and the
Western text are compared we find on the one hand a set of
essentially ritual requirements: abstinence from sacrificial food, from
fornication (which could be understood to mean forbidden marriages
or sacral prostitution), from non-kosher food, and from food contain-
ing blood; and on the other hand an ethical code, requiring the
rejection of idolatry (which, put positively, means adherence to the
exclusive worship of the one true God), avoidance of all unchaste
736 COMMENTARY ON ACTS

behaviour, avoidance of murder, and obedience to the fundamental

moral law, expressed in the Golden Rule. With very few exceptions,
textual critics have maintained that the ritual form must be primitive,
and have been transformed (by the omission of πνικτοῦ and the
addition of the Golden Rule) into an ethical code, perhaps in the
second century, certainly at a time when the older code was no longer
understood. There is enough force in this argument to make one
prefer the Old Uncial text as Luke’s work. As often expressed,
however, it is an over-simplification. It can be shown (see FS
Osborn, 50-59; Taylor 210) not only that the Decree continued for
centuries to be observed (and often to be found to require explana-
tion—or explaining away), but that it was interpreted in different
ways in different circumstances. Just as the circumstances that
evoked it had both ritual and ethical elements, so the Old Uncial
Decree had both ritual and ethical elements, especially in the form in
which James proposes it, which shows that the real concern in the
prohibition of ἀλισγήματα τῶν ειδώλων (εἰδωλόθυτα) was the
avoidance of idolatry itself (and hence the cultivation of true
worship), πορνεία may have certain specific associations, but it
means fornication, and the Decree forbids the sin of fornication.
πνικτόν does refer to non-kosher food, and this is a non-ethical
element in the Decree, but since the consumption of blood is already
forbidden in the prohibition of πνικτόν it is by no means impossible
that blood means murder, violence. It is not surprising that those who
accepted the authority of this mixed Decree emphasized sometimes
one sometimes the other aspect of it. The Western editor(s) came
down strongly on the ethical side.
The Golden Rule, in its positive form, appears in Mt. 7.12; Lk.
6.31. In the Matthean form there is the appended note, For this is the
law and the prophets. It was already established in Judaism (though
mostly used in the negative form) as a summary of man’s moral
obligation; see e.g. Tobit 4.15 (δ μισείς, μηδενὶ ποιήσης); b.
Shabbath 31a (Hillel: What is hateful to yourself, do not to your
neighbour); Philo, Hypothetica 7.6 (in Eusebius, Praeparatio Evan-
gelica 8.7.5, ἅ τις παθεῖν ἐχθαίρει, μή ποιεῖν αυτόν). See further
Bousset, RJ 138; StrB 1.459f.
For the further history of the Decree see H. J. Schoeps, Theologie
und Geschichte des Judenchristentums (1949), 303; Aus frühchris-
tlicher Zeit (1950), 78. See further on v. 29 and 21.25; also above.

21. Moses (as not only the author of the Pentateuch but the founder
of the Jewish religion) has (έχει, described by Moule (IB 8) as a
‘present of past action still in progress’ and translated by him as
has had) from generations of old (this—cf. StrB
2. 739f.—goes further back than ἀφ’ ήμερων αρχαίων in v. 7) in
every city (distributive use of κατά; cf. 2.46, and other passages)
 39. COUNCIL IN JERUSALEM. 15.6-29. 737

those who proclaim him, τούς κηρύσσοντας αυτόν. James thus uses
a verb frequently used of the proclamation of the Christian Gospel,
the proclamation of Christ (cf. 8.5; 9.20). That there were those who
proclaimed Moses as a cult figure is certainly true; see 7.22 and the
notes there and on I.338f. James however explains what he means in
the participial clause that follows: Moses is proclaimed in that he is
read (aloud) in the synagogues every Sabbath. For the synagogue
service see on 13.15. Throughout the Empire, and beyond, wherever
there were Jews the words of Moses were heard every week. Cf.
Josephus, Apion 2.175, ἐκάστης ἐβδομάδος των άλλων ἔργων
ἀφεμένους επί τήν ἀκρόασιν ἐκέλευσε τού νόμου συλλέγεσθαι καί
τούτον ακριβώς ἐκμανθάνειν; cf. 282. See Bousset (RJ 72, 172);
Tcherikover (269-332). ‘Jewish communities of greater or lesser
extent and significance had settled in almost every part of the then
civilized world’ (NS 3.3; for details, 3-86). It may be that the use of
κηρύσσειν implies (as it does with Christian preaching) some kind of
authority; Moses is not merely read but asserted as an authority, so
that what is read (including for example Leviticus 17 and 18) must be
obeyed. This inference seems very doubtful.
The sentence begins with for (γάρ), and therefore presumably
gives the ground for an earlier statement, but it is not clear which
statement it supports. (1) It may look back to v. 19: It is not for us to
trouble the Gentiles since Moses already has enough preachers to
take his part. Gentiles, moreover, who want the Law can go to the
synagogue for it (Williams 184). (2) It may look back to v. 20. It is
necessary to lay some obligations on the Gentiles, since Moses has
everywhere so many followers—so many observant Jews. (3) It may
look back to vv. 15-18: We can see that Amos’s prophecy is already
being fulfilled in the large number of those who preach the Law.
Dibelius (92) suggests that v. 21 may be a marginal gloss on the final
words of the quotation. There is no serious basis for this suggestion,
and as it stands v. 21 is too remote from the quotation for a reference
to it to be likely. The general tone of James’s speech is favourable to
Gentiles and we may therefore accept (1), but the whole passage is so
obscure that it may be that Luke intended both (1) and (2): It is not
necessary to trouble the Gentiles seriously (by demanding circumci-
sion) but it is necessary to trouble them a little (by the Decree).
Alternatively, Luke may be placing traditional material in a new
setting and thereby giving it a new meaning. He is no doubt
responsible for James’s speech as it now lies before us, but it is
improbable that he simply invented it—the Decree itself is certainly
traditional, though it was not (see p. 712) James’s work. If the Decree
is omitted the speech would impose no conditions upon Gentiles and
connection (1) would then be virtually the only possibility, though it
might be combined with Filson’s observation, ‘It seems clear from
Acts 15.21 that [James] expected synagogue worship to go on in
738 COMMENTARY ON ACTS

every city and that he also expected Jewish Christians to keep the
Mosaic Law’ (Decades 82). Perhaps somewhat less than clear; see
the good discussion by Wilson (Law 83f.).
Calvin’s comment (2.51f.) is worth quoting: ‘He warns that it is
not possible for the ceremonies to be abolished so quickly, as if at
one fell swoop, because the Jews had already been accustomed to the
teaching of the Law for many generations, and Moses had his
preachers; that agreement therefore must be gained for a short time
until the freedom, procured by Christ, should gradually be more
clearly understood; in other words, as the common saying goes, that
the ceremonies had to be buried with some decency.’ In the early
church there were those, notably Paul, who were quick to see the
theological principle involved in every arrangement, but no doubt
there were many more, who may or may not have included James but
certainly included Luke, who were happy to proceed on (what
seemed to be) common-sense lines.

22. The apostles and elders were those who had met to consider
the question raised at Antioch (v. 6). Here they meet σὺν ὅλη τῇ
ἐκκλησία, the whole body of Christians, that is, of the Jerusalem
church; the church of Antioch had sent only a small delegation
(15.2), and no other church seems to have been represented. The rank
and file Christians have disappeared in the next verse (but see the
note there on αδελφοί) and are given no share in the writing of the
Council’s letter. It goes too far to say, ‘Clearly only the presence of
the congregation gave legal validity to the resolution, although the
members of the congregation did not take part in the discussion and
did not vote in the decision’ (B. Reicke, in Stendahl, Scrolls 146).
The report of the discussion is extremely scanty and no vote is
recorded. Luke is in any case not thinking in terms of legal validity;
what was important was the claim (v. 28) that the Holy Spirit had
directed the proceedings. For comparison of this ‘democratic’ ele-
ment with the structure of the Qumran community see Braun 2.328f.
The representation of monarchical, oligarchical, and democratic
elements in the NT church needs, as Braun shows, qualification;
categorization in such ready-made terms is dangerous.
The assembled company decided (ἕδοξε; the impersonal use of
this verb, very common in ‘official’ Greek, appears in the NT at Lk.
1.3; Acts 15.22, 25, 28, 34; 25.27; Heb. 12.10—a Lucan character-
istic) to write to the church at Antioch, sending the message by men
whom they chose to accompany Paul and Barnabas. ἅνδρας is made
precise by the appositional ’Ιούδαν τὸν καλούμενον (cf. e.g. 1.23)
Βαρσαββᾶν (the reading of D, Βαραββᾶν, is no more than an
unfortunate slip of the pen) καὶ Σιλᾶν. The name Judas occurs
several times in Acts, Judas Barsabbas (for this surname, used also of
Joseph, see on 1.23) only in the present context, at vv. 22,27,32, and
 39. COUNCIL IN JERUSALEM. 15.6-29. 739

34 (si v.1.). Nothing else is known of him. Silas becomes an

important companion for Paul, in succession to Barnabas: see on vv.
22,27,32,34, (si v.1.), 40; 16.19,25,29; 17.4,10,14,15; 18.5. He is
sometimes identified with the Silvanus (perhaps a latinized form of
the Semitic ) of the Pauline epistles (2 Cor. 1.19; 1 Thess. 1.1; 2
Thess. 1.1), also with the Silvanus of 1 Pet. 5.12, though he can
hardly be invoked as responsible for the good Greek of that epistle;
he was (according to the present verse) a leading member of the
Jerusalem church, and there is no more ground for supposing him
capable of stylish Greek than Peter himself. The identification with
Paul’s Silvanus is probable. For the opposite view, a distinction
between two men both bearing the name Silas, the identification of
one of them with Titus, the identification of Titus with Luke, and
some other curiosities, see the article on Silas by P. W. Schmiedel, in
EBib 4514-21.
For ἐκλεξαμένους, (P74) 33 323 614 945 1739 pc (syp) have
ἐκλεξαμένοις, which agrees with the case of τοῖς ἀποστόλοις καί
τοῖς πρεσβυτέροις. In such a sentence Greek usage is tolerant of
either case; see Zerwick § 394; BDR § 410 n. 2.
ἡγούμενος (present participle of ἡγεῖσθαι) was on the way to
becoming a noun; cf. 7.10; 14.12; and see LS 763, s.v. II 3, with
many papyrus references; also MM 277. One may guess that Judas
and Silas were among the πρεσβύτεροι.

23. γράψαντες is another false concord (cf. ἐκλεξαμένους in v.

22), perhaps better described as anacolouthon; so BDR § 468.2, n. 3,
who point out that Luke continues as if he had written οι ἀπ. καί οἱ
πρεσβ. ἐβουλεύσαντο ... πέμψαι... γράψαντες, and draw attention
to a close and interesting parallel in Thucydides 3.36.2 (ἐδοξεν
αὐτοῖς ... ἀποκτεῖναι... ἐπικαλούντες ...).
There is considerable textual confusion in the opening words of
the verse. διά χειρός αυτών may be (Begs. 4.179) a Semitism
but it is common in Acts and there is no special reason why the short
and simple text of P45vid P74 A B pc bo (γράψαντες διά χειρός
αυτών) should have caused offence to copyists. But E (33) <a> syh
add τάδε as an object for γράψαντες. (C) D gig w (syp) (sa) have γρ.
ἐπιστολήν διά χ. αὐ. περιέχουσαν τάδε. 614 pc syhmg have γρ. διά χ.
αὐ. ἐπιστολήν καί πέμψαντες περιέχουσαν τάδε. Ψ has γρ.
ἐπιστολήν διά χ. αὐ. ἐχουσαν τὸν τύπον τούτον. It is easy to
understand the addition of a simple object for γράψαντες, not easy to
understand the elaboration and variety. There seems to be no better
explanation than that copyists and editors recognized and wished to
emphasize the central importance that the letter of vv. 23-29 has in
Acts.
On the use and construction of περιέχειν see in addition to the
lexica BDR § 308 n. 5.
740 COMMENTARY ON ACTS

The text of the letter follows. The whole ἐκκλησία of v. 22 now

disappears and the letter is written in the name of the apostles and
elders, though they (or at least the elders) are characterized as
brothers. The brothers do appear as a tiers état in ΕΨ sy bomss,
which have καί οἱ αδελφοί. At the other extreme αδελφοί is omitted
by pc vgms sa Orlat; but P73 P74 A B C D 33 81 pc lat Irlat are to be
followed with the simple αδελφοί (after πρεσβύτεροι): the elders if
not the apostles are at the same time members of, brothers within, the
local church, and it is as such that they address the members of the
other churches—including Gentiles. Bruce (1.302) quotes Torrey to
the effect that the opening words are faultless Aramaic idiom
and that brothers applies to both apostles and
elders; if we may suppose that there was an Aramaic original this is a
valid observation. Bultmann (Exegetica 416; so earlier Preuschen 96)
thinks that the letter referred to its senders only as αδελφοί, (οἱ) ἀπ.
καί πρεσβ. being a redactional addition by the author. He thinks
further that the specific address, to Antioch, Syria, and Cilicia, points
to the use of a written source. It is perhaps more probable (see pp.
710f., 741) that Luke wrote the whole paragraph apart from the
Decree itself, notwithstanding ‘some old features which suggest that
it may be pre-Lukan’ (Wilson, Gentiles 187). The place names
indeed are a feature that may be original. Luke himself believed that
the letter (or at least the Decree) was of wider relevance and
circulation. Paul and Silas (16.4) delivered it more widely, and v. 19
suggests that it would apply to all areas where Jews lived in the
Dispersion. Syria and Cilicia suggest more than the hinterland of
Antioch, but the omission of Pisidia is inexplicable if it was intended
to suggest the mission area of chs. 13 and 14. For Antioch see 11.19;
13.1; 14.28; for Cilicia see on 6.9. Syria, at this time a Roman
province, was the large tract of land in northern Palestine. Till AD 72
eastern Cilicia was administered by Syria; Vespasian made all Cilicia
a separate province (CAH 11.603). This however does not prove
(Hemer 179) that Acts was written in the period of the double
province. The two names are mentioned as if they referred to two
distinct administrative areas. The use of one article (κατά τήν A. καί
Σ. καί Κ.) groups them together but does not make a unit of them, or
exclude reference to other areas.
For the recipients, the letter takes up the use of αδελφοί noted as
Luke’s in 15.1. They are ἐξ ἐθνών, but they are nevertheless brothers
because they are Christians.
Χαίρειν, an infinitive standing as a kind of imperative (BDR § 389
n. 2), is a common (and not specifically Christian—Paul does not use
it) greeting. As BDR show, this epistolary idiom is elliptic: (We wish
you) to rejoice.

24. According to BDR § 464 n. 4, vv. 24-26 form the only true
 39. COUNCIL IN JERUSALEM. 15.6-29. 741

period in Acts. Does this mean that Luke found this piece of fine
writing in his source and copied it out? or did he rate the importance
of the Decree so highly that he thought it worth while to provide it
with an ornate frame? The language of these verses suggests the
latter alternative, but not strongly, ἐπειδή occurs twice in Lk. (also
ἐπειδήπερ), 3 times in Acts, 5 times in Paul. δοκεῖν is used
impersonally once in Lk., 4(5) times in Acts, not elsewhere in the
NT. ἐκλέγεσθαι occurs 11 times in Lk-Acts, 15 times in the rest of
the NT. ὁμοθυμαδόν occurs 10 times in Acts, once in Romans,
nowhere else in the NT. υπέρ τού ὀνόματος is used 4 times in Acts,
3 times in the rest of the NT. Luke certainly regarded the Decree as a
most important part of the tradition that he received; it is not
unreasonable to suppose that he created a fitting setting (including, it
may well be, not only the letter but the whole account of the Council)
for it. On the other hand, one must ask in what sort of context the
Decree itself might be handed down.
The present verse looks back to 15.1; the trouble-makers are
emphatically disowned. They had no official backing: οἷς ού
διεστειλάμεθα. Cf. Mk 7.36; 8.15; the verb means to give instruc-
tions. This is one of the points at which other NT evidence, and
evidence within Acts itself, lead the reader to ask questions that Luke
does not answer. If the circumcisers of 15.1, 5 had had no backing,
would they have caused so much trouble and precipitated a high-
level conference? If they had no backing at the time, had they later,
in Galatia and Corinth? Who provided the commendatory letters of 2
Cor. 3.1? These are questions that must be kept in mind in any
attempt to write the history of the first decades of Christianity. We
must not forget the possibility that men genuinely commissioned
may have gone beyond their brief. This is particularly suggested in 2
Corinthians (see 2 Corinthians 31f.). And in διεστειλάμεθα we must
distinguish between, ‘We gave no charge at all’, and ‘We gave no
charge to insist on circumcision’.
For ἐτάραξαν cf. Gal. 1.7; 5.10. ἀνασκευάζειν has a variety of
meanings; see LS 120. A logician’s destruction of his opponent’s
position is as good a parallel as any; the Antiochenes did not know
where they were or what to believe. It is the opposite of
κατασκευάζειν. Thus Aristotle, Rhetoric 2 24.4 (1401b), τὸ δεινώσει
κατασκευάζειν ἢ ἀνασκευάζειν; Quintilian 2.4.18, opus destruendi
confirmandique quod ανασκευή et κατασκευή vocatur.
The disturbers of Antiochene peace went out from us, ἐξ ἡμῶν
ἐξελθόντες. It was the fact that they went from us that gave them, in
the eyes of the Antiochenes, a disturbing authority, ἐξελθόντες is
however omitted by B 1175 pc; it is read by P33 P74 A C D Ε Ψ <a>
latt sy (sa) bo Itlat. There is a strong case for omission, because
though the MSS that omit are few they are old, and the short text is
lectio difficilior, for it makes the disowned envoys ‘some of us’
ΊΑ2 COMMENTARY ON ACTS

(rather than ‘some sent from us’). There is much to be said for
omitting ἐξελθόντες. See Metzger (436).
25. The report of trouble in Antioch led to a decision in Jerusalem:
ἐδοξεν ἡμῖν. The impersonal use of ἔδοξεν occurs twice in this letter
(also v. 28). It is possible (see above) that Luke found the whole letter
extant in this form and borrowed the expression for narrative use in v.
22; perhaps more likely that he himself elected to use it in all these
sentences; it was frequently used to introduce decrees. ‘Frequently of
a public resolution... especially in decrees and the like’ (LS 242, s.v.
II 4b, with many illustrations, including Thucydides 4.118.7, ἔδοξε
τω δήμω (see the whole paragraph); Aristophanes, Thesmophor-
iazusae 372 (379), ἔδοξε τῇ βουλή). Cf. also Josephus, Ant. 16.163.
Luke knows and uses the right word for his purpose, which is rather to
utter a decree than to express an opinion.
ὁμοθυμαδόν is a Lucan word, often used by Luke (see on 1.14)
with more than its usual meaning of unanimity, denoting also
physical assembly; here it probably has both meanings. The apostles
and elders met and reached a common mind. This means that,
according to Luke, even the extremists agreed (Wilson, Law 107).
ἐκλεξαμένοις. In a similar sentence in v. 22 the accusative was used,
though in each case there is textual variation. Here the dative is
read by P45vid A B L Ψ 33 81 614 945 1175 1739pm, the accusative by
C D E Η P 36 323 1241 2495 pm. See Metzger (437), who finds the
textual question ‘difficult to decide’; it looks as if he, against the
majority of his committee, preferred the accusative. Barnabas and
Paul (the order may be significant—Barnabas was an early Jerusa-
lem disciple; cf. v. 12) are τοῖς αγαπητοῖς B. και Π.... It is stressed
that there is no division between the Jerusalem apostles and the
Antiochene missionaries. The reading of D (υμών, for ἡμῶν) might
suggest a different attitude (‘Dear to you, not necessarily so to us’),
but is probably a simple error. The two pronouns were identical in
pronunciation and were often confused. On the relation between Paul
and the Jerusalem church see Introduction, pp. xxxix-xli, et al.
26. Further description of Barnabas and Paul follows. They are
ἀνθρώποις παραδεδωκόσι τάς ψυχὰς αυτών. At first glance this
might be taken to mean, Men who have given their lives, but this is
obviously not the sense intended. Accordingly the suggestion has
been made that the meaning is, Men who have hazarded their lives;
so LS 1308, but with no other example of this meaning, which is
rightly judged by Begs. 4.180 to be ‘indefensible’. The phrase must
mean, They have devoted their lives (so apparently BA 1242,
‘Menschen, die ihr Leben für den Namen des Herrn hingegeben
haben'); and this, though followed in Greek by υπέρ, must in English
be followed by, to the name of our Lord Jesus Christ. Barnabas and
Paul are Christians of unreserved devotion. τάς ψυχάς αυτών is used
 39. COUNCIL IN JERUSALEM. 15.6-29. 743

as a reflexive. Cf. Thucydides 5.16.3, τύχῃ αυτόν παραδίδωσι, trust

oneself to fortune. StrB (2.740) seem to take the expression to mean
‘give one’s life for’, since they equate it with but Mekhilta
Exod. 20.6 (75b) which they quote seems rather to mean ‘devote’:
.
At the end of the verse D E 614 pc 1 syhmg add εις πάντα
πειρασμόν. This probably reflects the view that the text as it stood
implied, They have handed over their lives to death, which needed
correction to, They have handed over their lives to every (kind of)
trial. Cf. Sirach 2.1.

27. οὖν connects with v. 25: Since it was decided to send men with
Barnabas and Paul we have therefore now sent ... The perfect
ἀπεστάλκαμεν implies that Judas and Silas (see v. 22) are now on
their way, presumably bearing the letter. They are however not
merely bearers but will confirm its contents (τά αυτά, the same
things that the letter contains, especially the disavowing of the
trouble-makers). Haenchen (436) comments: ‘Der Leser weiss ja
schon, was das Dekret enthalten wird, die Briefempfänger aber noch
nicht. Man sieht, dass die Briefkomposition auf Lukas zurückgeht.’
The conclusion is probably correct (see above) but it is difficult to
follow the logic by which Haenchen reaches it. The hearers will take
in the letter as a whole and find that the hearers say the same things.
The construction of the second part of the sentence is obscure, διά
λόγου, standing in contrast with an implied δι’ επιστολής, is clear
enough; λόγος, though it means much more, can on occasion refer to
speech uttered aloud. But καί αυτούς ... ἀπαγγέλλοντας is strange
Greek, even when taken from the same standpoint as the perfect
ἀπεστάλκαμεν. BDR § 339.2a, n. 8 note that the present participle is
sometimes used instead of the final future participle, and cite in
addition to this verse not only 21.16 (συνῆλθον ... άγοντες) but also
Thucydides 7.25.7, ἔπεμψαν ... πρέσβεις οι Συρακόσιοι ...
αγγέλλοντας ..., they sent ambassadors to report ... See also
Euripides, Suppliants 154, ταῦτ’ ἐκδικάζων ἦλθον. This use of the
present participle is rare, and the sense of purpose seems usually to
be more or less qualified, but it may be taken to cover Luke’s
ἀπαγγέλλοντας. More difficult is καί αυτούς. This is reminiscent of
nothing in Greek so much as of the Hebrew circumstantial clause,
which can be introduced in this way, the clause being completed
often by a participle; see e.g. Gen. 18.8; 24.62. BDR § 442.4b, n. 12
(noting the Vg qui) rightly say that καί may be taken in the sense of a
relative, but this does not amount to an explanation. We have seen
reason to think that the letter Luke gives here is, apart from the
Decree, his own composition; otherwise it would be attractive to
think that he had retained a trace of the Hebrew in which such a letter
would originally have been composed. If not this, we may think that
744 COMMENTARY ON ACTS

Luke heard of the letter in a partly Hebrew-speaking environment. If

original at all it must have come from such an environment.

28. ἔδοξεν: cf. vv. 22,25. The apostles and elders are now joined in
their decision by the Holy Spirit. In view of the importance of the
decision this must be regarded as the outstanding example of Luke’s
insistence that all developments in the church’s life were directed by
the Spirit. For participation in a decision by a divine being cf. Dio
Chrysostom 80(30).8, ἔδοξε τω θεω; Appian, Syria 58.303, ἐδόκει
τοῖς θεοῖς. It is not suggested that a decision made by the church must
be a decision of the Holy Spirit; the unanimity of the church bore
witness to a decision already reached by the Holy Spirit. Luke does not
claim that any of the participants in the discussion was, when he spoke,
moved, or filled, by the Holy Spirit. Whether the claim that the Holy
Spirit on this occasion spoke through the church is affected by the fact
that Paul (it seems) did not accept the decision is a question that should
be borne in mind with regard to such ecclesiastical assertions.
It was not decided that no requirements should be laid upon Gentile
Christians, only that such requirements should be limited. Certain
demands were to be made, but nothing more, μηδέν πλέον. ‘Att
πλέον nur Lk. 3.13; Jn 21.15; Act 15.28 gegenüber 18mal πλεῖον,
aber immer πλείων, πλείονος’ (BDR § 30.2, n. 3). The word βάρος (a
natural one to use; cf. Polybius 1.315, τὸ βάρος των ἐπιταγμάτων,
but note ζυγός in v. 10) occurs in a similar clause at Rev. 2.24, οὐ
βάλλω (cf. ἐπιτίθεσθαι) ἐφ’ υμάς άλλο βάρος. The same letter (to
Thyatira) condemns the false prophetess Jezebel, who teaches the
Christians πορνεῦσαι καί φαγεῖν εἰδωλόθυτα: similarly 2.14. Noth-
ing is said about blood and non-kosher food, but it seems likely that
the Decree, or something like it, circulated in these churches and was
approved by them, or at least by leading members in them. This is a
matter on which commentators on Revelation do not agree. R. H.
Charles (1.74) thinks that the prohibition of non-kosher food and
blood was deliberately omitted. ‘The use of the word άλλο in itself
points to the exclusion of the two latter. Thus our author had clearly
the Apostolic decree in his mind.’ On the other hand, Lohmeyer
(HNT, Apokalypse 29) writes, ‘Die formale Uebereinstimmung mit
Act 15.29 ... beweist nicht, dass hier auch inhaltlich an das Apos-
teldekret zu denken sei.’ It is interesting to note F. J. A. Hort (31), on
βάρος: ‘A curious coincidence with Acts 15.28, but probably acci-
dental’; and Η. B. Swete (45), ‘A scarcely doubtful reference to the
Apostolic decree in Acts 15.28... the rest of the prohibitions imposed
in the years 49-50... are not reimposed’. This allusion to the Decree,
if such it be, is probably the earliest, antedating the post-biblical
references given in FS Osborn 55-9.
πλήν is used as an improper preposition, taking the genitive; cf.
8.1; 27.22, except.
 39. COUNCIL IN JERUSALEM. 15.6-29. 745

The exception is denoted by τούτων των έπάναγκες. The mean-

ing is clear; the words mean these necessary (requirements) and refer
to the four abstinences called for in the next verse. But the expression
is unusual: the demonstrative pronoun and article, followed by the
adverb (the neuter singular of the adjective έπανάγκης forming the
corresponding adverb). According to Haenchen (436), this is ‘ganz
ungewöhnlich’; similarly Begs. 4.180: the reading ‘can hardly be
right, έπάναγκες is used in Attic Greek as an adverb ... but not with
the article’. These statements call for some qualification. It is quite
possible in Greek to use an article with an adverb, but this
construction seems to be mainly if not exclusively confined to
adverbs of time and place; e.g. οί τότε, those who lived at that time;
οί έξω, those who are outside; τά κάτω, τά άνω, things below, things
above. The possibility of extending this use of article and adverb
must always have existed even if it was seldom taken up. Its
unnaturalness in the case of έπάναγκες is attested not only by
modern attempts at explanation and conjecture but also by ancient
variants in the text. It is uncertain whether έπ’ άνάγκαις (X A C ) is a
genuine variant; αι is a not uncommon itacistic equivalent to ε. If it is
intended as έπί with the dative plural of ανάγκη it will mean these
matters which arise out of necessities, των έπάναγκες τούτων (P33v,d
e ÏR) scarcely differs from the text cited above (though Metzger
(437) describes it as ‘the easier sequence’). If τούτων έπάναγκες
(omitting των) is read the way is open to repunctuation, with a colon
or full stop after τούτων: ‘... no other burden than these things: it is
necessary to abstain from ...’. There is support for this (see Begs.
4.180f.) in Clement of Alexandria, Stromateis 4.15, where indirect
speech is used: έμήνυσαν γάρ έπάναγκες άπέχεσθαι δειν ειδωλο-
θύτων, κτλ. ‘Moreover the Didascalia seems here to have felt that
somehow it ought to read πλήν τούτων τό έπάναγκες άπέχεσθαι
...’. The reading των έπάναγκες ((P74) A 36 453 1241 pc) Metzger
ascribes to accidental omission of τούτων.
It is probably best to accept the reading of X2 B C Ψ 81 614 945
1175 1739 2495 al (τούτων των έπ.) and the unusual construction
that it involves. But however the text is taken and construed it
includes the notion of necessity, compulsion (see LS 607; cf. e.g.
Josephus, Ani. 16.365; Epictetus 2.20.1; and see Wilson, Law 82). It
is not said, You Gentiles are completely free of legal requirements,
but as a matter of courtesy to your Jewish brothers you might be so
kind as to abstain from ... This is important for the understanding of
the Decree; Luke at any rate understood it as a matter not of courtesy
but of compulsion, and therefore presumably as a condition of
salvation (cf. 15.1, 5).
Bauemfeind (198) thinks that abstention from πνικτόν and αιμα
will have been a real burden to Gentiles, yet one that Luke will have
regarded as compatible with the law-free Gospel.
746 COMMENTARY ON ACTS

29. This verse repeats what is given in v. 20 as James’s opinion

with the following variations: (1) instead of ἀλισγήματα τῶν
ειδώλων we have εἰδωλόθυτα; (2) the order is idolatry, blood, things
strangled, fornication, instead of idolatry, fornication, what is stran-
gled, blood; (3) the plural πνικτῶν is used instead of the singular
πνικτοῦ; (4) there is a concluding exhortation. There are textual
variants similar to those noted in v. 20; see once more Metzger
(429-34). This time fornication is omitted only by some Vulgate
MSS.
40. PAUL AND BARNABAS RETURN TO ANTIOCH 15.30-35

(30)So they, when they had been dismissed, went down to Antioch, and
when they had gathered the company together they delivered the letter. (31)
When they had read it they rejoiced at the comfort1 that it brought. (32) Judas
and Silas, who also were prophets, speaking at length, encouraged2 and
strengthened the brothers. (33) After spending some time [in Antioch] they
were released3 by the brothers in peace [to return] to those who had sent
them. (35) But Paul and Barnabas stayed on in Antioch, teaching and
preaching4 the word of the Lord along with many others.

Bibliography
E. Delebecque, RHPhR 64 (1984), 47-52.
T. Holtz, as in (38).
B. N. Kaye, NovT 21 (1979), 13-26.
E. Richard, as in (38).

Commentary
This short paragraph is the necessary complement to the preceding
one. The results of the Council had to be communicated to the church
in which the problem arose, and its letter delivered. Nothing is said
here that could not have been inferred from the letter of 15.23-29
except that Judas and Silas were prophets. Luke may have been
dependent on information derived from Antioch, but he is probably
rounding off in his own words the important story of the Council and
preparing for Paul’s change of partners; Silas is introduced as one
who is acceptable both to Jerusalem and to Paul.

30. Luke develops his story into a new phase with μὲν οὖν; see on
1.6. They (the group mentioned in 15.23) then having been dis-
missed; the sense makes ‘those who had been dismissed’ impossible.
D* (1) add ἐν ἡμέραις ὀλίγαις; it is hard to see what the Western
editors gained by inserting these words, or what would have been
gained by omitting them if they were original; nor is it clear whether
1RSV, exhortation; NEB, NJB encouragement.
2RSV, exhorted.
3NEB, with the good wishes of the brethren.
4NJB, proclaimed the good news, the word of the Lord.
747
748 COMMENTARY ON ACTS

they refer to time spent before departure or to time taken on the

journey. The point may have been that the messengers did not delay
on the journey, not even stopping to preach in the towns through
which they passed. The general effect is to speed up the narrative.
κατῆλθον: they went down from the (Jewish) capital; cf. ἀναβαίνειν
in 15.2, with the note on the customary use of verbs of going up and
going down. They came to Antioch whence they, or some of them,
had been sent to represent the Antiochene point of view. For the city
of Antioch see on 11.19.
They gathered together the πλήθος, here without doubt (cf. 15.12)
the whole company of Christians in Antioch. Nothing is said of any
elders in Antioch; contrast Jerusalem (15.6). In 13.1-3 elders are not
mentioned; in the present paragraph there is no reference to prophets
and teachers (though local prophets may be implied in v. 32). They
delivered (ἐπέδωκαν) the letter—the task with which they had been
entrusted, ἐπιδιδόναι is frequently used for handing over a letter or
other document; see MM 238, BA 252 (with references including
Josephus, Ant, 15.170, ἐπιδίδωσιν Ηρώδη τήν ἐπιστολήν, as well as
papyri), ἀναδιδόναι was also used for the purpose: 23.33. It had also
been expected that the messengers would supplement the letter with
oral communication (15.27); see v. 32.

31. The subject of ἐχάρησαν is evidently the Antiochenes, and it is

to them also that the participle ἀναγνόντες will apply. It must
however be understood to mean that one or more of them read the
letter aloud in the hearing of the rest. All rejoiced; that is to say, they
felt that they had got what they wanted, παράκλησις has a variety of
meanings; it may be exhortation or encouragement (cf. 13.15); here
however (as the cause of rejoicing) it probably means comfort; they
were relieved that the leading Jewish Christians in Jerusalem had not
insisted that they should be circumcised. Cf. however the use of the
verb παρακαλεῖν in v. 32.
Had they got what they wanted? Had their objection to the
demands of the Judaizers (15.1, 5) been simply to the offensive and
painful rite of circumcision, with food laws an insignificant factor?
Or had they objected to the principle of legal requirements (however
slight and pointless) as a condition of their being accepted as
Christians, a condition of salvation? Or are 15.1, 5 misleading
statements, the real point at issue being not, On what terms may
Gentiles be saved? but, On what conditions may Gentile Christians
eat with Jewish Christians? Or again, are these historical questions
improperly asked? Can we ask more than, How did Luke view the
Decree? Here we can answer plainly: he regarded it as a
παράκλησις, no doubt in comparison with a demand for circumci-
sion and full observance of the Law. It must be remembered that
(according to Luke) Paul was one of the party who conveyed and
 40. PAUL AND BARNABAS RETURN TO ANTIOCH. 15.30-35. 749

presumably commended the letter. Luke is claiming that Gentile

Christianity is free from the Law and that the unity of the church had
been preserved. His view of the matter is certainly the first question
into which we must inquire, but since Luke chooses to write as a
historian we are free to ask further questions, and are bound to
observe the evidence of the Pauline epistles that there were those
who did not consider that Gentile Christians were free from the Law,
and that the unity of the church was at least gravely imperilled.
32. Judas and Silas (see 15.22) joined in the operation, καί αυτοί
προφῆται ὄντες; this is another circumstantial clause (cf. 15.27),
commented on by Black (AA 83). For prophets in Jerusalem see
11.27. καί (if not simply the relic of a Semitic circumstantial clause)
must mean also; in addition to Paul and Barnabas (who are not
usually described as prophets) or in addition to the native Antiochene
prophets? D adds πλήρεις πνεύματος ἁγίου (of course, for prophecy
is a gift of the Spirit—D underlines a theme already present in Acts)
and omits πολλοῦ (with λόγου). This may be no more than a case of
homoeoteleuton, or D may be affected by 15.27 and wish to bring out
the fact of oral addition to the written letter. But Blass 172 rightly
observes ‘Sed manente quoque πολλοῦ id fieri potest.' διά is of
attendant circumstances (Moule, IB 57). The speaking is described
by the word παρεκάλεσαν. In v. 31 the cognate noun παράκλησις
appeared to mean comfort; here exhorted, or perhaps encouraged,
seems a more natural meaning.
They exhorted the brothers; for αδελφοί as a word for Christians
see 1.15; again, as in v. 30, there is nothing to suggest a hierarchy of
ruling members in the church, ἐπιστηρίζειν is used again at 15.41; cf.
14.22. The metaphorical use of the verb requires no explanation.
33. For ποιεῖν χρόνον cf. 18.23; 20.3; also Prov. 13.23
(ποιήσουσιν ... ἔτη πολλά). The use is classical (Demosthenes
19.163 (392); Plato, Philebus 50d; and cf. the Latin facere, Cicero,
AdAtticum 5.20 (79)). Hebrew is used similarly (Jastrow 1125).
ποιήσαντες χρόνον here means nothing more definite than that the
envoys did not simply deliver the letter and return immediately. How
long they stayed before the Antiochene Christians let them go
(ἀπελύθησαν) on their return journey cannot be determined. They
left μετ’ εΙρήνης; not only on good terms with their hosts but in a
general situation of Christian well-being; cf. 9.31. The Christians in
Antioch are still, as in v. 32, brothers. The messengers returned to
those who had sent them, the apostles and elders of 15.22. Solidarity
with Jerusalem is affirmed (Davies, Land 276).
If this verse is read on its own it suggests that both Judas and Silas,
perhaps Paul and Barnabas too, returned at this point to Jerusalem.
This is clearly not Luke’s intention; in v. 35 he declares that Paul and
Barnabas stayed on in Antioch and v. 40 suggests without quite
750 COMMENTARY ON ACTS

proving that Silas stayed on too. This means either that the plural in
the present verse is a mistake, or that others, unnamed in 15.22,
accompanied the four who are named. See on v. 34.
Metzger (439) notes a variant not mentioned in NA26. Instead of
πρὸς τούς ἀποστείλαντας αυτούς (P74 A B C D vg sa bo al), H L
P S many minuscules syp h bo arm eth Bede have πρὀς τούς
αποστόλους, ‘... to bring the apostolate into greater prominence’. It
is just as likely to be a corruption of αποστειλανταςαυτους The
divided witness of the Bohairic version will be observed.

34. This verse is not in the text of NA26; it is omitted by P74 A B

Ε Ψ <a>vgst bo. It is probably a Western ‘improvement’ designed to
explain the presence of Silas in 15.40 after v. 33 had suggested his
departure from Antioch to Jerusalem. It occurs in several forms with
minor differences. All begin ἔδοξε δὲ τῷ Σίλα (or Σιλέα) ἐπιμειναι
(so C D 33 36 323 453 614 (945) 1175 1739 1891 al gig 1 w vgcl
syh** sa bomss). Next C has αυτού (it seemed good to Silas to stay
there; cf. 18.19). Dc gig 1 w vgcl have πρὸς αυτούς, to stay with them,
D* has αυτούς (it seemed good to Silas that they should stay—
unless αυτούς is a mistake for αυτού (Begs, 4.182)). After this D gig
1 w vgcl have μόνος δέ Ιούδας ἐπορεύθη, and w vgcl add further
hierosolyma.
The language of the verse is Lucan. ἔδοξε has been used three
times in the immediate context (15.22,25, 28; cf. Lk. 1.3); ἐπιμένειν
is a Lucan word (6 times, not counting this verse; 10 times in the rest
of the NT).

35. Paul and Barnabas did stay (διατρίβειν is another Lucan word:
8 times in Acts; twice (but once with the variant μένειν) in Jn;
nowhere else in the NT) in Antioch, joining with many others—
unnamed; we think of the prophets and teachers of 13.1, and of those
who first spoke the word in Antioch, 11.19—in teaching and
preaching (preaching as good news, εύαγγελιζόμενοι) the word of
the Lord. For teaching the word of the Lord see 18.11; for preaching
the word of the Lord see 8.4; for the word of the Lord see e.g. 8.25;
15.36.
 XI
PAUL’S MISSION BREAKS NEW GROUND
(15.36-18.23)

41. TERRITORY OF THE FIRST JOURNEY REVISITED

15.36-16.5

(36) After some days, Paul said to Barnabas, ‘Let us return1 and visit the
brothers from city to city, every one of them, in which we proclaimed the
word of the Lord, to see how they are faring.’ (37) Barnabas wished2 to take
with them also John, called Mark; (38) Paul however took the view that they
should not take with them this man who had parted from3 them in Pamphylia
and failed to accompany them to the work. (39) There arose a sharp
disagreement so that they separated from each other, and Barnabas took
Mark and sailed away to Cyprus, (40) while Paul chose Silas and went off,
committed by the brothers to the4 Lord’s grace. (41) He passed through Syria
and Cilicia, strengthening the churches.
(1) He now reached Derbe and Lystra; and there was a certain disciple
there whose name was Timothy, the son of a Jewish woman who was a
believer and of a father who was a Greek.5 (2) This disciple had a good
reputation with the brothers in Lystra and Iconium. (3) Paul wished him to
go with him, and took and circumcised him on account of the Jews who were
in those parts, for they all knew that his father had been6 a Greek.7 (4) As
they passed through the cities they delivered to the believers8 for their
observance the decrees that had been decided by the apostles and elders who
were in Jerusalem. (5) So the churches were confirmed in faith and increased
in number daily.

Bibliography
T. Baarda, NovT 34 (1992), 250-56.
P. Borgen, Paul Preaches Circumcision (1983), 33-7 (= FS Barrett,
37-46).
1NEB, Ought we not to go back now.
2NJB, suggested taking John Mark.
3NEG, NJB; deserted.
4NJB, grace of God.
5NEB, Gentile.
6NEB, was.
7NEB, Gentile.
8Greek, them, but the word cannot refer to cities.

751
752 COMMENTARY ON ACTS

C. Bryan, JBL 107 (1988), 292-4.

S. J. D. Cohen, JBL 105 (1986), 251-68.
D. Daube, Ancient Jewish Law (1981), 22-32.
E. Delebecque, as in (40).
R. Μ. Grant, VigCh 46 (1992), 105-111.
B. N. Kaye, as in (40).
G. D. Kilpatrick, JTS 41 (1990), 98.
Y. Tissot, as in (39).
A. Wainwright, JSNT 8 (1980), 66-70.
W. O. Walker, ExpT 92 (1981), 231-5.

Commentary
Paul proposes to Barnabas that they should revisit the Christian
brothers in all the cities in which they had preached in the journey
described in chs. 13 and 14. According to 14.23 elders had been
appointed in every church; that is, there were organized Christian
communities, brought into being by the initiative of the church in
Antioch. It was an elementary Christian duty to make sure of their
well-being. Barnabas agreed, but wished to take with them John
Mark, who had accompanied them on the earlier journey. Paul
disagreed, on the ground that John Mark was not a man to be trusted.
The disagreement led to sharp dissension and separation; Barnabas
and Mark sailed to Cyprus, Paul chose Silas as his colleague and
went off by land through Syria and Cilicia.
Approaching the old mission field in reverse order Paul and Silas
came to Derbe and Lystra, where they encountered Timothy, whom
Paul wished to add to his team. Since Timothy was of mixed
parentage and uncircumcised, Paul, because of the local Jews who
knew that Timothy’s father was a Greek (that is, a non-Jew),
circumcised him. The journey continued, the decrees of the Council
were disseminated, and the churches flourished.
The outline of the passage is thus straightforward, but it raises a
number of problems, which are stated and discussed in the notes. The
sensible intention of revisiting the churches of the first journey is
speedily given up; at 16.6 the party is breaking new ground. That
Paul and Barnabas should disagree about the wisdom of taking Mark
with them is understandable, but in Gal. 2.13 a different ground of
dispute is given. That this has no place in Acts must be taken with the
fact that Acts provides no hint of the message from James that split
the church at Antioch. Was Luke aware of this? Did he deliberately
omit it and find himself obliged to produce a different reason for the
separation of Barnabas from Paul? The question of the authorship of
Acts is thus raised in acute form. It is of course possible for two men
 41. TERRITORY OF THE FIRST JOURNEY REVISITED. 15.36-16.5. 753

to be in dispute on more issues than one; and these two issues may
have been related to each other. Why did Mark turn back on the first
journey? One possible answer is that the mission was turning
increasingly to the Gentiles, and he disapproved; he may have carried
Barnabas with him to the extent that in Antioch Barnabas sided with
Peter rather than with Paul. This however does not alter the fact that
serious questions of authorship are raised not only by Luke’s account
of the Council itself but also by his story of the subsequent events.
A second serious problem arises in 16.1-3. The Council had
decided that it was not necessary for Gentile converts to Christianity
to be circumcised. It was a notable victory for Paul’s opinion. Very
shortly afterwards Paul, who had (according to Gal. 2.3) vigorously
resisted the attempt to have Titus circumcised, himself circumcised
Timothy. It is just credible that Paul did this, but only just. He might
have reasoned: As he is, Timothy is neither Jew nor Gentile; we
cannot undo the Jewishness of his mother; we can give him the
circumcision that his Gentile father would not permit. Thus all
ambiguity is resolved. This is possible; not probable. The same may
be said of other suggestions; see below. If it is rejected, if Luke has
given us at best a partial account of Paul’s break with Barnabas and
an incorrect account of his dealings with Timothy, the traditional
authorship of Acts by one of Paul’s circle is called in question, and
the best we can say of this paragraph is that Luke was dependent on
popular stories, which he did not, perhaps could not, check with Paul,
with Barnabas, with Timothy, or with the church in Antioch.

36. The process of teaching and evangelizing (15.35) lasted some

time—τινάς ἡμέρας; Luke does not know how long but means to
suggest a serious operation. Paul now takes the lead (εἶπεν πρὸς
Βαρναβᾶν) with a new suggestion: he proposes that they should
revisit the areas in which they had preached in chs. 13 and 14. There
are αδελφοί there; the word is used of Christians at 14.2 (not 13.15,
26, 28), frequently elsewhere in Acts. It will be a good thing to
see how they are faring, πώς ἔχουσιν; for ἔχειν with an adverb
(here interrogative) see on 7.1. Nowhere else in the NT does
ἐπισκέπτεσθαι, introduce an indirect question, but the word means
visit in the sense of review, and the question follows naturally,
though an extra word is needed in English (to see how...). Haenchen
(456) suggests that ἐπισκεψώμεθα will have made Luke’s readers
think of file word ἐπίσκοπος (20.28); it is doubtful whether the
elders (equivalent to bishops in ch. 20) engaged in wide-ranging
visitations.
κατά is used distributively (cf. 15.21); πάσαν must be emphatic—
from city to city, every one of them. Paul proposes a comprehensive
visitation of every city in which (ἐν αἷς, the plural understood ad
sensum, BDR 296.1, n. 1; D has ἐν οἷς, which must refer to τούς
754 COMMENTARY ON ACTS

αδελφούς, among whom) we proclaimed (κατηγγείλαμεν: 11 times

in Acts, 7 times in the rest of the NT ) the word of the Lord (see
15.35).
δή adds some emphasis: Come, let us return ... (Begs. 4.183)
There is obvious contact with 14.28 but this does not in itself mean
that the intervening material is simply an episodic addition. From the
literary point of view it may well be so, but it is possible to think of
the second visit as taking place in the light of the Council, and
therefore on a firmer, more established footing than the earlier, more
or less experimental mission. This at least will have been Luke’s
view.
There is a sense in which Roloff (236) is right in the view that this
paragraph, with Galatians 2, marks the end of Paul’s close associa-
tion with Antioch. It is probable that there was a change in Paul’s
association with Antioch, but the present paragraph does nothing to
suggest it; what it suggests is an effort to maintain contact between
Antioch and the daughter churches.

37. Βαρναβάς δέ ἐβούλετο. Turner (Insights 95) stresses the

imperfect tense, which he would translate ‘had half a mind to’. But a
half-hearted suggestion of this kind could not have given rise to the
παροξυσμός and separation of v. 39. Rather more convincing, but
still to be accepted only with caution, is the distinction between the
aorist infinitive συμπαραλαβείν here (but P74 A 1175 pc have the
present, συμπαραλαμβάνειν) and the present in v. 38. So Μ. 1.130:
‘Barnabas, with easy forgetfulness of risk, wishes συμπαραλαβεῖν
Mark—Paul refuses συμπαραλαμβάνειν, to have with them day by
day one who had shown himself unreliable.’ Zerwick (§ 249) agrees,
referring to Moulton. Μ. 3.79 looks at the matter in a similar light:
‘All Paul may have objected to was παραλαμβάνειν (Mark’s being
with them throughout the journey).' This, like Turner’s other attempt
to minimize the conflict between Paul and Barnabas, is unconvinc-
ing; there is in the story no hint that there might have been an
arrangement by which Mark would turn back half-way round.
καἰ τόν ( B 81 614 pc) is probably correct. P74 A C E Ψ 36 945
1175 1241 2495 pm have only καί; L 424 have only τόν; D 323 1739
1891 al 1 omit both words. But καί is natural—Not only me, John
too; τόν is anaphoric and is needed to point back to the earlier
references to John.
John is referred to incidentally in 12.12 to identify his mother. He
appears in the company of Barnabas (to whom according to Col. 4.10
he was related) and Paul at 12.25, and again, on the ‘First Missionary
Journey’, at 13.5. From Perge, however, he returned (not to Antioch
but) to Jerusalem, leaving the two senior partners to continue the
mission without him. See the notes on 13.13; 15.38; see also Col.
4.10; 2 Tim. 4.11; Philemon 24.
 4L TERRITORY OF THE FIRST JOURNEY REVISITED. 15.36-16.5. 755

38. For John Mark’s defection on the earlier journey, and for
Pamphylia, see on 13.13.
ἀξιούν in the sense of ‘to count something, especially a course of
action, right’ is in the NT peculiar to Acts (15.38; 28.22; contrast Lk.
7.7), but the usage is classical (LS s.v. III 2, p. 172). Μ. 3.65 thinks
that the imperfect tense ‘may mean that Paul’s suggestion about
Mark was only tentative at first’; contrast BA 156, ‘er bestand darauf
(Ipf), diesen nicht mitzunehmen’. The latter seems the more proba-
ble view; there is no hint in the narrative of a progressive hardening
of opinion.
ἀποστάντα suggests (as ἀποχωρήσας in 13.13. does not) the
blameworthiness of Mark’s departure (cf. e.g. Xenophon, Anabasis
2.5.7, εις ἐχυρόν χωρίον ἀποσταίη, to withdraw from battle), and
this is borne out by the following words: John Mark failed to
accompany Paul and Barnabas εἰς τό ἔργον, that is, the work of
evangelizing the interior of Asia Minor. This showed that he was
not to be trusted, and Paul accordingly thought it right μή
συμπαραλαμβάνειν (cf. v. 37) τούτον, the pronoun taking up τόν
ἀποστάντα ... καὶ μή συνελθόντα. The pronoun is unnecessary, and
was omitted by P45 gig vgcl.
D rewrites the verse without making any significant change: Π. δέ
οὐκ ἐβούλετο λέγων τόν ἀποστάντα ... καί μή συνελθόντα (om.
αὐτοῖς) εἰς τό ἔργον εις δ ἐπέμφθησαν (a natural but unnecessary
description of the work in question) τούτον μή είναι συν αὐτοῖς.
The addition of εις ὅ ἐπέμφ. is readily explicable in terms of the
tendency of D and its allies to pious expansion; the variants at the
beginning and end of the verse are not so explicable, and may
suggest that Western editors were working on a different base. The
additions ‘considerably weaken the force of the B-text’ (Metzger
439).
For possible significance of the tenses of συμπαραλαμβάνειν see
on v. 37.
A different view-point is given in the certainly fictional Acta
Bamabae 8 (L.-B. 2.2.294): πολλή τοίνυν φιλονεικία μεταξύ αυτών
ἐγένετο. Βαρναβάς δέ παρεκάλει κἀμέ συνακολουθῆσαι αὐτοῖς
... ὁ δέ Παύλος κατέκραζεν τού Βαρναβά ...
Pesch (2.93) thinks it possible that Mark may have been responsi-
ble for the conflict in Antioch by informing James of what had
happened in chs. 13 and 14. This is pure guesswork, but it could be
true.

39. The disagreement between Paul and Barnabas was sharp;

exactly how sharp it is not easy to say. The only other occurrence of
παροξυσμός in the NT has a good connotation: Christians are to
provoke one another to love and good works (Heb. 10.24). It is
possible however that this good use hints at a bad one: You are in
756 COMMENTARY ON ACTS

fact provoking one another to anger; enough of this! ‘Provoke’ one

another to something different. The cognate verb παροξύνεσθαι also
has different connotations: at 1 Cor. 13.5 Paul says that love οὑ
παροξύνεται; at Acts 17.16 it is said that Paul’s spirit παρωξύνετο
ἐν αὐτω at the sight of Athenian idolatry. Here it seems that relations
between Paul and Barnabas were embittered; how far the bitterness
was expressed in outbursts of anger is not clear. The result was clear;
the partnership that is described in chs. 13 and 14 was broken and the
two separated (ἀποχωρισθήναι; in 13.13 John Mark separated,
ἀποχωρήσας, from Paul and Barnabas) from each other. BDR §
391.2, n. 6 write, ‘In einem Satz wie Apg 15.39 ... würde ein Attiker
wegen der mangelnden engen Verknüpfung und wegen des
Gewichtes, das auf die eingetretene Folge fallt, eher den Ind. gesetzt
haben.’ Haenchen (457) would also prefer the indicative. This
probably states correctly the Attic distinction between infinitive and
indicative (see Μ. 1.209f.), but Luke’s sentence leaves no doubt
about the connection between the quarrel and the separation, and the
importance of the result. No other cause is given for the separation
(but see below), and Luke does nothing to suggest that the outcome
of Paul’s future missions would have been different if he had been
accompanied by Barnabas and Mark rather than by Silas and
Timothy.
Barnabas took (παραλαμβάνειν; cf. συμπαραλ. in vv. 37, 38)
Mark and they sailed away (έκπλεῖν occurs again at 18.18;
20.6—there is a good deal of sailing in the second part of Acts) to
Cyprus, revisiting this part of the field covered in chs. 13 and 14, and
leaving the rest to Paul (v. 41). Barnabas was originally from Cyprus
(4.36); it would however be unfair to suggest a less bold and
adventurous attitude on his part. Paul made for Cilicia—of which the
capital was Tarsus.
A different ground of controversy, also located in Antioch, is
given is Gal. 2.13. When Peter withdrew from table fellowship with
Gentile Christians in Antioch even Barnabas was carried away by his
hypocrisy, and though only Peter is addressed in the words quoted in
Gal. 2.14-18(21) they would apply equally to Barnabas. Nothing is
said here about Mark. It may be that there were two distinct quarrels
between Paul and Barnabas; it may be that there was one quarrel that
had two elements; it may be that Paul gives the true ground of the
separation (Paul had little to do with Antioch after this point) and that
Luke preferred to find a personal rather than a theological ground for
the split. If the disagreement about Mark was a pure invention on
Luke’s part it was very cunningly worked out, the ground being
prepared in ch. 13, and to some extent in ch. 12. There is no reason to
think that the story about Mark is totally untrue, for there is no reason
why it should have been invented. Luke could have remained silent
about the trouble described in Galatians 2, simply saying that, after a
 4L TERRITORY OF THE FIRST JOURNEY REVISITED. 1536-16.5. 757

period of shared ministry (v. 36), Barnabas accompanied by Mark

decided to go to Cyprus, and Paul, accompanied by Silas, to go to
Syria and Cilicia. 1 Cor. 9.6 shows that Paul continued to recognize
Barnabas as a colleague, and Col. 4.10; 2 Tim. 4.11; Philemon 24,
whether written by Paul or not, show that at a somewhat later date
Mark was understood to be a member of the Pauline circle.
D (gig pc) rewrite part of the verse as follows: τότε Βαρναβάς
παραλαβών τόν Μάρκον ἔπλευσεν. The best explanation of this
reading is that a copyist copied τόν τε as τότε, accidentally omitting
the v, and then rewrote the rest of the sentence as economically as
possible. One could equally suggest that τότε was erroneously
written τόν τε.

40. Paul was now left without a travelling companion, and it

would have been foolhardy to set out alone on the kind of journey (to
be inferred from the epistles as well as Acts) that he had in mind. He
must choose one from those available, and he chose (ἐπιλέγεσθαι is
used in a different sense at Jn 5.2) Silas, on whom see the note on
15.22. It is pure conjecture, but it may be correct, that Paul set out on
a new mission with a new companion, because he had lost his battle
at Antioch (Gal. 2.1-14). This consideration could well have
contributed to his change of companion (see above), but Paul needed
no additional motivation to continue his work for the Gospel (1 Cor.
9.16, 23). He began his new mission in what might be described as
the Antiochene mission field.
Paul set out, commended by the brothers (singular, παραδοθείς,
but it is impossible not to think that Silas was included) to the grace
of the Lord. Cf. 15.33, and especially 14.26, παραδεδόμενοι τῆ
χάριτι τοῦ θεοῦ. In the present passage θεού is read by P45 C Ε Ψ <a>
gig w vgcl sy bo; κυρίου by P74 A B D (*) 33 81 pc d vgst sa. θεού was
no doubt introduced through assimilation to 14.26. The parallel is
instructive. Luke does not say, ‘The Lord (Jesus) is God’, but he
writes in such a way as to suggest the possibility to his reader, if not
to himself.
According to Luke, Paul set out with the warm approval of the
church; he does not say so much of Barnabas.
Lüdemann 175 observes that difficulties regarding the movements
of Silas are resolved if 15.22, 27, 32f. may be regarded as redac-
tional, v. 40 as traditional. Or (he asks), is there simply negligence on
Luke’s part? A further possibility is that the negligence belonged to
the tradition as Luke collected it, perhaps not all from one source of
information. Thirty years on, who could remember precisely the
movements of a secondary character?

41. Paul (no doubt we should think of him as accompanied by

Silas, but the verbs, διήρχετο ... ἐπιοτηρίζων, are singular) passed
758 COMMENTARY ON ACTS

through Syria and Cilicia. In the city of Antioch Paul was in Syria,
and set out northwards. Cilicia may be roughly described as occupy-
ing the northeastern angle of the Mediterranean, joining Syria to the
south east (and for the most part administratively united with it; see
6.9; 15.23) and stretching to the west perhaps as far as Pamphylia
and thus adjoining the territory of Paul’s first journey. Syria and
Cilicia were thus new territory as far as missionary work recorded in
Acts is concerned. In 9.30 Paul went from Jerusalem to Tarsus, and
therefore presumably passed through Syria and Cilicia; cf. Gal. 1.21.
Whether at that time he preached in Syria and Cilicia we do not
know, but according to the present verses ἐκκλησίαι had been
established by someone, since they were now strengthened
(ἐπιστηρίζειν; on this word see 15.32). This is confirmed by the
address of the Council’s letter, to Antioch, Syria, and Cilicia. This
may in fact be the source of the present reference to Syria and Cilicia.
Someone had to take the letter there and it might be assumed that
Paul did; cf. Conzelmann (88). It is not clear how Preuschen (98) can
regard this part of the journey as a duplicate of chs. 13-14.
For διήρχετο as possibly referring to a preaching mission see 13.6.
Apart from this nothing is said about Paul’s work in Syria and Cilicia
beyond the vague and general ἐπιοτηρίζων τἀς ἐκκλησίας, so that
the contrast (BDR § 327.1, n. 1) between 15.41 (διήρχετο, imper-
fect) and 16.6 (διῆλθεν, aorist) seems somewhat contrived.
την Συρίαν καὶ τήν Κιλικίαν is the reading of B D Ψ 36 453 pc.
A C E <a> omit the second article; P45 has διά τής Συρίας καί τής
Κιλικίας. The omission of the second article may reflect the time (see
on 15.23) when Syria and Cilicia were united as one province. At the
end of the verse D (gig w vgcl syhmg) add παραδιδοὺς τὰς ἐντολάς
τῶν πρεσβυτέρων, forgetting the apostles, but otherwise tying in the
new journey with the Council and its Decree.

1. The new subparagraph makes a new beginning, probably the

beginning of a new source of information. The reference to Syria and
Cilicia in v. 41 was probably based on the address of the letter
(15.23); if Luke had any further information about the journey
through these areas he chose not to give it. Probably he had none. But
he not only recalls Derbe and Lystra from the first journey; he has an
important new fact. His narrative is more probably based on local
information than on a written source.
The Western text, represented by D (gig syhmg), connects this verse
more closely with the previous one: διελθὠν (borrowed from 16.6?)
δέ τά έθνη ταυτα κατήντησεν. Even if διέρχεσθαι has the sense (see
13.6) of conducting a preaching tour, τά έθνη is a surprising
expression. It recalls the use of the word as equivalent to the Latin
provincia (LS, s.v. 2c, p. 480). This is relatively late (Appian,
Herodian, Dio Cassius, Dio Chrysostom are cited, with papyri of
 41. TERRITORY OF THE FIRST JOURNEY REVISITED. 15.36-16.5. 759

similar date—see also MM 181), but possibly relevant if the Western

text is a second century revision. In both forms of the text the
singular is used; the writer—or his source, information found in
Derbe and Lystra—is interested in Paul, not in Silas. After δέ, καί
should probably be read, with P45 A B Ψ 33 453 614 1175 1739 2495
al syh. After he had passed through these provinces he reached
also ... For this use of καταντάν εις cf. 2 Macc 4.21, 24, 44.
Καί ιδού recalls the Hebrew but Luke is not translating; he is
using his ‘Semitic’, biblical style as he introduces a story reminiscent
of the OT.
Timothy is to play an important part in Acts (16.1; 17.14,15; 18.5;
19.22; 20.4) and in the Pauline letters (Rom. 16.21; 1 Cor. 4.17;
16.10; 2 Cor. 1.1, 19; Phil. 1.1; 2.19; Col. 1.1; 1 Thess. 1.1; 3.2, 6; 2
Thess. 1.1; 1 Tim. 1.2, 18; 6.20; 2 Tim. 1.2; Philemon 1; cf. Heb.
13.23). The name occurs frequently in Greek from the time of
Aristophanes. He was a disciple (μαθητής, by which Luke seems
always to mean Christian; see on 6.1; 9.36, 38, here reinforced by
αδελφοί in v. 2; cf. πιστής, below). Nothing is said of him on the
occasion of Paul’s first visit to this area (14.8-20, 21); one may
guess therefore that he had not been converted by Paul, though Paul
evidently thought well of him, and in 1 Cor. 4.17 speaks of him as his
child.
Timothy’s mother too was a Christian (according to 2 Tim. 1.5 her
name was Eunice); his father was not ("Ελλην here means simply
non-Jew, not specifically Greek). Kosmala (425) argues that πιστής,
believing, does not necessarily mean that Eunice was a Christian.
The argument is unconvincing. The wording of the present verse
suggests that he came from Derbe.
Timothy is described as of mixed parentage. His mother (notwith-
standing the Greek name she bears in 2 Tim. 1.5) was a Jewess, his
father a Greek. Their marriage therefore was in Jewish law illegal
(see StrB 2.741, quoting Jebamoth 45b). The same passage makes
the point that, in mixed marriages, the child followed the nationality
of the mother. Preuschen (99) quotes as an analogy Cicero, De
Natura Deorum 3.18(45), ut enim iure civili qui est matre libera liber
est. From this it would follow that Timothy would be regarded as an
Israelite, and should in accordance with the law have been circum-
cised on the eighth day after birth. Of course, a Gentile father might
very well object to the rite. This understanding of the position,
however, was called in question by D. Daube (Ancient Jewish Law
(Leiden, 1981), 22-32). who showed that the principle of matrilineal
descent was not in operation in the first century; it was brought in in
the second century (out of concern for Jewish women who bore
children as a result of rape by heathen soldiers, who naturally
were not present to claim and bring up their children). This con-
clusion was—rightly—accepted by S. J. D. Cohen (JBL 105 (1986),
760 COMMENTARY ON ACTS

251-68) and C. Bryan (JBL 107 (1988), 292-94); see below on v. 3

for the bearing of this on Paul’s circumcision of Timothy.
In the majority of MSS Timothy’s mother is described as a
Jewess—104 (pc) have χήρας Ίουδαίας, which is probably a
conflate reading, and gig p vgmss have viduae. This may be an
inference from ὑπήρχεν (v. 3; see the note), or possibly (Begs. 4.184)
due to confusion in the Latin texts in question of iudeae and
uiduae. If she was a Gentile widow (reading uiduae) the story of
Timothy’s circumcision takes on a new significance; but this is most
improbable.
Silas now disappears from the story till 16.19; Timothy (apart
from the next two verses) till 17.14.

2. μαρτυρεῖσθαι in the sense of having a good reputation with is

common in Acts (6.3; 10.22; 16.2; 22.12). Timothy was well spoken
of by the brothers, that is, the Christians (for αδελφοί see on 1.15) in
Lystra (presumably in Derbe also, if this was Timothy’s home; see v.
1) and Iconium. The three towns were not (in the days of foot travel)
very close together; see on 13.51; 14.6. The Christians in each (no
doubt relatively few) may however have found it rewarding to keep
in touch with one another. Luke is locating the incident regarding
Timothy in the area of the ‘first journey’.

3. Paul wished him [Timothy] to go with him. 1 Thess. 1.1; 3.2, 6

are sufficient to show that a man called Timothy accompanied Paul
on the ‘second missionary journey’ and assisted in a mission that
included Macedonia. The epistles cannot be said to confirm the
statement that Paul selected him for the purpose from among the
Christians of Lystra (or Derbe) and wished him to set out with him
from there; equally there is nothing to cast doubt on the statement.
The existence of the two epistles to Timothy, though these were not
written by Paul, shows that Timothy was known to have survived
Paul, and if the author of Acts took seriously the task of finding out
what he could about Paul’s ministry Timothy would have been a
useful and probably available source of information. The first seven
words of the present verse give rise to no difficulty; the same cannot
be said of the rest of the verse.
Paul had just been contending, on behalf of the church of Antioch,
for Gentile freedom from the Law, specifically the law of circumci-
sion, and had elicited the decision that circumcision was not
required. The epistles make it clear that the contention had been
sharper than Acts allows us to see. ‘If you are circumcised, Christ
will do you no good at all’ (Gal. 5.2). ‘If anyone was called in a state
of circumcision, let him not undo it; if anyone was called in a state of
uncircumcision, let him not be circumcised’ (1 Cor. 7.18). He had
resisted attempts in Jerusalem to have Titus circumcised (Gal. 2.3-5;
 41. TERRITORY OF THE FIRST JOURNEY REVISITED. 15.36-16.5. 761

for the text and interpretation see Freedom and Obligation, p. 112. n.
12). Is it likely that Paul would, apparently without pressure,
circumcise his intended companion?
That he did so is stated in Acts with some emphasis: λαβών
περιέτεμεν. This is claimed by Black (AA 125, following up a
suggestion by J. Jeremias; see also Wilcox, 125, and cf. 9.25; 27.35)
as a Semitism; in both Hebrew and Aramaic take is
used in a superfluous way, doing no more than accompany and
perhaps emphasize the action of the significant verb. This is not
convincing; Paul took Timothy as an assistant and travelling com-
panion, and circumcised him. This is plain Greek, and it is quite
clear; there is no need to conjecture Semitic influence. Paul did this
διὰ τούς Ιουδαίους τούς όντας ἐν τοῖς τόποις ἐκείνοις. These must
be the Jews resident in the Derbe-Lystra area, not those in Jerusalem.
It is true that the Jerusalem Jews might have made trouble for Paul if
they had known that he was travelling with an uncir-
cumcised half-Jew, but only the local Jews would have known the
facts about Timothy’s parentage (ᾔδεισαν γἀρ ἄπαντες ... ). It is
argued that though Paul believed that circumcision was not necessary
for salvation—and that if treated as if it were necessary would make
salvation impossible—he might have circumcised his companion
because he was proposing in his mission to make use of Jewish
synagogues and places of prayer (16.13; 17.1; etc.) and could not
have taken an uncircumcised man with him; this was part of being to
the Jews as if he were a Jew (1 Cor. 9.20). But it is rightly pointed out
(Schmithals 145) that Paul is not said to have taken Timothy into a
synagogue, and that in any case Gentiles, favourable to Judaism but
not circumcised as proselytes, seem to have been admitted to
synagogues (see on 10.2). Again, it is argued that Gal. 5.11 bears
witness to a current opinion that Paul had, in at least one notable
case, practised circumcision. P. Borgen (Paul Preaches Circumci-
sion and Pleases Men (1983), 33-37) contends that this means that
Paul preached ethical circumcision (circumcision of the heart), and
his opponents held that this needed physical circumcision to make it
complete. Even if this is correct as an explanation of the difficult
verse in Galatians (on which see now in addition T. Baarda, in NovT
34 (1992), 250-6) it will hardly justify, or make credible, the
statement in Acts. It is hard to contradict the judgement of Born-
kamm (4.159), ‘So wird man die Notiz Apg 16.3 anfechten und aus
der zu Genüge bekannten Tendenz des Lukas erklären müssen, die
Loyalität des Paulus gegenüber dem Gesetz im Gegensatz zu allen
anders lautenden Vorwürfen zu beweisen.’
The reason given for Paul’s alleged action would be under-
standable if it could be assumed (see on v. 1) that Timothy was
legally a Jew by descent. Local Jews who knew that Timothy’s father
had been (ύπήρχεν probably implies that he was now dead; had he
762 COMMENTARY ON ACTS

been alive the verb would have been υπάρχει—see Zerwick § 346,
n.1; BDR § 330, n. 2) a Gentile would probably suspect if they did
not know that circumcision had been omitted on the eighth day, and
would certainly—and not unreasonably—make trouble if Paul tried
to pass him off as a Jew. The answer to this would be not to treat him
or present him as a Jew. This would have been the right kind of tact,
not the wrong kind, which Paul carefully avoided in the case of Titus
(Gal. 2.3). It does not seem likely that Luke simply invented the
incident; he had heard it, possibly guessed it on the basis of
Timothy’s sympathy with Judaism. And a bare possibility must
remain that Luke was right; that Paul, confronted with a companion
who was half Gentile and half Jew, decided to ‘make an honest Jew
of him’; ‘neque salutis aetemae causa T. circumciditur, sed utilitatis’
(Blass 175). No other explanation of Luke’s account is at all
convincing, though Bauemfeind’s (204) epigram is worth consider-
ing: ‘Er [Paulus] wollte christliche Juden und nicht entwurzelte
Juden.’ So is Schneider’s quotation (2.200) of Baumgarten, writing
in 1889: ‘Man vergisst, dass man bei solchem Eifer für die Freiheit
Freiheit selber wieder in eine Knechtschaft verwandelt.’ Pesch (2.99)
writes similarly that Paul’s freedom would be limited if he could not
in such a way as this accommodate Jews. Lüdemann’s suggestion
(183), that Acts is right in saying that Timothy was circumcised but
wrong in the timing—Timothy was circumcised on conversion,
makes matters more rather than less difficult. There was no point in
mentioning the nationality of Timothy’s mother (v. 1) unless it was
thought (by Luke if not by Paul) that this, if it did not determine legal
descent, nevertheless so strongly suggested Judaism as to make it
desirable to treat Timothy as a Jew. But would Paul have accepted
and acted on such a desirability? Bornkamm’s negative answer is
almost certainly right. As Schille (333) points out, circumcision that
had nothing to do with gaining salvation would be senseless in
Jewish thought at the time. Luke (not Paul) may have been preparing
for 21.21—the charge is disproved before it is brought (Johnson
290).
Weiser (400) reaches the conclusion that ‘Lukas in den Versen lb,
3a, eine Einzelepisode aus der Timotheus-Überlieferung aufgenom-
men und die Verse 1a, 2, 3bc, 4, 5 redaktionell gebildet hat.’
Concluding himself that the circumcision did not happen, Weiser
gives on 402 careful lists of those who agree and those who disagree.
To those who think that Timothy was circumcised by Paul must now
be added Trebilco (23).
The Western text continues its rewriting in this verse, and indeed
has some claim to originality. P74 N A B C Ψ 33 36 81 945 1175
1739 al vg co have ᾔδεισαν γάρ ἅπαντες ὅτι Έλλην ό πατήρ αυτού
ὑπῆρχεν. But P45vid D E <a> (gig) sy have ᾔδ. γάρ ἅπ. τόν πατέρα
αυτού (614 2495 pc have a different order) ὅτι Έλλην ὺπ. This
 41. TERRITORY OF THE FIRST JOURNEY REVISITED. 1536-16.5. 763

anticipation of the subject of ὑπῆρχεν as the object of ἤδεισαν is

Semitic, and could be original.
4. Temporal ώς is frequent in Lk. and Acts, see BA 1792. With the
imperfect διεπορεύοντο it means while they were passing through,
as they passed through. On their way, in town after town, they
delivered the decrees. παρεδίδοσαν is the ‘correct’ third person
plural imperfect of (παρα)διδόναι; contrast 27.1, παρεδίδουν; see
Μ. 2.202. φυλάσσειν is an infinitive of purpose, so obvious that it
hardly needed to be made explicit. The decrees were for keeping, not
for mere interest. They have now been taken outside the area
specified in 15.23 to the churches of the first journey. It is clear that,
in Luke’s view, they were intended to be ‘applicable to all Gentile
Churches’ (Wilson, Law 81; cf. Haenchen 445).
δόγματα is the noun that naturally corresponds with the verb
ἔδοξε of 15.28 (cf. 15.22, 25). It is used of orders laid down by a
duly constituted person or group. Thus δόγματα συγκλήτου (Poly-
bius 6.13.2) are Senatus consulta; cf. τά τῶν Άμφικτυόνων
δόγματα (Demosthenes 5.19(62)). See ND 4.146 and cf. Lk. 2.1;
Acts 17.7. At 3 Macc. 1.3; Josephus, Apion 1.42 (θεοῦ δόγματα) it is
used of the Jewish Law. The word does not suggest a request for
considerate behaviour, aimed at making possible harmonious com-
mon meals at which both Jewish and Gentile Christians might unite,
rather ordinances that must be observed by those who wish to be
members of the group that ordains them. With this τά κεκριμένα (cf.
James’s ἐγώ κρίνω, 15.19) agrees. Cf. Epictetus 2.15.7, τοῖς
κριθεῖσιν ἐμμένειν δεῖ.
The verse is rewritten by D syhmg Ephraim: διερχόμενοι δέ τἀς
πόλεις ἐκήρυσσον (4- καὶ παρεδίδοσαν αὐτοῖς, D) μετά πάσης
παρρησίας τόν κύριον 'Iησοῦν άμα παραδιδόντες καί τάς ἐντολάς
αποστόλων καί ... The intention may have been to prepare for
v. 5.
Begs. 4.185 comments: ‘Whatever may have been the facts Luke
obviously wishes to represent Paul as the delegate of the apostles in
Jerusalem in a manner which is incompatible with the Epistle to the
Galatians.’ This seems to be a valid observation.
It was from the apostles and elders that the decrees originated; cf.
15.2, 6, 23. Nothing is said now of the πλήθος (15.12) or the
ἐκκλησία (15.22).
5. This verse is a summary, reminiscent of the summaries that
punctuate the earlier parts of Acts (2.42-47; 4.32-35; 5.12-16; 6.7;
9.31). The plural αἱ ἐκκλησίαι, the various local churches, recalls the
Western reading at 9.31. That they were strengthened, made firm in
[the] faith recalls 14.22, the increase in numbers 6.7 (ἐπληθύνετο ὁ
ἀριθμός), and καθ’ ημέραν 2.47. τῆ πίστει is omitted by D.
This verse should probably be regarded as Luke’s conclusion of the
764 COMMENTARY ON ACTS

paragraph that began at 15.1. As on other occasions (notably ch. 6), we

see the emergence of a problem, its speedy and effective solution, and
its outcome in greater progress and expansion for the churches. In this
way Luke affirms the truth of the Gospel, demonstrated by the fact that
God so evidently cares for and furthers its expansion that believers
increase in numbers and believe more firmly.
 42. GUIDED BY THE SPIRIT TO TROAS 16.6-10

(6) They passed through Phrygia and Galatian territory, having been
forbidden1 by the Holy Spirit to speak the word in Asia; (7) they came
opposite Mysia and tried to go into Bithynia, but the Spirit of Jesus did not
permit them to do so. (8) They arrived at2 Mysia and came down to Troas.
(9) A vision appeared to Paul in the night: a man of Macedonia was standing,
begging him with the words, ‘Come across into Macedonia and help us.’ (10)
When Paul had seen the vision, immediately we sought to leave for3
Macedonia, concluding that God had called us to evangelize them.

Bibliography
W. P. Bowers, JTS 30 (1979), 507-11.
F. F. Bruce, BJRL61 (1978/9), 337-54.
O. Glombitza, ZNW 53 (1962), 77-82.
R. Μ. Grant, as in (41).
C. J. Hemer, Tyndale Bulletin 26 (1975), 79-112.
C. J. Hemer, JTS 27 (1976), 122-6.
C. J. Hemer, JTS 28 (1977), 99-101.
G. Μ. Lee, NovT 9 (1967), 41f.
G. Μ. Lee, Bib 51 (1970), 235-7.
G. Μ. Lee, NovT 17 (1975), 199.
G. Stählin, FS Moule, 229-52.

Commentary
The journey begun in 15.41 reaches Troas by the end of v. 8. The
course of it as described in vv. 6-8 raises several notoriously difficult
questions; these will be considered in the notes and need not be dealt
with here. Luke emphasizes that at every stage the travellers receive
supernatural guidance. Such guidance is renewed in dramatic form in
vv. 9, 10; Paul is directed by a vision to cross over from Troas into
Macedonia. Dibelius (129, 148f., 200) is right to point out the stress
which the narrative lays on this movement into Greece, but it is
1NEB, because they were prevented; NJB, because they had been told ... not to.
2RSV, passing by; NEB: they skirted (mg. traversed); NJB, they went through.
3NEB, set about getting a passage; NJB, lost no time in arranging a passage.

765
766 COMMENTARY ON ACTS

mistaken (as is pointed out in the notes) to make much of a move

from Asia into Europe; Paul and his colleagues remained within the
one Greco-Roman world.
What must be noted here is the introduction in the narrative of the
first person plural, ἐζητήσαμεν (v. 10). The first impression given by
this word is that the person who wrote it was present at the events he
describes. The first impression is not necessarily correct, and many
different views are held about the so-called We-passages. These are
discussed in the Introduction, pp. xxv-xxx, in the light of all the
evidence. If attention is restricted to the present passage there is no
reason to doubt the prima facie meaning of the text. A person whose
words are reported, whether by himself or an editor, was with Paul in
Troas; he was told of a vision seen by Paul and concurred with the
conclusion that the vision constituted a divine call to evangelize
Macedonia. Even if the scantily attested nos venimus is read in v. 8,
Troas marks the beginning of the association of the narrator with
Paul. The Western reading (συνεστραμμένων δέ ήμών) at 11.28
gives a ‘We’ in Antioch. Very few would accept this reading; see
however I.564. Nothing that is said in the present passage points to
the identity of the fellow traveller implied by the first person plural.

6. The subject changes from that of 16.5 (probably Luke’s

additional note) and must now be Paul, Silas, and Timothy. These
passed through, perhaps preaching on the way; for Luke’s use of
διέρχεσθαι see 13.6. The area through which they passed is descri-
bed as τήν Φρυγίαν και Γαλατικήν χώραν. It is natural at first sight
to take this phrase, with its one article (some MSS have τήν before
Γαλατικήν, but this reading is to be rejected—Metzger 441), to
mean the Phrygian and Galatian region—one region defined by two
adjectives. Less probable might seem the Phrygian [region] and the
Galatian region; or, taking Φρυγίαν as a noun, Phrygia and the
Galatian region. The last finds some support by analogy in Lk. 3.1;
so Blass (176) concludes, ‘Χώρα ad Γ. tantum pertinet, v. 18.23.’
After inconclusive argument the matter seemed to have been settled
about as far as settlement of such a question could be expected by
Lake, who (Begs. 5.231; see the whole note on Paul’s route,
5.224-40) observes, ‘Φρύγιος, the adjective formed from Φρύξ,
was “of three terminations” in earlier Greek, but Lucian uses it as of
only two [Harmonides I], and I know of no instance of the
nominative with the feminine termination in Greek contemporary
with the New Testament ... In any case Φρυγία had undoubtedly
become a substantive proper name, and the first thought of any reader
would be to interpret it so.’ This observation and conclusion were
however questioned by C. J. Hemer (JTS 27 (1976), 122-6; see also
Acts 112, 277-307; there is a supplementary note in ND 4.174). T
am far from denying that Φρύγιος may be two-terminational, but the
 42. GUIDED BY THE SPIRIT TO TROAS. 16.6-10. 767

alternative certainly exists, and is probably the more usual' (124).

Hemer gives nine examples, ranging from the fourth century BC to
the fourth century AD. Probably the best, as regards date and clarity
is νόθον ἐκ Φρυγίας γυναικός (Apollodorus, Bibliotheca 2.4.5).
That Φρυγίαν in Acts 16.6 is an adjective must be accorded a higher
measure of probability than Lake allowed, but the matter is not
closed in view of the occurrence of the similar expression, τήν
Γαλατικήν χώραν καἰ Φρυγίαν at 18.23, where the word must be a
noun. This suggests that in 16.6 also Φρυγία is a noun, but cannot
prove it (cf. Schneider 2.205). Luke may be using, in the two verses,
different sources, in which the word was taken in different ways; or
he may be deliberately changing its sense by changing the order of
words. There remains however some probability that Luke intended
in each place to refer to the same area. If, with Hemer, we take
Φρυγίαν as an adjective we are presented with a χώρα that was both
Phrygian and Galatian. This is explained by Ramsay (Church 81) as
that part of Phrygia that had in 25 BC been incorporated (with a
number of other districts) in the new province of Galatia. It could be
‘idiomatically rendered “the Phrygo-Galatic territory’” (80), by
analogy with Pontus Galaticns, a term used ‘to denote a large district
of Pontus which was added to the province of Galatia a few years
B.C.’ (ibidem). It must be noted that though there is good evidence
for the name Pontus Galaticns there is none for Phrygia Galatica. It is
perhaps unlikely that we shall ever possess sufficient literary and
epigraphical evidence to settle this question by a simple considera-
tion of names and the way in which they are used. A stronger
consideration, which Lake (Begs. 5.235f.) regards as ‘perhaps deci-
sive against Ramsay’s theory’ is ‘the fact that if ή Φρυγία καί
Γαλατική χώρα means the regio of the province of Galatia called
Phrygia Galatica, it is impossible that Paul’s route through this
district brought him out anywhere near Mysia. Ramsay has to argue
that Paul, after passing through Phrygia Galatica, journeyed through
the province of Asia until he came to the neighbourhood of Mysia,
for though the Spirit prevented him from preaching in Asia, it did not
prevent him from travelling through it.’ There is a similar geo-
graphical argument in Thornton (267f.). Ramsay’s explanation is
hardly convincing. But to follow Lake’s point it is necessary to go
further in Luke’s text.
Phrygia and Galatia were both ancient kingdoms in Asia Minor.
After a varied history of subjection to various races, ‘in 116 B.C. the
greater part of [Phrygia] was absorbed in the province of Asia, and in
25 B.C. the remaining eastern portion became a region of the
province Galatia’ (W. Μ. Calder in OCD 829). The Roman province
of Galatia was thus formed in 25 BC, based on the ancient kingdom
of Galatia but including also parts of Phrygia, Lycaonia, and Pisidia.
It was later expanded by further additions.
768 COMMENTARY ON ACTS

Paul and his companions, then, passed through τήν Φρυγίαν καἰ
Γαλατικήν χώραν, κωλυθέντες υπό τού αγίου Πνεύματος λαλήσαι
τόν λόγον ἐν τῆ ’Ασία. A reader unfamiliar with the geography of
the region in question would certainly translate this, ‘They passed
through the area described, having been forbidden, that is, after they
had been forbidden, or because they had been forbidden, to speak the
word in Asia,’ following the common rule that an aorist participle
(κωλυθέντες) refers to action before the time of the main verb
(διήλθον). This suggests that the journey through Phrygian and
Galatian territory was an alternative to a journey through Asia, and
led to a point over against Mysia whence entry into Bithynia was
possible, though it was prevented by a further intervention of the
Spirit. A map however will show that a journey through Galatic
Phrygia (‘South Galatia’—a frequently used but undesirable term)
would have led the travellers only up to the frontier of Asia, so that
Mysia could have been reached only by passing through Asia. If
however they passed through Phrygia and old Galatic territory
(‘North Galatia’, in the corresponding phrase) they would approach
both Bithynia and Mysia, and would be able to reach Troas (v. 8). If
the Phrygia Galatica view described above is to be maintained it is
necessary to take the aorist participle κωλυθέντες to refer to a time
after that of the main verb; so for example Ramsay, Church 89. They
were forbidden to preach the word in Asia, and (then) passed through
the Phrygian-Galatian territory. Ramsay’s position is summed up by
him (Church 89) as follows: ‘It has been contended that the participle
κωλυθέντες gives the reason for the finite verb διήλθον, and is
therefore preliminary to it in the sequence of time. We reply that the
participial construction cannot, in this author, be pressed in that way.
He is often loose in the framing of his sentences, and in the long
sentence in verses 6 and 7 he varies the succession of verbs by
making some of them participles. The sequence of verbs is also the
sequence of time: (1) they went through the Phrygo-Galatic land; (2)
they were forbidden to speak in Asia; (3) they came over against
Mysia; (4) they assayed to go into Bithynia; (5) the Spirit suffered
them not; (6) they passed through Mysia; (7) they came to Troas.’
Whether this is a possible way of taking the participle is a question
on which opinion is divided. It is of course true that the aorist tense
describes a kind of action (Aktionsart) rather than the time at which
an action takes place, and aorist participles that describe action
coincident in time with that of the main verb though not common are
sufficiently familiar. Μ.1.132-4 discusses the question and con-
cludes by agreeing with P. W. Schmiedel (EBib 1599), ‘It has to be
maintained that the participle must contain, if not something ante-
cedent to “they went” (διήλθον), at least something synchronous
with it, in no case a thing subsequent to it, if all the rules of grammar
and all sure understanding of language are not to be given up.’ Moule
 42. GUIDED BY THE SPIRIT TO TROAS. 16.6-10. 769

(IB 100), considering on the same page Acts 25.13 (see the note),
observes Lake’s topography and translation (see above) and says that
this ‘is certainly the natural explanation of the Greek’. Μ. 3.80 on the
other hand writes, ‘Even time which is future to the main action
seems to be denoted by the aor. ptc.', and cites Mt. 10.4; Jn 11.2
[these examples are quite unconvincing]; Acts 16.6; 25.13. Zerwick
(§ 265) notes that Ramsay’s interpretation means that ‘supponendus
esset sensus subsequens participii κωλυθέντες ( = έκωλύθησαν
δέ)’. But he evidently regards this as a minority view. G. Μ. Lee
(NovT 9 (1967), 41, 42) cites προσφύς in Babrius (Fable 143.1-4)
as an example of an aorist participle of subsequent action, but it
probably does not strain the facts to say that in this passage all the
action (a viper’s bite) happens so quickly that one could legitimately
think of coincident action. The same author returns to the matter in
Bib 51 (1970), 235, 236. His Latin analogy (NovT 17 (1975), 199) is
totally unconvincing. G. W. Hansen (in Book of Acts 1.378f.) takes
the reference to be to ‘the Phrygian-Galatian region’ but does not
deal with the grammatical question. It is hard to escape the conclu-
sion in Begs. 4.186: ‘It is impossible to translate the passage without
doing violence to the Greek, unless we recognize that the phrase
means that Paul first contemplated preaching in Asia, and, being
prevented from doing this, passed through τήν Φρυγίαν καί
Γαλατικήν χώραν.’ It is worthwhile to add that Luke’s geographical
knowledge, though better than that of many of his contemporaries,
was probably a good deal less precise than that of a modern scholar
equipped with good modern maps. This would be especially true of
more remote regions which he (Luke) had not himself penetrated.
διελθόντες (<a> vg) for διήλθον (P74 A B C D Ε Ψ 33 81 323
614 945 1175 1739 2495 al gig) makes more easily possible a
reversal of the order of this verb and κωλυθέντες. The addition (by
D) of μηδενί after πνεύματος improves the Greek and underlines the
prohibition. τόν λόγον alone (for the Gospel) is used e.g. at 4.4;
the addition (by D gig vgcl syp bo Speculum) of τοῦ θεού is
conventional.
The province of Asia was a wealthy and highly civilized part of
Asia Minor, based upon a considerable number of city states, still at
least partially independent. ‘From 48 B.C. till c. A.D. 297 Asia
included all the territory from Tyriaion to the sea, with the adjacent
islands; it was bounded on the north by Bithynia, on the south by
Lycia, and on the east (after 25 B.C.) by Galatia’ (W. Μ. Calder and
E. W. Gray, OCD 131). See further on other aspects of Asia below,
on ch. 19.
It is characteristic of Luke to ascribe the direction of the mission to
the Holy Spirit; cf. v. 7. How the Holy Spirit issued the prohibition in
question is not explained. Luke possibly thinks of a communication
through prophets; cf. 13.1-3.
770 COMMENTARY ON ACTS

7. Forbidden to speak the word in Asia the missionaries came

κατά τήν Μυσίαν, opposite, over against, Mysia (LS 883, s.v. κατά,
B I 3, with numerous illustrations). This presumably means that they
came up to the Mysian border but entered Mysia no more than they
had Asia, or rather, since Mysia was a (northern) part of the province
of Asia (but see on v. 8), having so far avoided Asia they still
avoided it when, moving northward, they came to the Mysian part of
Asia. Bithynia lay to the east of Mysian Asia, and stretched eastward
along the southern coast of the Black Sea. Pompey the Great
organized Bithynia as a province in union with Pontus (further east).
During the first century AD it was a senatorial province, though its
great importance as lying on the route to the east led to an unusual
measure of imperial interference, and finally Marcus Aurelius made
it an imperial province.
For ἐπείραζον, D syp have ήθελαν (sic). For the (unusual) Spirit of
Jesus C* gig bomss have πνεύμα κυρίου and <a> sa have simply
πνεύμα. Luke means nothing different from what he means by the
Holy Spirit in v. 6; why he has a different expression is not known. It
is unlikely that he has a new source of information. He may simply
have been seeking variety of expression as he makes his unchanging
assertion: the Christians did what they did under the instruction and
guidance of God, who worked through his Spirit, here more narrowly
defined as the Spirit of Jesus.

8. παρελθόντες τήν Μυσίαν is usually taken to mean, They

passed by Mysia (see v. 7). This translation, however, though it
accords well with the usual meaning of παρέρχεσθαι, and recognizes
that Mysia was, as a part of Asia, forbidden country, encounters a
geographical difficulty in that Troas was situated in Mysia and could
only be approached by traversing Mysia, which Luke may possibly
have distinguished from Asia. BA 1265 (followed by Haenchen and
Conzelmann) takes the view that in this passage παρερχ. means not
to pass by but to pass through, citing Appian, Bellum Civile 5.68
§ 288 (... made their way through a crowd), and 1 Macc. 5.48. Some
support may be found also in the use of παρερχ. διά in Mt. 8.28;
PAmh 154.2 (6th/7th century AD), μή παρελθεῖν τινα διά τῶν
ἐποικείων αυτού). BA also note, convincingly, that the variant in D,
διελθόντες, which Dibelius (92) says ‘appears logically justified’,
can be explained if it is supposed that the Western editor (a) did not
know that παρέρχεσθαι could mean traverse (but 17.15 D suggests
that he did!), and (b) did know that Troas could not be reached
without traversing Mysia. Luke may have used παρέρχεσθαι
because διέρχεσθαι suggested to him (see 13.6) a preaching tour,
whereas on this occasion Paul and his companions travelled without
preaching. An alternative explanation, simpler and perhaps better, is
that παρέρχεσθαι means simply to arrive, as at Acts 24.7 (an
 42. GUIDED BY THE SPIRIT TO TROAS. 16.6-10. 771

observation relevant even if this verse is not an original part of the

text of Acts). Thus in v. 7 the travellers came opposite Mysia (κατά
τἠν Μυσίαν, at v. 8 they entered it (Cf. Metzger 442). ‘παρελθόντες
( etc.) sensui adversatur: non praetereunda, sed transeunda erat
Mysia, ut ad Aegaeum mare venirent’ (Blass 176). Otherwise
Rackham (276); Pesch (2.101).
κατέβησαν, they came down, because Troas lay near the coast;
κατήντησαν (D) lacks the clear sense of descent (notwithstanding
the compounded κατά), but otherwise scarcely differs; but nos
venimus (Irlat) introduces the ‘We’ material two verses before the
point (v. 10) at which the main textual tradition does so. This is not a
slip; the context (Adv. Haereses 3.14.1) shows that Irenaeus knows
what he is doing: ‘... this Luke was inseparable from Paul... he says
that when Barnabas ... had sailed to Cyprus, we came to Troas’. The
person who associates himself with Paul in the first person plural still
appears for the first time at Troas, but he arrives there with Paul. See
on v. 10.
Troas, in full ’Αλεξάνδρεια ή Τρωάς, was founded by Antigonus
(323-301 BC). It was a free city; under Augustus it became a colony.
See Hemer 112f. In addition to 16.11; 20.5, 6, cf. also 2 Cor. 2.12; 2
Tim. 4.13.
On vv. 6-8 see also Taylor (5.234-9).

9. The route of the missionaries is still (cf. vv. 6, 7) determined by

supernatural means, now however by a vision (ὃραμα; cf. 7.31; 9.10,
12; 10.3, 17, 19; 11.5; 12.9; 16.10; 18.9; elsewhere in the NT only
Mt. 17.9; the distribution reflects Luke’s insistence on the divine
control of the early Christian movement). The vision is seen διά τής
(but τής is omitted by A B D 6 36 1175 pc) νυκτός; Moule, IB 56,
noting that νυκτός alone would mean by night, translates in the night;
cf. Zerwick (§ 115: spatium temporis, intra quod ...). Luke’s
handling of cases and prepositions is perhaps not so precise as this
distinction would require.
ὤφθη with the dative will mean appeared to (rather than was seen
by; cf. 2.3 and other passages). Wilcox (125) regards ἑστώς καί
παρακαλών as perhaps an Aramaism; cf. 1.11; 5.25. It has been
asked how Paul was able to tell that the man was a Macedonian
(ἀνήρ Μακεδών τις); if there is a historical event behind the story it
is enough to say that in a dream one knows this kind of thing;
otherwise the urgent appeal to come and help us (ήμῖν) in Mace-
donia might justify the implied inference. The man ην ἑστώς; this
adds vividness; cf. 11.13. The participles are descriptive rather than
members of a periphrastic tense: the man was standing and begging
and saying. διαβαίνειν implies coming from one side of a barrier
(here the Thracian Sea, the northern part of the Aegean) to the other;
cf. Lk. 16.26 (χάσμα μέγα); Heb. 11.29 (τήν ἐρυθράν θάλασσαν).
772 COMMENTARY ON ACTS

βοηθεῖν is a surprisingly general word (elsewhere in Acts only

21.28, in a different sense; cf. 27.17); one would expect, Come and
preach the Gospel to us, or the like. The intention may be to indicate
that the Macedonians do not yet know what the Gospel is; they are
aware of a need of help, not of the particular help that Paul had to
offer.
A vision that appears in the course of the night is probably a
dream; for the communication of supernatural instructions by dreams
see references in Betz 53f., e.g. Lucian, De Morte Peregrini 26 (an
intention changed by dreams). ‘Dies ist ein fester Topos in den
apokryphen Apostelakten’ (Bornkamm, 4.185).
According to Davies (Land 278), Luke ‘had an acute awareness of
the point at which the Gospel passed over to Europe from Asia... the
entry upon a new area of the Christian mission is due to vision’.
Davies is right in noting Luke’s ‘sensitivity to geography’, but
probably overstates the significance of the transition from Asia to
Europe. Both Mysia and Macedonia were in the first century AD
associations of Hellenistic cities which had become Roman prov-
inces (in the case of Mysia, part of the province of Asia). In Philippi
Paul would speak the same Greek that he had spoken all the way
from Antioch.
Macedonia had been a province since 146 BC. More details will be
given when individual towns are mentioned.
There is in this verse a good deal of minor textual variation which
shows nothing more important than Western readiness to handle the
text freely, perhaps inviting reflection on the kind of authority it had;
see Introduction, pp. xxif.
The omission of τής was mentioned above; the following variants
may also be noted:
ὅραμα] έν ὁράματι D e syp
ἀνήρ] ώσεί ἀνήρ D syp sa
ᾖν] om. D* E pc
καί] κατά πρόσωπον αὐτοῦ D 614 pc (syh**) sa
Thus the text of D runs: καὶ ἐν ὁράματι διά νυκτὸς ὤφθη τω
Παύλῳ ώσεί ἀνήρ Μακεδών τις ἑστὠς κατά πρόσωπον αυτού
παρακαλῶν καί λέγων Διαβὰς ... The editor (copyist) is not trying
to make a point; he is not concerned to reproduce his text with
elaborate precision.

10. For temporal ὡς as a characteristic of Lucan style see 16.4;

also BDR § 455.2 and BA 1792; also Μ. 4.70, but ὡς is seldom if
ever due to the Aramaic
Paul saw the vision, and ευθέως (here expressing unhesitating
obedience to a divine command) ἐζητήσαμεν, we sought ... first
person plural. This (unless variants are accepted at 11.28; 16.8; see
 42. GUIDED BY THE SPIRIT TO TROAS. 16.6-10. 773

the notes) is the first occurrence of the first person plural which is a
striking characteristic of some of the narratives in the second part of
Acts. For a discussion of their interpretation, significance, and
bearing on the authorship of Acts see Introduction, pp. xxv-xxx.
There is no doubt that the immediate (but not for that reason
necessarily correct) impression given by the present passage is that at
this point, or shortly before it, Paul, Silas, and Timothy were joined
by another person, who proceeds to describe (presumably at least up
to the end of the Philippi story, as well as at later points in the book)
events as he himself witnessed and recalled them. He may have been
the author of the whole book—‘Luke’, or the author of a source
incorporated by the author of the whole book. Or Luke may at this
point have switched from third person to first person narrative for
some reason of his own—because it suited his theme, because it
added greater vividness, or because it gave the untrue suggestion that
he was present so that his narrative might be accepted as an eye-
witness account of events. These possibilities can be seriously
discussed only when all the material has been surveyed. It should
however be borne in mind that, if there was an eye-witness, we may
have already met him. Schneider (2.204, cf. 1.89-95) points out that
the occurrence of ‘we’ in this verse does not necessarily mean that
the writer has only just joined the party. He might be Timothy or
Silas. Pesch (e.g. 2.98) thinks the eye-witness material may come
from Timothy. Roloff (239), more cautiously, ‘Die Wir-Stücke
stammen sicher aus dem Umkreis des paulus, aber der Verfasser ist
nicht Lukas.’ On this question see especially Thornton, passim.
Worth mentioning, but not to be accepted, is the suggestion that Luke
had come from Philippi with an invitation to Paul, and was himself
the ‘vision’ of v. 9.
συμβιβάζειν is used in a variety of ways; cf. 9.22; 19.33. Here it
can hardly mean anything other than conclude (cf. Plato, Hippias
Minor 369d, διαπυνθάνομαι καί ἐπανασκοπώ καί συμβιβάζω ...).
‘συμβιβάζειν (9.23) hic ut att. colligere' (Blass). The conclusion was
inevitable. God had called us to evangelize the Macedonians, πρός
adds to καλεῖν a sense of specific direction; cf. 13.2. The verb is
followed by an infinitive of purpose, on which Radermacher (153)
comments, ‘Dieser Infinitiv ist überall möglich, wo irgendeine innere
Beziehung zwischen Hauptsatz und Absichtssatz vorhanden ist.’ For
Luke’s use of εύαγγελίζεσθαι see 5.42; 8.4; etc. Schneider (2.207)
connects it here with Isa. 61.1f.
Again (cf. v. 9) there is a free Western rewording: διεγερθεὶς οὖν
... διηγήσατο τό όραμα ήμῖν (of course, how else could we know
of it?), καί ἐνοήσαμεν ότι προσκέκληται ἡμάς ό κύριος
εὐαγγελίσασθαι τούς ἐν τή Μακεδονία. In reading κύριος in place
of θεός (P74 A B C 33 81 al), D is joined by <a> gig sy sa Irlat.
 43. PAUL AND SILAS AT PHILIPPI 16.11-40

(11) We set sail from Troas and made a straight run to Samothrace, and on
the next day to Neapolis, (12) and thence to Philippi, a leading city1 of2 the
province of Macedonia, a colony. We stayed in the city several days. (13) On
the Sabbath day we went outside the [city] gate by a river, where3 we
supposed that there was a place of prayer. We sat down and4 spoke with the
women who had gathered. (14) A woman, Lydia by name, a5 dealer in purple
from the city of Thyatira, who worshipped God, listened; the Lord opened
her heart to give attention to the things spoken by Paul. (15) When she and
her household were baptized she asked us, ‘If you have judged me to be6
faithful to the Lord, come into my house and stay;’ and she constrained us to
do so.
(16) It happened that as we were on our way to the place of prayer, a slave
girl who7 had8 an oracular spirit met us; she made much profit for her
owners by9 giving oracles. (17) She followed Paul and us, crying out, ‘These
men are slaves of the Most High God; they are proclaiming to you a way of
salvation.’ (18) She did this for many days. Paul could endure it no longer,
turned, and said to the spirit, ‘I command you in the name of Jesus Christ to
come out of her.’ It came out at once. (19) When her owners saw that the
hope of their profit had departed, they seized Paul and Silas and dragged
them into the10 Agora before the rulers. (20) Having brought them to the
magistrates they said, ‘These men, who are Jews, are greatly disturbing our
city, (21) and are proclaiming customs which it is not lawful for us, who are
Romans, to receive or practise.’ (22) The crowd joined in the attack upon
them and the magistrates tore off their clothes and commanded11 [their
officers] to beat them. (23) When they had laid many stripes upon them they
cast them into prison, ordering the gaoler to guard them securely. (24) He,
having received such a command, put them in the inner prison and secured
their feet in the stocks. (25) At midnight Paul and Silas in their prayers were
singing psalms to God, and the prisoners were listening to them. (26)
Suddenly there was a great earthquake so that the foundations of the prison
were shaken. Immediately all the doors were opened and the bonds of all [the
prisoners] were loosed. (27) The gaoler woke up, and when he saw the doors
of the prison standing open he drew his sword and was about to kill himself
1NJB, the principal city.
2NEB, NJB, that district.
3NEB, thought there would be; NJB, this was a customary place for prayer.
4NJB, preached to.
5NJB, in the purple-dye trade.
6NJB, a true believer in.
7NJB, was a soothsayer.
8RSV, a spirit of divination.
9NEB, telling fortunes.
10RSV, NJB, market place; NEB, main square.
11RSV, gave orders to beat them with rods.
774
 43. PAUL AND SILAS AT PHILIPPI. 16.11-40. 775

because he supposed that the prisoners had escaped. (28) But Paul called out
with a loud cry, ‘Do yourself no harm, for we are all here.' (29) [The gaoler]
asked for lights and sprang in. He was trembling, and fell down before Paul
and Silas. (30) He brought them outside and said, ‘Gentlemen, what must I
do to be saved?’ (31) They said, ‘Believe in the Lord Jesus and you will be
saved—you and your household.’ (32) And they spoke to him the word of
the Lord, together with all who were in his house. (33) He took them in that
hour of the night and washed them clean from the effect of the blows they
had received, and he was baptized, he and all his family, immediately. (34)
He brought them up into his house, prepared a meal, and rejoiced with all his
household, because he had come to faith in God. (35) When day broke, the
magistrates sent the lictors, saying, ‘Release those men.’ (36) The gaoler
reported these words to Paul, saying, ‘The magistrates have sent that you
should be released; so now depart and travel on in peace.’ (37) Paul said to
them, ‘They have beaten us publicly, uncondemned, men who are Romans;
they have put us in prison; and will they now put us out secretly? No, indeed;
rather let them come themselves and lead us out.’ (38) The lictors reported
these things to the magistrates. When they heard that they were Romans they
were afraid, (39) came and placated12 them, and having brought them out
asked them to leave the city. (40) When they had come out of the prison they
went into Lydia’s house, and when they had seen the brothers they
encouraged them, and left.

Bibliography
R. Borger, ThR 52 (1987), 36-8.
L. Bormann, Philippi, SuppNovT 78 (1995).
C. Burchard, ZNW 69 (1978), 143-57.
G. Delling, ΝονΤ 7 (1964-5), 285-311.
Μ. Hengel, FS Kuhn, 157-84.
P. van Minnen, JSNT 56 (1994), 43-52.
Th. Mommsen, ZNW 2 (1901), 82f.
D. Noy, JTS 43 (1992), 188-22.
D. R. Schwartz, Bib 65 (1984), 357-63.
W. Stegemann, ZNW 78 (1987), 200-29.
P. R. Trebilco, JSNT 36 (1989), 51-73.
W. C. van Unnik, Sparsa Collecta I, SuppNovT 29 (1973), 374-85.
A. Vögeli, ThZ9 (1953), 415-38.
C. S. de Vos, NTS 41 (1995), 292-6.
P. Weigandt, NovT 6 (1963), 49-74.
A. Weiser, FS Pallottis, 118-33.
Μ. Wilcox, JSNT 13 (1981), 110f.

12RSV, NEB, apologized to.

776 COMMENTARY ON ACTS

Commentary

The use of the first person plural which begins in 16.10 continues
through the earlier part of the new section, which falls into six parts.
16.11,12: the missionary group makes it way from Troas to Philippi,
where it is to remain for some time. 16.13-15: they make contact
with a group of women who are evidently in the habit of frequenting
a (Jewish) place of prayer; one of them, Lydia, is converted, and
takes the missionaries to her home. 16.16-18: Paul exorcises a
demon from a girl. 16.19-24: Paul and Silas are brought before the
magistrates, flogged and imprisoned. 16.25-34: there is an earth-
quake, Paul and Silas, and other prisoners, are set free, and the gaoler
becomes a Christian. 16.35-40: the magistrates, learning that Paul
and Silas are Roman citizens, release them and ask them to leave the
city.
Verbs or pronouns in the first person plural occur in vv. 11, 12,13,
15,16,17; that is, the narrator expressly represents himself as present
on the journey from Troas to Philippi, as visiting the place of prayer
and entertained by Lydia, and as addressed by the prophesying girl.
From this point ‘We’ disappears; this however means, or need mean,
no more than that is was only Paul and Silas who were arrested and
imprisoned—the narrator no longer played an active part in the story.
If we are to think of the author (real or supposed) of a source (or of
the book itself) it is reasonable to think that he joined Paul’s party at
Troas, possible that he had come to Troas from Philippi (or at least
from Macedonia), and probable that having reached Philippi he
remained there throughout Paul’s stay, and after it. See on 20.4, and
Introduction, pp. xxv-xxx. Whoever wrote Acts 16 seems to have
known Philippi (especially if in v. 13 ἐνομίζομεν is read), and to
have been not unfamiliar with its administration (στρατηγοί, v. 20).
Against such realistic features of the narrative must be placed the
earthquake (v. 26). There is nothing incredible in an earthquake, but
the reader does not expect one to be violent enough to release all the
prisoners in the town gaol, yet gentle enough to do them no harm,
and sufficiently localized to be, it seems, unnoticed by the town
officials. The development of a story within local tradition creates no
difficulty; one would think however that the memory of an observer,
whose words were recorded in the narrative we have, would have
exercised a critical check on such developments. There are two
possible ways of escaping this problem. One is to suppose that the
eye-witness implied by ‘We’ was no longer present after v. 17; his
narrative ends at that point. We should have to admit that if he was
no longer in Philippi we have no idea what became of him until he
reappears at 20.5; he would then have the opportunity of adding local
tradition to his own recollections. The other would distinguish
between the witness implied in ‘We’ and the author of Acts; the latter
 43. PAUL AND SILAS AT PHILIPPI. 16.11-40. 777

used a ‘We-document’ up to v. 18 and added local tradition to it. As

an example of more radical treatment of the chapter Weiser’s
analysis may be noted. He rates as standing on their own the travel
details of vv. 11, 12a; in addition, tradition supplied the raw material
for the conversion of Lydia, the exorcism, the accusation of causing
disturbance, flogging and imprisonment, release and dismissal, and
the conversion of the gaoler. ‘Lukas hat diese Einzelnachrichten mit
den Mitteln des dramatischen Episodenstils szenisch ausgestaltet und
miteinander zum Bericht eines sich steigernden Geschehensablaufs
verbunden. Insbesondere stammen von Lukas: der Wir-Stil, die
Kennzeichnung Philippis (V 12b), die Angabe der Aufenthaltsdauer
(V 12c), die konkreten Ausgestaltungen der Lydia- und Exor-
zismusgeschichte (VV 13-15, 16—18), die Anklage wegen
Geschäftsschädigung (V 19a) und der Verderbnis römischer Sitten
(V 21), die Angaben über die Strenge der Haft (VV 23b, 24), die
Befreiungswundererzählung mit den Konkretionen des Berichts über
die Bekehrung des Gefängniswärters (VV 25-34), die Diskussion
über das römische Bürgerrecht und die davon Abhängigen Aussagen
(VV 37-39a), die abschliessende Erwähnung der Einkehr im Hause
Lydias (V 40)’ (431). It need not be questioned that the author has
contributed substantially to the narrative as we have it, or that he was
elaborating material that he derived from tradition. The important
question left is whether the final author of the book was responsible
for adding not only a number of lively details but also the ‘We-style’.
For this general question see Introduction, pp. xxv-xxx. We may
well have to distinguish in this chapter, which provides an important
introduction to the general question, (a) eye-witness recollection,
using the first person plural; (b) local tradition; (c) the elaboration of
local tradition; (d) the editorial work of the final author. Cf. Taylor
241: vv. 1 l-13a, 16b-23a, 35-40 come from the Journal de Voyage;
Luke adds vv. 13b-16a and 25-34; Act III introduces various
modifications, especially in vv. 23b, 24.
There is an important discussion of honour and shame as themes
of this paragraph in Johnson (303f.); more fully, Rapske Book ofActs
(3.303f.).

11. ἀναχθέντες: see 13.13. The participle is followed by δέ in P74

A Ε Ψ 6 33 81 326 1175 pc vg, by οὖν (linking more closely with
what precedes) in B C <a> gig syh. D(*) 614 syhmg rewrite: τή δέ
επαύριον άναχθέντες.
ἀπὸ Τρφάδος: see 16.8.
εὐθυδρομεῖν occurs again at 21.1, not elsewhere in the NT. It is
used especially of ships: to sail a straight course; so Philo, Legum
Allegoriae 3.223 (but at De Agricultura 174 transitively of the wind
driving a ship).
Samothrace is an island at the northern extremity of the Aegean,
778 COMMENTARY ON ACTS

not far from the coast of Thrace. It had long possessed a notable
sanctuary of the ‘Great Gods’, of Thracian origin but widely popular
in the Hellenistic age. Many came to be initiated into the two grades
of the mysteries practised at the temple. Paul and his companions
presumably spent the night there though they may not have left the
ship; there was a beach but no harbour.
τῇ δἐ ἐπιούση (sc. ἡμέρα; cf. 7.26), on the next day the journey
was continued to Neapolis, the modern Cavallo, the port of Philippi.
At the time of the battle of Philippi (42 BC) Brutus and Cassius used
Neapolis as their naval base; later a colony was established there (cf.
Philippi, v. 12). Ignatius, Polycarp 8.1, records that he sailed from
Troas to Neapolis. Acts describes no missionary work at Samothrace
or Neapolis; possibly neither place had a synagogue or place of
(Jewish) prayer; cf. v. 13.
Neapolis appears as Νεάπολιν in C D* E Ψ <a> but the better
text is Νέαν Πόλιν, in P74vid AB Dc 1175 1739 pc. This is the
older form (M.2.151,278); for inscriptions and coins see Hemer 113,
n. 30.

12. κἀκεῖθεν εις Φιλίππους: but at Neapolis the travellers must

have left their ship and travelled about 10 miles inland. The battle of
42 BC, in which the Triumvirate defeated the republican forces under
Brutus and Cassius, was fought a little to the west of the town. After
the battle Μ. Antonius founded the colonia (see below), which was
subsequently augmented by his former colleague Octavian (Augus-
tus) who defeated him at Actium (31 BC). The full name of the new
town was Colonia Julia Augusta Philippensium. The description of
the town in the present verse is obscured by textual and other
problems. The main readings are:
(a) πρώτη τής μερίδος τής Μ. <a>
(b) πρώτη τής μερίδος Μ. P74 A C Ψ 33 36 81
323 945 1175 1891 pc
(c) πρώτη μερίδος τής Μ. Β
(d) πρώτη τής Μ. 614 1241 1739 2495 pc
syh
(e) πρώτη μερίς Μ. E samss
(f) κεφαλή τής Μ. D syp
(f) is probably a translation variant, πρώτη was taken into Latin as
caput (so d) and into Syriac as and then came back into Greek
in a literal and unidiomatic rendering. See however Clark (362-5).
When Μακεδονίας is preceded by τής the meaning is probably
partitive—that district (part) of Macedonia. When it is not,
Μακεδονίας is probably in apposition—of the district (i.e. province)
Macedonia, τής before μερίδος is demonstrative—of that part of Μ.
(see Moule, IB 111). πρώτος is used of leading or capital cities (e.g.
 43. PAUL AND SILAS AT PHILIPPI. 16.11-40. 779

Thucydides 2.8.1), but if Luke had intended to describe Philippi as

the capital of Macedonia he should have written ή πρώτη ... πόλις
(with the article). The article is wanting, and with the grammatical
goes the further difficulty that Philippi was not the capital of
Macedonia or of any part (district, μερίς) of it. Macedonia had
however been divided into four districts (Livy 45.29: ... deinde in
quatuor regiones dividi Macedoniam). Sherwin-White (93), BDR §
164.3, n. 7, and many others (including recently Tajra 5) conjecture
as the original text πρώτης μερίδος Μ. πόλις, a city of the first
district of Macedonia. Livy (loc. cit.) describes the first region as
follows: quod agri inter Strymonem et Nessum amnem sit: acces-
surum huic parti trans Nessum, ad orientem versum, qua Perseus
tenuisset vicos, castella, oppida, praeter Aenum, et Maronem, et
Addera; trans Strymonem autem vergentia ad occasum, Bisalticam
omnem cum Heraclea, quam Sinticen appellant. Philippi is in this
area. The Committee whose work is reported in Metzger (444-6)
also approved this conjecture, but there is an appended note, signed
K. A[land] and B. Μ. M[etzger]: ‘Despite what have been regarded
as insuperable difficulties in the commonly received text (πρῶτη τῇς
μερίδος), it appears ill-advised to abandon the testimony of P74 A
C 81 al, especially as the phrase can be taken to mean merely that
Philippi was “a leading city of the district of Macedonia”; cf. Bauer’s
Griechisch-Deutsches Wörterbuch, 5te Auflage (1958), s.v. μερίς’.
BA (successor to the dictionary referred to) 1023f. however takes the
view that ‘Die Über. erste Stadt betreffenden Bezirkes von Maz... ist
nur schwer erträglich.’ The article does not consider (an equivalent
of) ‘a leading city ... ’, but prefers the NA26 text πρώτη[s] μερίδος
τής Μ.—the conjecture referred to above. This is surprising, not least
in view of the fact that BA 1452 (s.v. πρώτος 1.c.a.) recognizes the
use of πρώτος in the sense of rank but Ohne superi. Wert’, citing
Eph. 6.2 (‘ein bedeutsames Gebot’); Lk. 15.22; 1 Cor. 15.3 ('unter
den ersten = wichtigsten Stücken’). The word is not infrequently
used in the plural (LS 1535), as for example Thucydides 2.8.1
(referred to above, τών πρώτων πόλεων). It would be less natural
but not impossible to refer to one of such cities as a πρώτη πόλις.
See Begs. 4.188f., and for an interesting note on the conjectured
reading R. Borger in ThR 52 (1987), 36-8. See further Hemer
(113f.), and for a bolder conjecture Hort (Introduction, Notes 96f.):
(‘It is not impossible that μερίδος should be read as πιερίδος (M for
Π I), for Philippi belonged to the Pieria of Mount Pangaeon, and
might well be called “a chief city of Pierian Macedonia”; so Steph.
Byz. Κρηνίδος, πόλις Πιερίας (codd. Σικελίας), ας Φίλιππος
μετωνόμασε Φιλίππους: cf. Herod. vii 212; Thuc, ii.99. The name ή
Πιερίς Μακεδονία does not seem however to occur elsewhere, and
would more naturally be applied to the more famous Pieria in the
S.W. of Macedonia. For the present the reading must remain in
780 COMMENTARY ON ACTS

doubt’ For Herodotus 7.212 we should no doubt read 7.112. It will

be noted that Hort seems to find no difficulty in taking πρώτη
πόλις as ‘a chief city’. Page (184) somewhat surprisingly approves
the suggestion of Erasmus: ‘prima occurrit a Neapoli petentibus
Macedoniam.’
Philippi was a colonia, A colonia was originally a settlement of
Roman citizens in conquered territory, intended to help to hold down
the local population; then a place to which surplus Italian population
could be consigned; then (as with Philippi) a place where discharged
soldiers were pensioned off with land. Coloniae enjoyed libertas
(autonomous government), immunitas (from tribute and taxation),
and lus Italicum, which meant, in effect, that they were considered to
be a piece of Italian soil, having a Roman form of administration
with Roman law and judicial procedure. Marshall (266) suggests that
details about Philippi are provided in order to prepare for Paul’s first
encounter with Roman administrators. He had however already
encountered a proconsul at 13.7. On Philippi see further L. Bormann,
Philippi: Stadt und Christengemeinde zur Zeit des Paulus, Supp-
NovT 78, Leiden, 1995.
The periphrastic imperfect, with the auxiliary separated from the
participle, ἦσαν ... διατρίβοντες, emphasizing the continuous
though not protracted stay (ἡμέρας τινάς; cf. 9.19; for a longer stay
Luke has ἡμέραι ἱκαναί) over against the events of the next verses.

13. τῇ τε ημέρα τῶν σαββάτων; for the expression see 13.14. Paul
and his companions (we) went out of the city gate παρά ποταμόν
(τόν ποτ., D). Μ. 1.82 quotes but doubts the view of Ramsay that the
absence of the article shows familiarity with the locality. The article
may have been dropped because the phrase was a common one (cf.
‘down town’). The river on which Philippi stood was the Gangites
(or Cangites). The party were, it seems, looking for a προσευχή, or
place of prayer (on the word see below), but there is considerable
textual variation.

(a) ἐνομίζομεν προσευχὴν εἰναι Αc €Ψ33 81 pc bο

(b) ἐνόμιζεν προσευχὴν εἰναι
(c) ἐνομίζομεν προσευχῇ εἰναι B pc
(d) ἐνομίζετο προσευχὴ εἰναι A*vid E
(e) ἐδόκει προσευχὴ εἰναι D
p74vid
(f) ἐνόμιζεν προσευχὴ εἰναι

(f) is probably a slip and P74 is often added to the MSS that have
(d), (e) probably comes from (d), perhaps by way of the Latin
videbatur (in d). (c), which must be translated ‘Where we were
accustomed to pray (to be in prayer)’, involves an unusual use of the
dative. Ropes (Begs. 3.155) prefers (d) with its less usual use of
 43. PAUL AND SILAS AT PHILIPPI. 16.11-40. 781

νομίζομαι: ‘Where a place of prayer was accustomed to be’, that is,

‘We went to an area where it was customary to find a place of
prayer’. For this (Ropes thinks) (a), which yields a similar sense, was
substituted: ‘Where we supposed there was a place of prayer’. Luke
however elsewhere uses νομίζειν in the active (Lk. 2.44; Acts 7.25;
8.20; 14.19; 16.27; 17.29; 21.29; the only exception is Lk. 3.23, and
this is a genuine passive and does not mean ‘is accustomed’), and this
may tip the scales in favour of (a), which is accepted with some
hesitation by Metzger (447), who thinks that ἐνόμιζεν in P74 may
testify to an earlier ἐνομίζομεν, and that προσευχή in P74 A B may
have arisen through the omission of the horizontal stroke in —χῆ (=
-χην). For νομίζειν cf. Josephus, War 7.128, 155; 2 Macc. 14.4.
Whether (a) or (d) is accepted, it seems to be implied that the
visitors expected to find a προσευχή near the river. προσευχή in the
NT is usually prayer, but it is used also for place ofprayer, often but
not necessarily a building. In what is described in ND 3.121 as the
‘earliest mention of a synagogue’ (CIJ 2.1440) the word used is
προσευχή. Josephus, Life 277 is explicit (συνάγονται πάντες εις τήν
προσευχήν, μέγιστον οίκημα) and the word was borrowed in Latin
(Juvenal, Satire 3.296, ede ubi consistas, in qua te quaero pro-
seucha?). See Philo, Legatio 132, 152, 346, 371; also NS 2.424-7.
The question whether it was customary to establish places of prayer
in the vicinity of water is discussed in Begs. 4.191 and NS 2.440-42.
That this was a universal practice cannot be proved, though Jose-
phus, Ant. 14.258 comes near to asserting it when he quotes a decree
of Halicarnassus permitting the Jews to make προσευχαὶ πρός τῇ
θαλάττη κατά τὸ πάτριον ἔθος. Cf. Ant. 12.106, with R. Marcus’
note; also Ep. Aristeas 304f.; PTebt 86.16-20. StrB 2.742 note, ‘Der
Brauch, die Gebetstätten in der Nähe von Gewässern zu errichten,
wird in der rabbin. Literatur nicht erwähnt’, though Mekhilta on
Exod. 12.1 notes that in the OT the word of God was given to
prophets in the vicinity of water (Dan. 8.2; 10.4; Ezek. 1.3). I.
Elbogen, Der jüdische Gottesdienst (1931/1967) 448 concludes, ‘Es
ist sehr unwahrscheinlich, dass selbst in der Diaspora die Synagogen
überall am Wasser lagen.’ See also Sukenik (Synagogues 49f.), who
speaks more positively. ‘Although official Judaism has preserved no
trace of a precept to that effect, there is abundant evidence that Jews
in Hellenistic countries built their synagogues by preference in the
proximity of water.’
It is not stated by Luke whether the place of prayer in Philippi was
a building; it is probable that it was not, since Luke frequently uses
συναγωγή for the building and his choice of προσευχή here may be
deliberate. The distinction made by Bowker (Targums 10) between
συναγωγή and προσευχή has to do not with building but with the
purpose for which the building or site was used; it is hardly
established by the evidence. For identity of the two see Μ. Hengel,
782 COMMENTARY ON ACTS

FS Kuhn 157-84. That the προσευχή was outside the city gate does
not prove that it was not a building; Jews were sometimes obliged to
practise their rites outside the walls.
The women who gathered (συνελθούσαις) at the place of prayer
were presumably Jewish (by birth, or proselytes) or at least Jewish
sympathizers, and they had presumably come to pray. ‘Man durfte
sie [synagogues] nicht einmal betreten, wenn man nicht beten wollte’
(Elbogen, 452). Elbogen notes a good many exceptions to this rule,
though if the προσευχή was not a building the women will not have
come there for shelter ‘bei Regenwetter’ or ‘brennendem Son-
nenschein’ (449). For Luke’s interest in and concern for women, and
in the conversion of women to Christianity, see 1.14; 8.3, 12; 9.2;
16.1; 17.4, 12, 34; 18.2, 26; 21.5; 22.4. For the ‘prominence of
women in Asia Minor’ see Trebilco (104-26).
καθίσταντες ἐλαλούμεν does not suggest a formal synagogue
service (contrast 13.5, 14f.), though sitting was a natural posture for
teaching.

14. Among the women was one called Lydia, a fairly common
name, used also in Latin (Horace, Odes 1.8). It may but need not be
significant that Thyatira (see below) was situated in Lydia. For the
name see Hemer (114,231), with the references also to ND 2.25-32;
3.54. The ethnic appellation might suit a freedwoman, the name
having been given while she was a slave—ή Λυδία (the one from
Lydia). But as Hemer points out the name can now be shown to have
been used by women of higher class. It occurs as a second name,
Νεωνἰς ή κ[α]ί Λυδία (ΤΑΜ 3.661).
Lydia was a dealer in purple from Thyatira. A large group of
compound words (LS 1451f.) based on πορφύρα, the purple-fish
(Musca trunculus or Purpura haemastoma), also the purple dye
obtained from it, bears witness to the importance of the dyeing
industry and of crafts and trades related to it. W. Μ. Ramsay (in the
HDB article, 4.757-9, still an excellent account, to be supplemented
by ND 2.25-32), however, accepts the view that ‘the dyeing in
Thyatira was performed in ancient times with madder-root, rubia...,’
and that ‘the purple stuffs which the Thyatiran Lydia sold in Philippi
(Acts 16.14) were dyed with what is, in modern times, called “Turkey
red’’ ’ (759). It is not clear whether Lydia was a sort of commercial
traveller in purple cloth, who visited Philippi frequently enough to
know her way to the place of prayer, or had opened a retail establish-
ment there. A fragmentary Latin inscription from Philippi appears to
refer to dealers in purple (CIL 3.664.1: pu]rpurari[us or i]). ND 2.27
gives examples from elsewhere. CIG 2519 implies the feminine form
used here, πορφυρόπωλις, and thus confirms the activity of women in
this business; PFlor 71.461, sometimes quoted, appears to read
πορφυροπώλου. For another business woman in the NT cf. Chloe
 43. PAUL AND SILAS AT PHILIPPI. 16.11-40. 783

(1 Cor. 1.11). The connection of Macedonia with Thyatira is illus-

trated by IG 10.2.1.291, in which ή συνήθεια τῶν πορφυροβάφων,
the gild of dyers of purple in Thessalonica, honour Μένιππον
Άμ(μ)ίου τόν καί Σεβήρον Θυατείρηνον.
The history of Thyatira is of greater importance for the inter-
pretation of Rev. 2.18-29 than of the present passage. Several local
inscriptions refer to gilds of dyers. In πόλεως Θυατείρων the
genitive of the place name is in apposition with πόλεως. A woman
trader appears as early as Homer, Iliad 4.14'lf.
σεβομένη τόν θεόν: for this term see the note on 10.2. It may be
that she was neither a born Jew (her name is not Jewish) nor a
proselyte but an adherent of the synagogue, a Jewish sympathizer.
On the other hand, the words do not necessarily mean more than that
she feared God; she was a devout woman. Wilcox (JSNT 13 (1981),
111) thinks that Paul would have been less likely to stay with her (v.
15) if she had been ‘a mere “adherent” and not a full member of the
synagogue’. As the informal meeting proceeded, ήκουεν, she was
listening; the imperfect here cannot refer to ‘ein häufigeres Hören’
(Delling, Studien 305); Delling mentions this only as a possibility,
and is right in the observation that no significant interval can have
intervened between the opening of Lydia’s heart and her baptism (v.
15). Lydia hears, but it is the Lord who opens her heart. For the
theological issue raised here see 13.48; cf. Stauffer (Theologie 163).
προσέχειν must in this passage (cf. 8.6) mean more than pay
attention to; something more like believe (Begs. 4.192). Cf. Jose-
phus, Apion 1.2, where προσέχειν is the counterpart of άπιστεῖν.

15. There is no indication that any period of instruction followed

Lydia’s hearing of the Gospel and preceded her baptism. On baptism
in Acts see Introduction, pp. xcif.; there is no reference here to the
name or to the use of water. When Lydia was baptized so was (πᾶς,
D pc (gig) w sams bomss) ὁ οἰκος αυτής. On the term household and
its meaning in such contexts as this see on 10.2; the use of οἶκος
cannot prove or disprove that small children were baptized. See
Delling (Studien 288-310) as well as the studies by Jeremias and
Aland.
For a woman as head of a household see Delling (Studien 302)
quoting Philostratus (Gymnasticus 23) ἐπί μητρί δὲ εἶναι τόν
οἶκον—after her husband’s death. See also ND 5.108.
BDR § 328.2, n. 3 would expect παρεκάλει rather than
παρεκάλεσεν (but cf. v. 39). Μ. 3.65 would translate ‘she invited’.
This does not seem to. be the sense that Luke gives to παρακαλεῖν,
but it agrees with παρεβιάσατο at the end of the verse. For the
omission of pronouns—παρεκ. (ἡμας) λεγ. (ἡμῖν)—see BDR §
278; it is correct style.
πιστήν seems to contain both a substantive and an adjective: if you
784 COMMENTARY ON ACTS

have judged me to be a believer, and a true, genuine, faithful,

believer.
εἰσελθόντες (plural, all the travellers were invited) εἰς τόν οἶκόν
μου. Unless this means, ‘Stay in, as members of, my family, my
household’, the word οἶκος has changed its meaning in the verse. It
probably does now mean dwelling house. In ND 4.93 papyri are used
to analyse the cases in which a woman owns the principal house of a
family. In some cases the husband has died and left the house to his
widow; it is much more frequently left to the children, though the
widow may be given the right to occupy the house. Infrequently, a
husband is found to reside in a house owned by his wife. The most
frequent case is that of divorce in which the agreement is that the
divorced wife shall own and occupy the house. ‘If the above factors
may legitimately be applied generally to the Mediterranean world, it
is quite likely that Lydia, Paul’s first convert in Europe, who appears
to be the head of a household (Acts 16.14, 15, 40), was in fact
divorced’ (D. C. Barker). This is possible, perhaps unlikely. It may
be that we should see here an early stage in the growth of churches
(ἐκκλησίαι) that met in people’s houses (e.g. Rom. 16.5).
παρεβιάσατο: a Lucan word, Lk. 24.29 and the present verse only
in the NT. It may denote physical or moral constraint; here clearly the
latter. Cf. Gen. 19.9; 1 Sam. 28.23,2 Kings 2.17. Luke is interested in
the moral consequences of conversion, and in hospitality.

16. ἐγένετο δέ ... παιδίσκην ... ύπαντήσαι. For the construction,

and Luke’s ways of using his (biblical) (καί) ἐγένετο (δέ) see
Introduction, p. xlvi.
ἡμών ... ἠμῖν. The construction begins as a genitive absolute but
runs into the dative; not one of Luke’s best sentences. It is natural to
interpret the first person pronouns to mean that the narrative is, or is
represented as being, reported by one who was present. Alterna-
tively, the pronouns may be proper to the framework of the story (the
itinerary) and have been imported from this into the episode,
originally an independent piece of tradition inserted by Luke into the
itinerary; so Bultmann (Exegetica 420). On the ‘we-passages’ in
general see Introduction, pp. xxv-xxx. The enriching of an outline
itinerary by incidents discovered in local tradition is a plausible
account of Luke’s method.
For the προσευχή see on v. 13.
παιδίσκη has various meanings (see I.584 and John 526); at Jn
18.17 the word must mean maidservant, or possibly female slave.
Here it must mean slave, since the girl has κύριοι. Prostitute, well
attested in earlier Greek, is not impossible, but would be unsuitable
for Rhoda, the παιδίσκη of 12.13. The Philippian παιδίσκη is said to
have had a πνεύμα πύθωνα; so P74 A B C* D* 81 326 pc,
πύθωνος is read by P45 C2 D1 E Ψ <a> πύθων was originally the
 43. PAUL AND SILAS AT PHILIPPI. 16.11-10. 785

name of the snake, or dragon, that inhabited Delphi (originally,

Pythia). This (a symbol or representative of the underworld) was
killed by Apollo, who thus acquired the title Pythian Apollo; his
priestess, who delivered the oracles at Delphi, was called the Πυθία
(sc. ἱέρεια). There may be some connection with πυνθάνεσθαι
(verbal stem πυθ-), to inquire (e.g. of an oracle). The word came,
however, at the beginning of the imperial period (for evidence see
especially W. Foerster, TWNT 6.917-19), to mean ventriloquist
(ἐγγαστρίμυθος). A person capable of this act might well be
believed (see Plutarch, De Defectu Oraculorum 9 (414e)) to be an
instrument through whom a god or spirit spoke; indeed, ἐγγαστ.
seems to have been used both of the human instrument and of the
deity or spiritual being using it. The word ἐγγαστ. sends us further to
1 Sam. 28.7, where it is used of the Witch of Endor (Hebrew,
); ἐγγαστ. becomes in the Vulgate python or spiritus
pythonicus, and the Hebrew is explained (Sanhedrin 7.7 (not
7.2 as in Foerster, who also gives other examples)) as
( = πύθων). It is best to suppose that Luke is describing
such a person, and himself held the popular view (which the girl
herself may well have shared) that the strange natural phenomenon
was due to possession; he evidently believes (v. 18) that some being
was driven out of the girl, and that this being had supernatural
knowledge of what Paul and his colleagues were doing (v. 17). The
construction remains unclear. The reading with the genitive is
probably an attempt to ease it: a spirit of Python. The accusative is
probably to be taken as in apposition: a spirit, a python; or, a spirit
called Python. Or we may (see BDR § 242 n. 2) take the whole to
mean a pythonic spirit. There is little difference between the first and
the last of these. For all its prophesying it is not a good spirit; cf.
Clementine Homilies 9.16: και πύθωνες μαντεύονται, ἀλλ’ ὑφ’
ἡμών ὡς δαίμονες ἐκριζούμενοι φυγαδεύονται.
The possessed girl provided much έργασία—business, or profit.
The meaning business is well supported by papyri (MM 252), but it
was no doubt the profit arising from it that interested the girl’s κύριοι
(who could have been man and woman—Μ. 3.22), and in a different
way Luke, who disapproved of profit made by the abuse of spiritual
agencies (see Kremer, Actes 287-91). The profit arose through the
girl’s μαντεύεσθαι, a word used here only in the NT; thus never of
OT or Christian prophets. Here, by giving oracles. It confirms the
interpretation of πνεύμα πύθωνα given above and reappearing in
Suidas, for whom πύθων is δαιμόνιον μαντικόν.

17. κατακολουθούσα (P45(*) P74 B D 36 453 pc), present

participle, goes better with ἐκραζεν, imperfect (the aorist ἐκραξεν is
read by very few MSS), than κατακολουθήσασα (A C Ε Ψ <a> co).
This might make it the textually inferior reading (a copyist’s
786 COMMENTARY ON ACTS

‘improvement’), but the combination of Old Uncial and Western

MSS means that it is probably correct and part of a harmonious
picture of a repeated event (cf. v. 18). There is no point in the
observation that ventriloquists are unable to make loud noises
(κράζειν); Luke was not recording decibels, and believed that the
girl was genuinely possessed (v. 18), though by an evil spirit. The
spirit has supernatural knowledge of the Gospel; cf. the knowledge
shown by the demons in the Synoptic Gospels (e.g. Lk. 4.34).
She followed Paul and us; Weiser (435) notes that though ἡμῖν is
used Paul is the only person who matters. ‘Das Wir nur noch
angehängt ist.’ He infers that in this verse the first person plural
comes not from the source but from the author. There is little force in
this argument; on any showing Paul was the most important person
in the group.
The omission of άνθρωποι by the Western text (D* gig Lucifer)
gives a more exclusive identification: not ‘These men are slaves’ but
‘These men are the slaves ...’. This was probably not intended by
Luke.
ὕψιστος is a designation of God common in the LXX of the
Psalms (rendering nearly always and of Daniel (rendering
and ); it is also frequent in Sirach. It is also used of Zeus (e.g.
Aeschylus, Eumenides 28f.: ... καλοῦσα καί τέλειον ὕψιστον Δία,
ἔπειτα μάντις ές θρόνους καθιζάνω—spoken by the Pythian proph-
etess). It occurs in Philo and Josephus, but its use in the Hellenistic
synagogue has been variously assessed. Thus G. Bertram, in TWNT
8.616f.: ‘Die weite Verbreitung und der populäre Gebrauch des
Gottesnamens der Höchste im hellenistischen Judentum ergibt sich
aus Synagogeninschriften mit der Widmung τῷ ύψίστῳ θεῷ, zB aus
Athribis in Ägypten Dittenberger OGIS 1.96.5’ (in fact, line 7;
Bertram gives the date of the inscription as about AD 200; it may be
somewhat later). See also Μ. Hengel (Judentum und Hellenismus
(1969), 545f., n. 244): ‘Im Gebrauch der Synagoge trat die Verwen-
dung dagegen wegen der Gefahr eines synkretistischen Missver-
ständnisses zurück, so erscheint der Begriff bei den Inschriften in
Rom überhaupt nicht.’ One may conclude that the word was used
sporadically in Hellenistic Judaism. For the interpretation of Greek
and Jewish usage and belief see especially C. H. Roberts, T. C.
Skeat, and A. D. Nock, ‘The Gild of Zeus Hypsistos’ in HThR 29
(1936), 39-88, with Hemer (231), ND 1.25-29, and Trebilco
(127-44). A resident in Philippi, with no first-hand knowledge of
Judaism, might well identify the one Jewish God with the highest
god in his own pantheon. For slaves (δoῦλoι) of God see 2.18; 4.29.
The word is Pauline (e.g. Rom. 1.1), and goes well with the
description of God as exalted in majesty and power.
The servants of God announce (καταγγέλλουσιν; for the word cf.
1 Cor. 11.26) to you (ὑμῖν: the spirit speaks; ἡμῖν, in A C Ψ <a> e
 43. PAUL AND SILAS AT PHILIPPI. 16.11-10. 787

sa is thus a natural but in fact inappropriate ‘correction’) οδόν

σωτηρίας. For οδός see on 9.2, for σωτηρία on 4.12. οδός here
however will refer not to a manner of life but to the way to, that is the
way to acquire, salvation. This answers in advance the question of v.
30, τί με δεῖ ποιεῖν ἵνα σωθώ; implying the reply given by Paul and
Silas, πίστευσον ἐπί τόν κύριον ’Ιησούν, καί σωθήση (ν. 31). See
below.
For καταγγέλλουσιν D(*) (so also the Peshitto, though this is not
noted in NA26) has ευαγγελίζονται. There is no difference in
meaning. D has the slightly more common but with οδόν σωτηρίας
slightly less suitable word. It is not correct to say (BDR § 309.1, n. 2)
that D* attests the active of the verb εὐαγγελίζειν. It is true that this
MS has what appears to be the plural of the active participle
εὐαγγελίζοντες, but this is probably an attempt to make some sense
of εὐαγγελίζοντε, an itacistic error for ευαγγελίζονται.

18. τοῦτο δέ ἐποίει (imperfect, cf. ἔκραζε in ν. 17), the process

was repeated, and for a considerable though undefined period.
διαπονηθείς: cf. 4.2. POxy 4.743.22, ἐγώ ὃλος διαπονούμαι is
rendered by the editor ‘I am quite upset’, but, especially in the aorist,
the word suggests 'I have reached the end of my patience’. Paul put
up with the girl’s behaviour as long as he could but at length could
stand it no longer. The suggestion (Kosmala 339) that Paul was angry
because the use of οδός (v. 17) suggested Essenism rather than
Christianity is fanciful and unconvincing; Paul was well aware of the
Christian use of Way (9.2)—at least, the author of Acts thought that
he was. Paul turned to the spirit, that is, to the girl possessed by the
spirit; clearly Paul (Luke) understood the ventriloquial phenomenon
to be the result of possession.
D (ἐπιστρ. δέ ό Π. τω πν. καί διαπον.) gives a less logical (and
perhaps therefore textually preferable) order, though it would be fair
to say that the two aorist participles signify more or less coincident
action: he turned impatiently and said ...
παραγγέλλω certainly means ‘Here and now I order you’, and
in this sense may be called an aoristic present: Μ. 1.119; BDR § 320
n. 2.
For the significance of the name see 1.198-200; here, with the
authority of.
ἐξελθεῖν, infinitive following upon a word of commanding (BDR
§ 392.1d). The progressive disuse of the infinitive is reflected in the
reading of D e gig, ἵνα ἐξέλθης, Μ. 1.240.
The response is immediate; the demon, or spirit, went out αὐτη τη
ώρα. There can be no doubt that Luke means at once; D rightly
interprets in the variant ευθέως. Black (AA 109-111) sees behind this
an Aramaic construction with proleptic pronoun, The Greek
is discussed by Μ. 2.432 (noting that here ‘Semitic sources are not in
788 COMMENTARY ON ACTS

question’; it is a ‘mannerism’; a ‘Lucan idiom’; cf. Wilcox 130) and

by Moule (IB 93, 122). There seems to be no good reason why we
should not translate ‘at that very moment’. LS (282 b top) quote
Thucydides 2.3.33, αὐτὸ τὸ περίορθρον, at the point of dawn.
The incident is clearly regarded by Luke as an exorcism (cf. e.g.
Lk. 4.35); to take it as an example of pre-baptismal exorcism
(Richardson, Theology 338; cf. 8.7) might be slightly more convinc-
ing if Luke went on to describe the baptism of the girl!

19. ἰδόντες δέ (B has καί ἰδ.) οἱ κύριοι αυτής (see ν. 16). D avoids
the participle and the pronoun, writing ὡς (for temporal ὡς see BDR
§ 455.2) δέ είδαν οἱ κ. τής παιδίσκης.
When the spirit went out the hope of profit also departed
(ἐξῆλθεν). Again D has the same sense in different words:
ἀπεστερήσθαι τής ἐργασίας αυτών, ἧς εἶχον δι’ αυτής. Loss of
profit is treated by Luke as the real cause of the action taken by the
girl’s owners; the charge they bring in v. 21 is thus a falsehood, and
Luke intends that it should be seen as such. Cf. the motive of anti-
Christian activity in 19.25. For Luke’s interest in money and the
abuse of money see e.g. I.413.
ἐπιλαμβάνεσθαι is a Lucan word (Lk. 5 times; Acts 9.27; 16.19;
17.19; 18.17; 21.30, 33; 23.19; in the rest of the NT 7 times, used in
good senses and bad. It is commonly used with the genitive, and the
present verse is not an exception in that the accusative τόν Π. καί τόν
Σ. is the object of εἱλκυσαν; so BDR § 170.2, n. 2 (cf. 9.27; 18.17).
For the resulting economy in pronouns see BDR § 278 n. 3. Silas was
last mentioned by name at 15.40; presumably he is to be included in
the first person plurals at 16.11,12,13,15,16,17. There is nothing to
suggest that the author of the ‘We-source’, or Timothy (if he was not
that author) was arrested too.
For εἵλκυσαν (cf. Epictetus 1.29.22, ἕλκει μ’εἰς τήν αγοράν) Ε has
ἔσυραν without significant difference of meaning.
For the procedure of the arrest and the charges brought (vv. 20f.),
see Sherwin-White (78-83).
εις τήν αγοράν: ‘ ‘ ‘The courthouse’ ’ rather than “the market-
place” would give the meaning, but the word is better transliterated
than translated’ (Begs. 4.194). The earliest meaning seems to be
assembly, especially an assembly of the people rather than a council
of chiefs (e.g. Homer, Iliad 2.93; Odyssey 2.69—cited LS 13);
thence it became place of assembly, but especially market-place. The
market-place of a Greek city, however, was much more than a place
for buying and selling (cf. 17.17, and passages there cited); it was
used for all kinds of public purposes, including judicial purposes. So
e.g. Demosthenes 44.36(1091), ἐν τή τῶν αρχόντων αγορᾷ; Lucian,
Bis Accusatus 4, προτίθεμεν αὐτοῖς αγοράν δικών ...; 12. Here the
άρχοντες (cf. Demosthenes, above) were to be found. The word was
 43. PAUL AND SILAS AT PHILIPPI. 16.11-40. 789

a general one and perhaps never quite ceased to suggest the present
participle of ἄρχειν: the αρχών was one who ruled, though the word
had particular associations in e.g. Athens. It can be equivalent to
praefectus (Polybius 6.26.5); here, magistrate. For the question
whether the άρχοντες are equivalent to the στρατηγοί see on v. 20.

20. They brought them to the magistrates, τοῖς στρατηγοῖς. Are

the στρατηγοί of this verse identical with the άρχοντες of v. 19? See
Begs. 4.195, with the references to Meyer and especially to Ramsay.
στρατηγός is a standard equivalent for the Latin praetor, though the
original sense of the Greek word was military, of the Latin judicial.
Under the Empire the office of praetor intervened in the cursus
honorum between that of aedile or plebeian tribune and the consul-
ship, and it was the duty of praetors to preside in the courts and
supervise the public treasury. But praetor was no longer (see
Sherwin-White 92) the term normally in use for the magistrates of
colonies; these were duoviri (fully duoviri iuri dicundo; occasionally
quattuorviri), though earlier the term praetor was sometimes con-
nected with the office; cf. Cicero, De Lege Agraria contra Rullum
2.34(93), ‘cum ceteris in coloniis duumviri appellentur, hi ac prae-
tores appellari volebant’ (the passage refers to Capua). στρατηγός
was thus a not unnatural word to use in Greek, especially in view of
the fact that it is virtually impossible to find a literal translation of
duumviri. The question however remains: Does Luke mean (a) they
brought Paul and Silas to the rulers, and then (perhaps because told to
do so by the rulers) took them on to the magistrates, or (b) they
dragged them to the magistrates (described as άρχοντες) and in
doing so brought them to the magistrates (described with their
judicial function more clearly expressed as στρατηγοί)? (a) requires
us to find different functions for άρχοντες and στρατηγοί, which is
not easy; (b) seems a pointless repetition (though Begs. loc. cit.
affirms that such repetitious variation is a Lucan characteristic; no
references are given but we may perhaps think of those who are
described in 20.17 as πρεσβύτεροι and in 20.28 as επίσκοποι).
On the whole, since the repetition in the second clause includes the
verb (εἵλκυσαν—προσαγαγόντες) as well as the noun, it seems
better to accept (a), giving άρχοντες some such meaning as ‘the
leading people’.
The first charge against the Christians is that of causing a
disturbance (cf. 17.6, οἱ τήν οικουμένην ἀναστατώσαντες; 24.5,
κινοῦντα στάσεις). For ἐκταράσσειν cf. Plutarch, Coriolanus 19
(223) ... τόν τε δήμον αὖθις ού παρέξουσιν ἐκταράττειν τοῖς
δημαγωγοῖς (tribunes); Aristophanes, Knights 863(867) ... σὺ
λαμβάνεις, ἢν τήν πόλιν ταράττης. Paul and Silas are described as
Jews (and therefore men who might be expected to cause trouble?),
but the substance of the charge is given in the next verse.
790 COMMENTARY ON ACTS

21. The remainder of the charge ‘so far from being anachronistic is
positively archaic’ (Sherwin-White 82). It implies that it was illegit-
imate for Romans (members of a colonia) to adopt foreign customs,
especially Jewish customs (v. 20). This was in accordance with the
ancient principle that Roman citizens must practise the state cult, and
might in addition practise only those cults that had been sanctioned
by the Senate—religiones licitae. This principle was however
relaxed in practice in the early Empire, and no objection was made to
religions that did not offend against public order and public morality.
Here however the ‘principle of incompatibility’ (Sherwin-White 80)
is invoked (the dative 'Pωμαίοις σὖσιν is to be taken with ἔξεστιν
ἡμῖν rather than with the infinitive: BDR § 410 n. 2). For this revival
of a mostly forgotten principle there may have been two reasons (a)
‘It is perhaps characteristic that it is in an isolated Roman community
in the Greek half of the Roman Empire that the basic principle of
Roman “otherness” should be affirmed, whereas in Italy the usual
custom prevailed of treating alien cults on their merits’ (Sherwin-
White 82). (b) The reference to the fact that Paul and Silas were Jews
may be significant. Roman policy was to be tolerant towards Jews in
the practice of their religion, but there is some ground for thinking
that there was at this time a reaction against any kind of proselytiza-
tion. See A. Momigliano, Claudius (ET 1934), 29-35. Sherwin-
White (81) is right to question ‘whether there was any precise
enactment against proselytism’; the evidence is not strong enough to
affirm this. But the distinction between a national religion and
attempts to turn this into a missionary religion is certainly in line
with general imperial policy, and the juxtàposition of ’Ιουδαῖοι
υπάρχοντες (v. 20) and 'Pωμαίοις σὖσιν suggests that Paul and his
companions were accused of illicit prosleytizing. See the note on
18.2. This raises the question whether Paul did in fact present his
Christian message as a version, the best version, the only true
version, of Judaism. Acts sometimes if not always suggests that he
did. He and Silas proclaimed ἔθη; on this word see on 15.1. See
W. C. van Unnik, Sparsa Collecta I, NovTSupp 29 (1973), 374-85.

22. D (καί πολύς όχλος συνεπέστησαν κατ’ αυτῶν κράζοντες)

makes the scene more vivid, but is not necessarily secondary. The
plural verb and participle, though defensible (with a noun of
multitude), are not likely to have arisen out of the συνεπέστη of the
other MSS. But the editor may have felt that κατ’ αυτών required a
supplement, ουνεφιστάναι does not occur elsewhere in the NT, but
ἐφιστάναι (a Lucan word: Lk. 7 times, Acts 11 times, the rest of the
NT twice) at 4.1 in a similar sentence takes a simple dative. For the
importance of the crowd here see Rapske (Book of Acts 3.121-3).
The magistrates react strongly, περιρήξαντες αυτών τά Ιμάτια.
Whose clothes? Begs. 4.195 points out that throughout the sentences
 43. PAUL AND SILAS AT PHILIPPI. 16.11-40. 791

αυτών (first occurrence), αὐτοῖς, αυτούς refer to Paul and Silas,

leaving some likelihood that the second αυτών also will do so.
Moreover, for judges to tear off their own clothes in horror is a
Jewish (e.g. Mk 14.63) rather than a Roman or Greek custom
(noticed already by Calvin, 2.82), whereas to tear off an offender’s
clothes before beating him has many parallels (e.g. Demosthenes
19.197(403): περιρρήξας τὸν χιτωνίσκον ὁ οἰκέτης ξαίνει κατά τού
νώτου πολλάς; Plutarch, Publicola 6(99) ... συλλαβόντες τούς
νεανίσκους περιερήγνυον τἀ ἱμάτια, τάς χεῖρας ἀπῆγον ὀπίσω,
<b>άβδοις ἔξαινον τά σώματα. On the use of single or double ρ see Μ.
2.101f. (with a reference to 14.14, διαρρήξαντες), 192f.; BDR §
11 1, n. 2.
ἐκέλευον: the aorist might have been expected. At Μ. 3.65 Turner
notes this; in Insights 96f. he says, ‘St. Luke carefully avoids saying
that they had explicitly and peremptorily ordered the flogging. The
tense of his verb represents a mere pretence at command, satisfying
the plaintiff commercial interests, and yet not constituting a breach of
Roman law. The magistrates possibly hoped that the gaoler might
have the good sense not to carry out what they had diplomatically but
half-heartedly commanded.’ He refers to the Western text of v. 39
(see the note there for his rendering). A more valuable point in v. 39
however is the use there of the aorist παρεκάλεσαν, where the
imperfect παρεκάλουν would have been more natural. The simple
fact is that Luke was not always as careful about his tenses as he
might have been. One cannot do more than note an exception, as do
BDR § 328.1, n. 2, ‘κελεύειν im Impf. nur Apg 16.22 ἐκέλευον
<b>αβδίζειν statt ἐκέλευσαν wie Vg iusserunt.’ Turner’s suggestion is
inconsistent with v. 24 (παραγγελίαν τοιαύτην).
<b>αβδίζειν, active, not passive; the agent and object must be
understood: ‘they commanded the officer to beat them.' See BDR
§ 392.4, n. 14. The present infinitive (taken with the aorist participle
ἐπιθέντες in v. 23) may be noted to Luke’s credit: the beating went
on until many blows had been inflicted. The verb describes a Roman
punishment; cf. 2 Cor. 11.24f., where it is distinguished from the
Jewish synagogue beating. It should not have been inflicted on
Roman citizens; see on v. 37 (and 2 Corinthians 297, on 2 Cor.
11.25).

23. The sentence, which began to falter a little in v. 22, since the
στρατηγοί, though they gave the command (ἐκέλευον), are unlikely
themselves to have tom the clothes off the backs of Paul and
Barnabas, now causes further difficulty. Only indirectly can
ἐπιθέντες refer to the στρατηγοί, ἔβαλον may be either ‘the
attendants physically put them in prison’ or ‘the magistrates com-
mitted them to prison’, but παραγγείλαντες must refer to the
magistrates who are giving an order to the δεσμοφύλαξ. The general
792 COMMENTARY ON ACTS

sense is perfectly clear; so perhaps is the grammar on the legal

principle qui facit per alium facit per se.
If Turner rightly understood the imperfect ἐκέλευον (v. 22), the
attendants certainly misunderstood their orders, for they applied
πολλάς πλήγας. The expected aorists take over with ἐπιθέντες
(contrast <b>αβδίζειν, and see on v. 22), ἔβαλον, and
παραγγείλαντες.
δεσμοφύλαξ, gaoler; LS’s earliest citation (380) is BGU
4.1138.12 (1st century BC). BA 352 refer to but do not specify
papyri from the 3rd century BC and add further literary references,
including Josephus, Ant. 2.61. The word occurs in Lucian, Toxaris
30; Lucian’s story has a few superficial parallels with Luke’s.
Beyer (102) observes that it would be possible to pass directly
from this point to v. 35, leaving the possibility that Luke might have
inserted the supernatural story of the earthquake into a much more
matter-of-fact event.
The special instructions to the gaoler are scarcely consistent with
any weakening of the force of ἐκέλευον in v. 22. The active infinitive
τηρεῖν corresponds to <b>αβδίζειν in v. 22. Did Luke expect the agent
and object (τῷ δεσμοφύλακι and αυτούς) to be read back into the
earlier part of the sentence?

24. δς παραγγελίαν τοιαύτην λαβών ... To resume a sentence

and continue a narrative by means of a relative (ὅς evidently refers to
the δεσμοφύλαξ) is characteristic of Luke in Acts. D sa have the more
ordinary ὁ δέ in place of ὅς. The gaoler acts in accordance with the
command (such a command—not such as is suggested by Turner on
v. 22!) that he has received. He picks up the ἀσφαλῶς of v. 23. The
prisoners are put εἰς τὴν ἐσωτέραν φυλακήν. The use of ἀνάγειν
(literally, to bring up) in v. 34 is not enough to prove that this was an
underground dungeon; the first action of the gaoler (v. 30) is given as
προαγαγών, which might suggest bringing the prisoners forth from
an inner room. The reference to the inner prison both fulfils the
magistrates’ instruction and prepares for the miraculous earthquake.
The prisoners’ feet were secured (the gaoler ἠσφαλίσατο, refer-
ring back to ἀσφαλῶς in v. 23) εἰς τό ξύλον. It is not certain whether
the ξύλον is to be thought of as a means of security or an instrument
of torture. In the classical passages often quoted (Herodotus 6.75.2;
Aristophanes, Knights. 367, 394, 1049(1046)) it is used with δέω,
which suggests confinement (though Knights 394, and possibly 1049,
also suggest torture). Eusebius (HE 5.1.27; 8.10.8; possibly 4.16.3)
speaks of the ξύλον as a means by which torture was applied to
Christian martyrs. The simple explanation is no doubt correct: the
same instrument was used for both purposes. In HE 5 and 8 Eusebius
speaks of a number (5, 4) of holes (cf. the πεντεσύριγγον ξύλον of
Knights 1049); apparently the victim’s legs could be more or less
 43. PAUL AND SILAS AT PHILIPPI. 16.11-40. 793

widely extended and discomfort turned thereby into severe pain.

Luke does not suggest that Paul and Silas were being tortured; the
context stresses security (ασφαλώς ... εις τήν ἐσωτέραν φυλακήν
... ἠσφαλίσατο); and when in ν. 37 Paul complains of the treatment
that he and his colleague have received he says δείραντες ... ἔβαλον
εις φυλακήν, but adds no reference to further ill-usage.
The Western text (D pc) has ἐν τω ξύλω, an improvement on εις
τὀ ξύλον; cf. Aristophanes, Knights 394, ἐν ξύλω δήσας. gig Lucifer
have in nervo, but nervus seems to mean fetter.

25. Knox (Hell. El. 18) follows Rutherford (Phrynichus 36) in

describing μεσονύκτιον as poetical, and Ionic. Use by Plutarch,
Lucian, Vettius Valens, Strabo, as well as in the LXX (see LS 1107)
is sufficient background for its appearance in NT prose. Here the
article is used; contrast 20.7 (μέχρι μεσονυκτίου; for the omission
there see Μ. 3.179). At 27.27, κατά μέσον τής νυκτός is used; here
also by D.
προσευχόμενοι: a natural recourse for Christians in distress. In
12.5 the church prays for Peter while he sleeps. James 5.13 enjoins
prayer upon anyone who κακοπαθεῖ; Paul and Silas in addition
ὕμνουν τόν θεόν (cf. James’s ψαλλέτω for the man in good spirits).
ύμνεῖν without object occurs at Mk 14.26 (= Mt. 26.30, the singing
of the Hallel Psalms at Passover); a pronominal object replaces God
at Heb. 2.12 (quoting Ps. 22(21).23). For ὑμνεῖν τόν θεόν parallels
are given in ND 1.71, 72. ύμνος occurs at Eph. 5.19; Col. 3.16. The
non-biblical usage suggests the celebration of some god or hero, but
determinative for Acts is the use of ύμνος along with ψαλμός for the
Psalms of David (cf. especially 2 Chron. 7.6, ἐν ὓμνοις Δαυίδ,
). Paul and Silas were singing OT Psalms or new composi-
tions on the same lines (cf. the Qumran also 5.41). Singing in
prison may be (see Begs. 4.196f., quoting Reitzenstein) a literary
convention, intended to show the coolness, courage, and faith of the
prisoners. See Tertullian, Ad Martyres 2, Nihil crus sentit in nervo
cum animus in caelo est. But, bearing in mind that ἐπακροασθαι
occurs nowhere else in the NT, the best parallel appears to be
Testament of Joseph 8.5: καί ὡς ἤμην ἐν τοῖς δεσμοῖς ... καί
ἐπηκροᾶτό μου πῶς ὓμνουν κύριον ... (text β). But the Testament
may be dependent on Acts. Bruce (2.317) quotes also Epictetus
2.6.26, ἐσόμεθα ζηλωταί Σωκράτους, ὃταν ἐν φυλακή δυνώμεθα
παιάνας γράφειν. But Begs. 4.197 rightly goes on to say that this
often really happened. Luke is building up a highly dramatic but not
unique story. Whatever we make of the earthquake and release it is
true that men in prison for their faith have praised God; this is not the
point on which the historicity of Luke’s story will stand or fall.
έπηκρόωντο—because at night all were put into the inner cell?
Rapske, Book of Acts 3.203f.
794 COMMENTARY ON ACTS

26. ἄφνω is a Lucan word: Acts 2.2; 16.26; 28.6 only in the NT.
The sentence σεισμός ἐγένετο μέγας occurs also (no doubt by
chance) at Mt. 28.2; Rev. 16.18; cf. Rev. 6.12; 11.13. In the eastern
Mediterranean earthquakes were as familiar in antiquity as they are
today (e.g. Thucydides 3.87.2). Luke does not directly ascribe this
one to divine providence, though he will certainly not have thought it
fortuitous.
For σαλευθήναι see on 4.31. τἀ θεμέλια: the word (an adjective—
θεμέλιος λίθος) is normally treated as masculine. See Μ. 2.122 (‘...
τά θεμέλια here shows the collective sense; contrast oἱ θ. in Rev.
21.19’). See also Radermacher (52).
Unlike δεσμοφύλαξ (v. 23), δεσμωτήριον is classical.
παραχρῆμα is a Lucan word (Mt. twice; Lk. 10 times; Acts 6
times); it is probable therefore that its omission by B gig Lucifer is
accidental. Stories of the supernatural characteristically emphasize
the immediacy of divine action (see 3.7 and the note).
θύραι could refer to only one door, but it is probable that the
prison had several and that the plural is a real plural; πᾶσαι makes
this virtually certain. See BDR § 141.4, n. 8.
τὰ δεσμὰ appears as the plural of ὁ δεσμός here and at Lk. 8.29;
Acts 20.23. οἱ δεσμοί is used at Phil. 1.13. In other passages the
gender is not made clear by case and construction. Cf. the note on τά
θεμέλια above, and see Μ. 2.121f. In this case the neuter plural is
hardly a collective. There is an important note in Rutherford,
Phrynichus 353f.
ἀνέθη ( D* have ἀνελύθη) is probably correct and signifies
further supernatural events; we are not to think (with Ramsay, Paul
the Traveller 221) of staples loosened from cracks in the building, so
that prisoners presumably ran out still manacled and fettered though
no longer attached to the walls.
Origen (c. Celsum 2.34; cf. 8.41f.), prompted by Celsus’ allusion,
refers to the story of Dionysus in Euripides, Bacchae, quoting 498,
λύσει μ’ὁ δαίμων αυτός ὄταν ἐγὠ θέλω; see also 443-8, 586-8,
especially 447f.: αυτόματα δ’αὐταῖς δεσμὰ διελύθη ποδῶν κλῇδές
τ’ἀνήκαν θύρετρ’ ἄνευ θνητής χερός. See also Ovid, Metamor-
phoses 15.669-671. For other possible connections between Acts
and Euripides see 21.39; 26.14.
The question, discussed by Knowling (351), why all the prisoners
did not immediately run away, is one that would not occur to Luke.
There is a more profitable discussion in Begs. 4.196f. of the
difference and relation between magic and religion. The observation
that the earthquake may well have happened—earthquakes do
happen—and that prisoners sometimes escaped without under-
standing the circumstances of their escape is fair enough; to be
added to it is Luke’s conviction that a special providence protected
Paul as the special agent of God’s plan for the spread of the Gospel.
 43. PAUL AND SILAS AT PHILIPPI. 16.11-10. 795

2 Cor. 11.24-27, a list of perils from Paul’s own hand, contains

evidence of enough well-nigh incredible escapes to persuade a first-
century Christian to believe in miracles (though Paul does not ascribe
his escapes to miracle).
27. έξυπνος, awake, seems always to be used with γίνεσθαι, so
that γενόμενος is not to be regarded as a Lucanism (Begs. 4.198).
After ὁ δεσμοφύλαξ (v. 23), 614 pc add ό πιστός Στέφανός, a
Western elaboration, recalling the apocryphal acts and having no
ulterior motivation. For the name cf. 1 Cor. 16.15, 17.
σπασάμενος, middle, rightly; see BDR § 310.1, η. 1. τήν is
omitted by P74 A Ε Ψ <a>, but B C D 81 1175 pc, which have it, are
right. The gaoler drew the sword, that is, the one that he had, that was
part of his equipment.
νομίζων ... Presumably the gaoler would think that punishment
for allowing the escape of the prisoners would be an alternative
worse than suicide. Cf. 12.19. Again dramatic effect is heightened
(and the question of v. 30 prepared for), but whether the detail is a
probable one is questioned. Begs. 4.198 observes that the gaoler
would have had a reasonable chance of catching the prisoners, and
that the earthquake would have been regarded by responsible
authorities as a reasonable excuse. See Digest 48.3. Bauemfeind
(210) on the other hand thinks that the gaoler’s superiors would not
have believed the story about the earthquake. One would have
thought however that an earthquake strong enough to put the prison
out of commission would hardly have escaped the notice of other
residents in the town.
28. ἐφώνησεν μεγάλη φωνῇ is not a normal Greek expression for
uttering a loud cry (μέγα φωνεῖν would be better). It has an
OT ring, though in the OT φωνεῖν with the cognate dative occurs
only at Dan. 5.7 LXX, ἐφώνησε φωνή μεγάλη, where the Aramaic is
... ... .
With πράξης a double accusative rather than accusative and dative
(σεαυτῷ) would be expected; but cf. Lk. 6.27. See Radermacher (99
' ... in denen sich Lukas keineswegs als Attizist erweist’). Paul’s
concern for the gaoler’s life recalls Stephen’s prayer for those who
were stoning him (7.60).
How, in the middle of the night and in the midst of an earthquake,
Paul knows that all the prisoners are present is a question that Luke
does not need to ask. Rapske (Book of Acts 3.203f.) (see on v. 25)
answers the question by his suggestion that for security all the
prisoners had been put for the night in the inner cell. Paul knows that
God is ordering all things for good (and does not stop to consider
whether others, like himself, might have been falsely imprisoned).
This is not quite the same thing as claiming (Schille 347), ‘Paulus
erscheint wie ein Theios-anēr.’ The divine man boasts, 'I can release
796 COMMENTARY ON ACTS

myself when I choose;’ Paul knows that all things are in the hands of
a God who is other than himself.
There is much textual variation at the beginning over the verse,
and NA26 may be unwise in following the reading of A alone. In
other witnesses we have (for μεγάλη φωνῇ ὁ Παῦλος)
Π. μεγ. φωνή B pc
μεγ. φωνή Π. C* 33 pc
φωνή μεγ. Π. ρ74
φωνή μεγ. ό Π. C3DE <a>
ό Π. φωνή μεγ. 36 pc
29. The subject changes without notice; no longer Paul but the
gaoler.
Naturally the gaoler calls for lights, φώτα in the sense of
λαμπάδας is a late use. Begs. 4.198 raises the question whether
φώτα should be taken as the accusative singular of φῶς (by analogy
with ἐρῶτα, ἱδρώτα). LS (s.v. φάος, 1916) and BA 1738f. give no
example of this and the suggestion has little point since there is
evidence for the use of φώς as lamp. The gaoler sprang in,
presumably into the dungeon where Paul and Silas were confined.
For εἱσεπήδησει cf. ἐκπηδᾶν in 14.14; (on both verses see C.
Burchard, ZAW 69 (1978), 155). He was trembling, for έντρομος (cf.
Heb. 12.21) γενόμενος cf. ἔξυπνος γενόμενος in v. 27; υπάρχων
(C* D Ψ 614 2495 pc) may be a better reading, γενόμενος arising by
assimilation.
προσέπεσεν, he fell down before Paul and Silas. The Western text
(D* gig sy) heightens the colour by adding πρὸς τούς ποδάς. For
such respect to apostles and other preachers cf. 10.25; 14.11-13; see
on these passages. Here Paul and Silas make no comment, not
because they wish to be thought θειοι ἄνδρες (see on v. 28) but
because Luke is moving swiftly on to what he regards as the real
climax of his story in vv. 30f.
30. προαγαγών ... έξω. See on v. 24. We hear no more of the
earthquake, and the conversation between the gaoler and Paul and
Silas takes place outside the prison.
There is no need to take κύριοι as in itself more than polite
address, but Luke undoubtedly intends to portray a man impressed by
supernatural events which he connects with the persons to whom he
is speaking. This does not mean that he views them as θεῖοι ἄνδρες,
though Euripides and the Bacchae come once more to mind (see v.
26). If he did so view them he is immediately corrected by Paul and
Silas who (v. 31) point away from themselves to the truly divine
Man.
The gaoler asks, τί με δεῖ ποιεῖν ἵνα σωθώ; On σῷζειν in Acts
see 1.227-31. The word could be used by Luke in a purely ‘secular’
 43. PAUL AND SILAS AT PHILIPPI. 16.11-40. 797

sense (notably 27.20, 31), and it is possible to understand it in such a

sense here. The gaoler, facing the possible escape of all the prisoners
committed to him, is in such straits that he intends suicide; what can
he do to evade the consequences he fears? It is however impossible
that Luke should have used the word without awareness of its
specific Christian sense (see the reply in the next verse). Cf. the
movement from 4.9, σέσωται, he has been cured, to 4.12, No other
name by which δεῖ σωθήναι ἡμᾶς—we, who are not lame beggars.
The form of the question suggests that it may have been a formula,
but no real parallel suggests itself. Cf. 2.37. ND 3.99 quotes PVindob
G 25683, ... τἱ [...] ποιῇσαι ἵνα σωθώ ό ἁμοιρτολός This however
is a sixth-century papyrus, perhaps dependent on Acts.
The Western text if represented by D(c) (syp h**) gives at the
beginning of the verse: καί προήγαγεν αυτούς ἔξω τούς λοιπούς
ἀσφαλισάμενος καί εἶπεν αὐτοῖς. This longer reading holds up the
action with a wooden and prosaic addition, and is not to be defended
by the observation (Black, AA 69) that with v. 29 it forms inelegant
parataxis: καὶ ... προσέπεσεν ... καί προήγαγεν ... καί εἶπεν. The
editor filled in a gap, without recognizing that he was (on this
occasion) taking life and movement out of the story by putting more
detail into it.

31. Whatever may be said about the form of the gaoler’s question
(see on v. 30), Paul and Silas reply in terms that rest upon an
accepted formula; cf. Rom. 10.9, also Phil. 2.11. To believe on the
Lord Jesus (Ίησουν: so P74vid A B 33 81 pc gig vg bo; Western and
Antiochian texts, C D Ε Ψ 0120 <a>sy sa, add Χριστόν) means to
accept him as κύριος, that is, as supreme authority (see 2.36), and by
implication to believe (in the words of Rom. 10) that God raised him
from the dead, since otherwise the crucified Jesus could not be
κύριος. It is implied that the word κύριος (repeated perhaps with
intention from v. 30) would to such a man as the gaoler denote a
recognizable category, and this, it may be noted, would not be a
category defined in terms of the familiar κύριος—mm equivalence
of the OT. Jesus is presented as a cult figure, the origin of the οδός
σωτηρίας that Paul and Silas had preached (v. 17). A divine being
who would be insulted when his servants were publicly ill-used and
was powerful enough to send an earthquake to release them was one
to command allegiance. This is a form of Christian belief neither
profound nor exalted, but the historian should remember that this is
what Christianity must have looked like to many in the first century,
whether believers or unbelievers.
σωθήση: see on v. 30.
σὴ καὶ ὁ οἶκός σου. For οίκος cf. ν. 15. It would be difficult to
maintain that the word here includes infants, since not only were οἱ
αὐτοῦ ἅπαντες baptized (ν. 33), all heard the word of the Lord
798 COMMENTARY ON ACTS

spoken by Paul and Silas (v. 32) and as a result the whole household
rejoiced (ἠγαλλιἁσατο πανοικεί). On this sequence of terms see
Delling (Studien 301), though his conclusion lays the stress else-
where—the author of Acts intends to include all ‘die zu dem
Hausstand des Betreffenden gehören’. Here as in vv. 14f. the
conversion of the head of the household carries with it the members
of the household. For the relation of the salvation of the household to
the faith of one member cf. 1 Cor. 7.14-16 (on which see 1
Corinthians 164-7).
32. ἐλάλησεν αὐτῷ τόν λόγον τού κυρίου (so P45 P74 A C (D) Ε
Ψ 0120 <a>lat sy co; B pc have θεού); speaking the word of God
(no difference is intended if Lord is read; it goes with the wording of
v. 31) is one of Luke’s standard terms for preaching the Gospel. The
basic proposition of v. 31 is filled out in a way that Luke’s readers
could develop for themselves—from the book itself if in no other
way.
συν πᾶσιν (P45 has σύμπασι, presumably omitting αύτφ—there is
a lacuna before λόγον) ... See on v. 31. All who were in his house
will be synonymous with ὁ οἶκός σου. Luke must mean all those
who were capable of both hearing and understanding. He is not
thinking of small children.
33. The gaoler’s first reaction is to minister to those who are
manifestly favoured by providence and have offered him a way of
salvation—which presumably he has accepted, since this verse
recounts his baptism.
παραλαβών perhaps (with C. Burchard, ZNW 69 (1978) 156f.),
took them in as his guests; cf. 21.24, 26. ἔλουσεν από τῶν πληγών
will mean ‘washed their wounded bodies, washed off the effect of the
blows (πληγαί, v. 23) they had received’. The expression is unusual
and not fully explained by the parallels cited by Deissmann (BS 227;
cf. Dittenberger, Syll. 982.5-7). BA 975 gives no further help; από
των πληγών shows that religious (i.e. cultic) washing is not in mind
(though there may well be an intended play on the ἐβαπτίσθη
that follows). This was seen by Chrysostom (Homily 36.2): ἔλουσεν
αὐτούς καί ἐλούθη' ἐκείνους μἑν ἀπὸ τῶν πληγῶν ἔλουσεν, αὐτός
δέ ἀπό τών αμαρτιών.
οἱ αὐτοῦ πάντες (or, with X B 614 1891 2495 pc, ἅπαντες) is the
reading of the great majority of MSS, and almost certainly correct.
P45 vgcl boms have ό οἶκος αὐτοῦ ὅλος; this is probably an
assimilation to v. 31. A alone has οἱ οἰκίοι αὐτοῦ πάντες. There is no
difference in meaning between all these readings. They cannot be
used to prove that infants were baptized in the NT church. On
baptism in Acts see Introduction, pp. xcif. The gaoler acted imme-
diately for the relief of his prisoners (ἐν ἐκείνη τη ώρα τής νυκτός)
and his baptism followed at once (παραχρῆμα, a Lucan word; see on
 43. PAUL AND SILAS AT PHILIPPI. 16.11-40. 799

3.7). Whatever be the historical value of this account of events in

Philippi it may probably be inferred that Luke and his readers would
find nothing incredible, or improper, in a baptism following imme-
diately, without instruction, upon conversion and profession of
faith.

34. ἀναγαγών implies nothing about an underground dungeon

(see on v. 24). The prisoners have already been brought out of the
prison (ἔξω, v. 30); the gaoler now takes them εἰς τὸν οἶκον, to his
house, ‘up home’. It has been suggested that he had a flat above the
prison; this might have fared badly in the earthquake.
παρέθηκεν τράπεζαν, he prepared a meal (that is, what was set
on the table). The expression is not uncommon: Homer, Odyssey
5.92, ὡς ἄρα φωνήσασα θεά παρέθηκε τράπεζαν; Herodotus
6.139.3, τράπεζαν ἐπιπλέην αγαθών πάντων παραθέντες; Thucy-
dides 1.130.2; Josephus, Ant. 6.338. The most natural way of taking
the text is to suppose that the gaoler generously entertained men who
had been wronged, whom he revered, to whom he was indebted; and
rejoiced to do so (ἠγαλλιάσατο). Reicke, DFZ 217, however, argues
that this meal, following upon baptism, will have been a eucharist
(cf. 2.46, κλῶντές τε κατ’ οίκον άρτον, μετελάμβανον τροφής ἐν
ἀγαλλιάσει); so also Schille (348), supporting his view by the use of
τράπεζα (cf. Didache 11.9). This is unlikely. τράπεζα is too
common a word to prove anything, and ἀγαλλιᾶσθαι (Mt. 5.12; Lk.
1.47; 10.21; Jn 5.35; 8.56; Acts 2.26; 16.34; 1 Pet. 1.6, 8; 4.13; Rev.
19.70 and ἀγαλλίασις (Lk. 1.14, 44; Acts 2.46; Heb. 1.9; Jude 24)
are anything but certain pointers. If Luke had wished to indicate that
the meal was a eucharist he would have done this clearly, perhaps
using the phrase, the breaking of bread (though whether this refers to
the eucharist is quite uncertain). Yet there is a sense in which Reicke
is right: every Christian fellowship meal was at least potentially a
eucharist. On the Christian meal in Acts see Introduction, pp. xciif.
The rejoicing of the gaoler and his family is connected more
clearly with his faith than with the meal. πεπιστευκώς is a causal
participle, because he had believed, or perhaps somewhat less
strongly, rejoiced in his new-found faith.
For πανοικεί cf. v. 31; the whole household is involved.
‘πανοικεσία, Άττικῶς. πανοικί, Έλληνικῶς’ (Moeris); cf. Lobeck
(Phrynichus 514). But Plato uses πανοικεί (-ί); it is also common in
papyri (MM 476f.). See also Josephus, Ant. 4.70.
For the last five words of the verse D has ἠγαλλιάσατο σύν τω
οἴκφ αὐτού, πεπιστευκώς ἐπἰ τόν θεόν. See Μ. 1.67f., 235. ἐπί τόν
θεόν is an improvement; the simple dative suggests (though Luke
cannot have meant) that he simply believed what God said.

35. The night meal takes the story on till dawn, when the
800 COMMENTARY ON ACTS

στρατηγοί (vv. 20, 22) are again at their business. It is rightly

pointed out (Beyer 102) that the present verse could follow directly
upon ν. 23 or v. 24; there would then however be no apparent motive
for the magistrates’ change of mind with regard to their prisoners. In
the story as Luke has presented it it is clear that they too have
experienced the earthquake, and connect it with those men. Accord-
ingly they send their officers, τούς ραβδούχους (in the NT only here
and at v. 38). The word denotes one who carries a rod or staff of
office (ράβδος). The Vulgate correctly translates lictores. In Rome
itself the lictors bore the fasces as their badge of office; those
attending for example Consuls and Praetors had, in addition to
ceremonial duties, the task of carrying out arrests and punishments;
those attached to officials in a colony such as Philippi would no
doubt have the same tasks assigned to them—here, unusually, that of
rectifying wrongful arrest and punishment.
After the plural τούς ραβδούχους the singular verb ἀπόλυσον is
surprising; presumably we must think of the order that the lictors
were charged to transmit to the gaoler.
The text of the verse in D (with varying support from a few other
Western authorities) runs: συνῆλθον οἱ στρατηγοί (this suggests,
though it does not require, a special meeting of the bench) ἐπί τό
αυτό (a Lucan expression; see 2.47) εἰς τήν αγοράν (see ν. 19) καί
ἀπομνησθέντες τον σεισμόν τόν γεγονότα (if it was as described
they would not merely remember it but see its effects about them)
ἐφοβήθησαν (not simply the earthquake itself but the supernatural
forces exhibited on behalf of the prisoners) καί ἀπέστειλαν τούς
ραβδούχους λέγοντες· ἀπόλυσον τούς ανθρώπους ἐκείνους οὓς
ἐχθές παρέλαβες (this, another singular verb, is clearly addressed to
the gaoler). This reading looks like an attempt to make the statement
fuller and clearer, leaving nothing to the imagination or intelli-
gence—another characteristic Western ‘glimpse of the obvious’ (cf.
8.1 and the note), επί τό αυτό is the sort of Lucan expression that an
editor could easily borrow.
Release, like imprisonment, was a matter of the magistrates’
coercitio (Tajra 26).

36. The lictors brought their message to the gaoler, and the gaoler
passed it on to Paul and Silas—or rather, as far as the wording goes,
to Paul, who stands out as the leading partner.
ἀπέσταλκαν (for the correct perfect form, –ασι). See Μ. 1.52 and
2.221, where Moulton corrected his earlier view. The tendency to
assimilate was strong, the third person plural ending in –ασι was the
only ending in which the perfect active differed from the weak aorist
active, and –αν had every chance of displacing the ‘correct’ form,
even among reasonably well educated scribes; in this case
ἀπέστειλαν in v. 35 may have affected the copyist’s thought. The
 43. PAUL AND SILAS AT PHILIPPI. 16.11-40. 801

fact is noted without comment (except a reference to unspecified

inscriptions and papyri) in BDR § 83.1, n. 1.
πορεύεσθε ἐν ειρήνη. D gig omit ἐν ειρήνη, which may perhaps
not be original; in any case, whatever be the foundation of the story,
Luke is responsible for the wording. '“Bibelgriechisch” lässt Lukas
sogar den neubekehrten Gefängniswärter in Philippi sprechen’ (BDR
§ 4.3, n. 8).

37. Paul did not find this somewhat casual treatment satisfactory.
For his (or Luke’s) motives see below.
δείραντες: the illegality of beating a Roman citizen is mentioned,
with a reference to 2 Cor. 11.25, on v. 22. The Lex Porcia forbade
this: Porcia tamen lex sola pro tergo civium lata videtur: quod gravi
poena, si quis verberasset necassetve civem romanum, sanxit (Livy
10.9.4). This is eloquently taken up by Cicero, In Verrem 2.5.66
(170), Facinus est vincili [or vincire] civem Romanum, scelus
verberari [or verberare], prope parricidium necari [or necare]. Cicero,
of course, knew that what ought not to happen sometimes did
happen. In the same oration (2.5.62(162)) he writes, Caedabatur
virgis in medio foro Mesanae civis Romanus, iudices, cum interea
nullus gemitus, nulla vox alia istius miseri, inter dolorem crepi-
tumque plagarum audiebatur nisi haec, ‘Civis Romanus sum’. This
incident however confirms that the victim knew that to claim
citizenship ought to have delivered him from his suffering. Cf.
further 2.5.57(147), and see Sherwin-White (58f.).
Paul stresses the enormity of the offence committed by the magis-
trates by adding δημοσία (publicly; cf. δημοσία τεθνάναι, to die at
the hands of the public executioner, Demosthenes 45.81(1126). They
had been publicly insulted and disgraced by the punishment. On the
theme of honour and shame see Rapske (Book of Acts 3.303f.) They
were also άκατακρίτους—uncondemned because there had been no
trial (re incognita). Cf. Rackham (291). This was a further point well
understood in Roman law and custom. Tacitus, Histories 1.6, con-
demns the execution by Galba of Congonius Varro and Petronius
Turpilianus: ‘inauditi atque indefensi tamquam innocentes perierant.’
Augustine (De Civitate Dei 1.19) bases an argument against suicide
on the principle: ‘Vos appello, leges iudicesque Romani. Nempe post
perpetrata facinora nec quemquam scelestum indemnatum inpune
voluistis occidi.’ See also Cicero, In Verrem 2.1.9(25): ‘Causa cognita
multi possunt absolvi, incognita quidem condemnari nemo potest.’ D
(syp) add (before δείραντες) αναιτιους, innocent, guiltless; this is not
appropriate in the same clause as άκατακρίτους.
The most serious point (in the view of the magistrates), however,
is that Paul and Silas are Romans; that is, citizens, enjoying by right a
considerable measure of immunity and having the power to seek
legal redress. For Paul’s claim to be a Roman citizen, and for the
802 COMMENTARY ON ACTS

means by which he acquired the citizenship, see 22.25, 28; in the

present passage the same claim is made for Silas. On the citizenship
as it relates to Paul in Acts see especially H. J. Cadbury in Begs.
5.297-338 and Sherwin-White (55-72, 144-54). It is not easy to
answer the question how he could have demonstrated his citizenship,
especially after his clothes, with no doubt any pockets they may have
contained, had been tom off (v. 22). Whether in any circumstances
he would have had a certificate to show is uncertain (Sherwin-White
148f.); he could hardly from the prison in Philippi ask for the register
in Tarsus to be consulted. It may be that the bare claim would be
enough to frighten the magistrates on the ground that though it might
be false it could possibly be true and that if it were they could be
involved in expensive legal proceedings with an unfortunate out-
come. There is no reference to the citizenship in any extant epistle of
Paul’s, and we know (2 Cor. 11.23, 25) that he received a Roman
flogging three times and was often imprisoned. These facts do not
prove that Paul did not possess the citizenship. We know (1 Cor.
9.12) that he was not in the habit of insisting on all his rights and
there is nothing improbable in Luke’s own suggestion in this chapter
that on occasion action was taken in hot blood and too swiftly to
permit an appeal to Paul’s rights, supposing Paul to have been
willing to make one. This suggests a further question. If it be granted
that Paul was a citizen yet often waived such rights as he possessed,
why does he in Philippi stand on his dignity, not accepting simple
release but demanding from the magistrates personal appearance and
(virtually) apology? The question in this form is not answerable,
since the historicity of the event is open to question and we have in
any case no direct access to Paul’s motives. We may however say
that Luke uses the incident as a warning to magistrates: they would
be well advised to give Christians a fair trial and prove their guilt
before punishing them. According to Barth (CD 3.4.685), Paul was
not motivated by pride or rancour, but by the ‘desire for a restoration
of disrupted order’. This is fair enough, if we accept both the
historicity of the story and that Paul’s actions were governed (as they
doubtless usually were) by theological principles. For relevant
principles see Rom. 13.1-7, with Romans 224-9.
On λάθρα (from λανθάνειν), Radermacher (168f.), notes the
Hellenistic tendency to prefer adverbs to participles.
ἐκβάλλουσιν neatly takes up ἐβαλον (v. 23): ‘put us in ... put us
out’.
ού γάρ: the use of γάρ in the answer to a question is classical
(BDR § 452.2, n. 3), and adds emphasis: No, indeed, αλλά: but, on
the contrary ...
For Jews as Roman citizens see Josephus, Ant. 14.228, 232, 234,
237, 240; also Juster, 2.25-27; Th. Mommsen, ZNW 2 (1901), 82f.;
NS 3.133, 135.
 43. PAUL AND SILAS AT PHILIPPI. 16.11-40. 803

38. The ραβδούχοι report back to the στρατηγοί; for the reaction
of the latter see v. 37, and cf. 22.29. For δέ, E 104 pc have τε. For
the effect of the claim to Roman citizenship see Cicero, In Verrem
2.5.57(147), Illa vox et imploratio, ‘Civis Romanus sum,’ quae saepe
multis in ultimis terris opem inter barbaros et salutem tulit.
The text of D (which has some Syriac support) in this verse shows
some confusion. It runs: ἀπήγγειλαν δέ αυτοιςοι (sic) στρατηγοῖς
οἱ <b>αβδοῦχοι τὰ <b>ήματα ταῦτα τἀ <b>ηθέντα πρὀς τούς στρατη-
γούς, οί δὲ ἀκούσαντες ὅτι 'Ρωμαῖοί εἰσιν εφοβήθησαν. It should
be noted that D differs again and even more widely from the other
MSS in the next verse; see the note there. It seems clear that D has
conflated two texts. τὰ ρήματα and τά <b>ηθέντα are doublets, and the
impossible αυτοισοι bears witness to αὐτοῖς (possibly αυτοί) and
τοῖς (στρατηγοῖς). The original Western text (cf. Ropes in Begs.
3.160; Clark 106f., 365) probably ran, ἀπήγγ. δέ οἱ <b>αβ. τά <b>ηθέντα
πρὀς τούς στρ., οἱ δὲ ἀκ. ὅτι 'Ρωμ. εἰσιν ἐφοβ. This is an innocent
and indeed pointless variation on the Old Uncial text, and suggests
the work of an editor who was at this point copying, but not too
attentively or respectfully. See the note on the next verse, and
Introduction, pp. xxi, lxix.

39. παρεκάλεσαν, a word of varying meaning in Acts (see on

15.32). Here, they asked—perhaps stronger, urged, or, begged. BDR
§ 328.2. n. 3 (cf. v. 15) hint that the more natural tense would be
παρεκάλουν, the imperfect. This tense in fact appears immediately
in ἠρώτων; could it be that παρακαλεῖν here means to placate?
They placated them and then set about their request.
ἐξαγαγόντες, brought them out—of what? In view of what
follows one is inclined to think that they were now brought out of the
prison, and then asked to go away from the city, από τής πόλεως.
But already at v. 30 they were ἔξω, and at v. 34 they were eating in
the gaoler’s house. Possibly we should, in view of ἀπόλυσον (v. 35),
ἐξελθόντες (v. 36), and ἐκβάλλουσιν (v. 37), suppose that Paul and
Silas had been brought back into the prison. See also v. 40. This is
not the general impression given by the story, which says nothing of
a return, but the link was not important, and Luke might well omit it,
or not think of it. But perhaps the best explanation is that the original
form of the story did not contain the account of the earthquake, vv.
25-34.
BDR § 209.1 note that here, exceptionally, ἀπό and ἐκ are not
distinguished; by ἀπελθεῖν από της πόλεως Luke means ‘ “aus der
Stadt hinausgehen” (nicht “aus der Nähe der Stadt Weggehen”)’.
In this verse the reading of D (partially supported by 614 pc syh**)
is so different from that of other MSS as to call for separate
commentary. There is nothing intrinsically impossible or indeed
difficult in it and it is of some importance as an outstanding example
804 COMMENTARY ON ACTS

of a place where the Western text presents a different picture from

that of the Old Uncials—different, however, not in essentials but in
additional detail. Which is to be preferred, and whether it is right to
prefer one as more ‘original’ than the other, are questions that can be
answered only as part of a general discussion of the two texts. See
1.20-29 and Introduction, pp. xix-xxiii; the view taken here is that
the Western text arose at a time and place in which the text was not
regarded as having full canonical authority and was therefore open to
free modification. The text of D is
και παραγενόμενοι μετά φίλων πολλών εις τήν φυλακήν
παρεκάλεσαν αυτούς ἐξελθειν εἰπόντες' Ήγνοήσαμεν τά
καθ’ υμάς δτι ἐστέ ἄνδρες δίκαιοι. και ἐξαγαγόντες
παρεκάλεσαν αυτούς λέγοντες* Έκ τῆς πόλεως ταύτης
ἐξέλθατε μήποτε πάλιν συστραφώσιν ήμεῖν ἐπικράζοντες
καθ’ υμών.
See Ropes in Begs. 3.160f.; Clark 107, 365f.
The στρατηγοί appear at the prison with many friends, who, they
presumably hope, will add weight to their request, Their presence
certainly adds to the dignity of Paul and Silas. It is here clearly stated
(cf. the note above) that the preachers are back in prison.
παρεκάλεσαν αυτούς may be a gloss introduced from the other text,
but here it has an explicit complement, whereas ἠρώτων ἀπελθεῖν is
wanting. εἱπόντες is an aorist participle of coincident action. Direct
speech follows.
To Ήγνοήσαμεν τά καθ’ύμάς ὅτι ἐστέ ἄνδρες δίκαιοι there is no
parallel in the Old Uncial text. Turner (Μ. 3.15) translates, ‘We acted
amiss at your trial in court’, adding in brackets the word πράγμα,
presumably to show that this is how he understands τά καθ’ύμᾶς.
This is fair enough, though since there was no trial in court your case
(in a legal sense) would be better (cf. 25.14). But the rendering we
acted amiss is questionable. It is true that ἀγνοεῖν, used absolutely,
can mean ‘go wrong, make a false step ... to be ignorant of what is
right, act amiss’ (LS 12, s.v. II). But (a) in the NT the word means
almost always not to know, not to understand (Heb. 5.2; 2 Pet. 2.12
are unlikely and in any case unimportant exceptions); (b) if the word
is taken in Turner’s way it is not easy to explain the following clause,
ὅτι κτλ.; (c) d has ignoramus. It is better to translate, We failed to
understand, in reference to your affair, that you were innocent
(δίκαιοι) men. The next words bring the Western text into line with
other authorities, but fresh material follows, again in direct speech,
and the magistrates explain why they are anxious that Paul and Silas
should leave. It is μήποτε πάλιν συστραφώσιν ήμῖν. They are
concerned for their own safety as well as that of the prisoners. Those
who brought the initial complaint may again gather together, with
aggressive intent; cf. the use of the cognate συστροφή at 19.40;
 43. PAUL AND SILAS AT PHILIPPI. 16.11-40. 805

23.12. They would indeed be shouting against Paul and Silas (καθ’
ὑμών), but aggressively urging the magistrates to take action against
them.

40. ἐξελθόντες: unless this is coincident with ἐξαγαγόντες (v. 39)

the chronology is confused once more. ἀπό τής φυλακής: ἐκ (P74 A
D E Ψ 0120 <a>) is a natural improvement, so that there is no need
here to think of a distinct Western tradition of events. εἰσήλθον
(ἧλθον, D e gig) πρὀς τήν Λυδίαν. Cf. ν. 15. The narrator has now
faded from the scene, because he had already left Philippi, because
the first person plural is a literary device, or because interest is
naturally focused on the two witness-bearers who have suffered in
the cause. εἰση. πρὸς τ. A. is short for εἰσελθόντες εις τὀν οἶκον τής
Λ. (ν. 15).
παρακαλεῖν is now used in a sense quite different from that of v.
39. Only the conversion of Lydia has been mentioned so far in the
chapter, so that we have no evidence to tell us who the αδελφοί were.
The phrase is a formula Luke uses from habit; he may think of the
brothers—an έκκλησία—as meeting in Lydia’s house. D extends the
formula by introducing a report: καὶ ὶδόντες τούς αδελφούς
διηγήσαντο ὅσα ἐποίησεν κύριος αὐτοῖς παρακαλέσαντες
αυτούς. Ε Ψ <a> introduce a second (pronominal) object: ἰδόντες
τούς ἀδ. παρεκ. αυτούς.
‘The Duoviri left themselves open to severe punishment for they
could be deprived of office and disqualified from any further
government service for having violated the rights of Roman citizens
in a Roman colony’ (Tajra 29, referring to Dio Cassius 60.24.4).
Luke is not interested in the fate of the Duoviri but only in the next
moves of Paul and Silas.
 44. FROM PHILIPPI TO ATHENS 17.1-15

(1) They followed the road through Amphipolis and Apollonia and came to
Thessalonica, where there was a synagogue of the Jews. (2) Paul, in
accordance with his custom, went in to them on three Sabbaths1 and argued
with them on the basis of the Scriptures, (3) expounding them and submitting
that it was necessary that the Christ should suffer and rise from the dead, and,
‘This man is the Christ, Jesus, whom I am proclaiming to you.' (4) Some of
them believed and accepted their lot2 with Paul and Silas, also a large
company of the3 god-fearing Greeks, and not a few of the leading women.
(5) The Jews became jealous,4 took to themselves evil men from among the
market-place louts, gathered a crowd, and set the city in an uproar. They
sought to bring them out5 to the people. (6) When they did not find them they
dragged Jason and some brothers before the politarchs, shouting, ‘These men
who have6 led the whole world into revolt have come here too. (7) Jason has
taken them in; and they are all acting contrary to the decrees of Caesar,7
saying that there is another king, Jesus.’ (8) They disturbed the crowd and
the politarchs, as they listened to these [charges], (9) and from Jason and the
others they took security8 and dismissed them.
(10) Immediately, in the course of the night,9 the brothers sent off Paul and
Silas to Beroea. When they arrived they went into the synagogue of the Jews.
(11) These [Jews] were more liberal10 than those in Thessalonica; they
received the word with all eagerness, searching the Scriptures daily [to see]
if these things were so. (12) So many of them became believers; also not a
few of the Greek women of good standing, and men. (13) But when the Jews
from Thessalonica knew that the word of God had been preached by Paul in
Beroea also they came there too stirring up and disturbing the crowds. (14)
Immediately, at that very time, the brothers sent Paul away11 to travel to the
sea. But Silas and Timothy remained there [in Beroea]. (15) Those who were
accompanying Paul brought him as far as Athens; they left with instructions
for Silas and Timothy that they should come to him as quickly as possible.

1RSV, for three weeks.

2RSV, NEB, joined.
3RSV, devout; NJB, god-fearing people and Greeks.
4NJB, full of resentment
5NJB, before the People’s Assembly.
6RSV, turned the world upside down; NEB, made trouble all over the world; NJB,
been turning the whole world upside down.
7NEB, flout the Emperor’s laws.
8NEB, they bound over Jason and the others.
9NJB, when it was dark.
10RSV, noble; NEB, civil; NJB, noble-minded.
11RSV, on his way.

806
 44. FROM PHILIPPI TO ATHENS 17.1-15. 807

Bibliography
K. P. Donfried, NTS 31 (1985), 342-6.
E. A. Judge, RTR 30 (1971), 1-7.
D. W. Kemmler, Faith and Human Reason, SuppNovT 40 (1975).
G. D Kilpatrick, JTS 11 (1960), 340.
J. Kremer, in Kremer, Actes 11-20.
J. Murphy-O’Connor, RB 99 (1992), 418-24.
C. Schuler, Classical Philology 55 (1960), 90-100.
Μ. J. Suggs, NovT 4 (1960), 60-8.

Commentary
The most significant stop made by Paul and Silas in Macedonia, after
leaving Philippi, is at Thessalonica. The mission begins with con-
siderable success (vv. 1-4). Opposition on the part of the Jews,
however, leads the politarchs to take relatively mild action against
the Christians (vv. 5-9); Paul and Silas are sent off to Beroea, where
they receive a more favourable welcome (vv. 10-12). Jews from
Thessalonica pursue the missionaries to Beroea. The Beroean Chris-
tians send Paul away to the sea and bring him to Athens; Silas and
Timothy remain in Beroea, but with instructions to rejoin Paul as
soon as possible (vv. 13-15).
All this makes a connected narrative that follows a logical course
and gives rise to no serious difficulty. It is probably to be regarded as
a continuous composition by Luke himself. A number of words are
characteristic of his vocabulary (e.g. διαλέγεσθαι, καταγγέλλειν,
σέβεσθαι, σύρειν, οἰκουμένη, ταράσσειν, ἐκπέμπειν, ἐξιέναι).
There are features of Lucan style, such as the litotes of v. 4 and the
continuation of a sentence by the use of a relative (v. 7). This is of
course not to say that Luke made it all up out of his imagination. That
the place names—Amphipolis, Apollonia, Thessalonica, Beroea—
were derived from an itinerary is a probable hypothesis which is
supported by the fact that no incidents are recorded for Amphipolis
and Apollonia. This may be due to the fact that Paul and his
companions passed through these towns without making any effort to
evangelize them (possibly because they had no synagogues?), but it
is at least equally probable that all the names stood in a list and that
Luke visited Thessalonica and Beroea and acquired information but
was unable to reach the other two towns—or if he reached them
gained no usable information. The name Jason suggests contact with
local tradition; the use of the correct word ‘politarch’ will suggest
either contact or a good knowledge of Roman provincial administra-
tion. Weiser (443) ascribes vv. la, 10a, 14f. to the itinerary; for
different analyses see Roloff (249); Lüdemann (194f.); Taylor
(5.265) (‘le récit de Act I se lit maintenant en 17.1b-10,12a, 13-15’).
808 COMMENTARY ON ACTS

The essence of Paul’s preaching is given in a sentence that recurs

frequently in varying forms in Acts: τόν Χριστόν ἔδει παθεῖν καί
ἀναστήναι ἐκ νεκρών. The (deliberate?) misunderstanding of Paul’s
message to mean that Jesus is another king (and thus a rival to
Caesar) is one that must have occurred frequently. It was easy to
reject, probably not so easy to dispose of.

1. The new paragraph resumes the account of the journey broken

off at 16.12 to describe events at Philippi. It is at least possible that
the stories of Lydia and the gaoler were inserted in an itinerary that
described a journey from Troas through Samothrace, Neapolis,
Philippi, Amphipolis, Apollonia, and Thessalonica; see Introduction
p. xxvi. Schneider (2.222f.) however goes too far in the belief that
17.1 cannot have followed upon 16.40. For the structure of the new
paragraph see above, p. 807.
Διοδεύσαντες: elsewhere in the NT διοδεύειν is used only at Lk.
8.1, where Jesus is said to go preaching and evangelizing in every
city and village. Were it not for this one would be inclined to think
that Luke had here abandoned his usual διέρχεσθαι (which some-
times at least refers to a preaching mission) because Paul and his
companions passed through Amphipolis and Apollonia without
stopping to preach there. This may in any case be true. ND 1.45
provide a recently published example of the verb, used of men on
military service; here it may well (by the compounded οδός) point to
the fact that the journey followed the Via Egnatia, the main land
route from Rome to the East (see Cicero, De Provinciis Consularibus
2(4): via illa nostra, quae per Macedoniam est usque ad Helle-
spontum militaris).
For the somewhat unpredictable use of the article with placenames
in Acts see BDR § 261.2, n. 2. Amphipolis and Apollonia lay on the
Via Egnatia between Philippi and Thessalonica. According to Livy
(45.29; see above on 16.12) Amphipolis was the capital of the first
district of Macedonia (Capita regionum, ubi concilia fierent, primae
regionis Amphipolim, secundae Thessalonicen, tertiae Pellam, quar-
tae Pelagoniam fecit). It was formerly called Εννέα οδοί (Thucy-
dides 1.100.3), but became an Athenian colony (4.102.1). It derived
its name from the fact that it was surrounded on two sides by the
river Strymon (ἐπ ἀμφότερα περιρρέοντος τοῦ Στρυμόνος, δια το
περιέχειν αυτήν), and it was visible (περιφανής) from both sea and
land (4.102.2). Apollonia was further inland, a day’s journey beyond
Amphipolis (id diei iter est, Livy 45.28). Hemer (115) thinks that
Amphipolis and Apollonia were ‘the places where the travellers
spent successive nights, dividing the journey into three stages of
about 30, 27 and 35 miles’. To cover these distances each in a single
day presupposes the use of horses (Begs. 4.202).
In due course they came to Thessalonica, πόλις ... μεγίστη καί
 44. FROM PHILIPPI TO ATHENS. 17.1-15. 809

πολυάνθρωπος (Theodoret, HE 5.17.1; cf. Lucian, Lucius 46). In

Acts ὅπου is used only here and at 20.6 (but here B (Ψ) <a> have ου
and D has ἐν ἧ καί). σὗ is used 9 times if 20.6 is included. According
to Zerwick (§ 217) ὅπου contains a causal element; Paul and his
company halted in Thessalonica because there was a synagogue there
(cf. 16.13); it would be implied that they passed by Amphipolis and
Apollonia because there was no Jewish colony in these cities.
Knowling (357) thinks that ὄπου is not to be distinguished from ου
but if it is it implies oppidum tale in quo esset. Thessalonica was
head of the second division of Macedonia (see above) and a free city
(civitas libera; Pliny, Natural History 4.36)—hence the politarchs
mentioned in v. 6. There is no confirmatory evidence for the
existence here of a Jewish community as early as Acts, but C1J 693
attests a synagogue or at least a Jewish family in the late second
century, or later (Hemer 115). For the significance of the synagogue
and its place in Luke’s understanding of the Christian mission see on
6.9; 13.5; et al. Taylor (269) quotes an inscription possibly relevant
in time (IG 10.2.1.72) which contains the words θεῷ ὑπσίστω κατ’
ἐπιταγήν ουες
For τήν Άπ. ἦλθον εις Θεσσ., D has κατήλθον εἰς Άπολλωνίδα
κἀκεῖθεν εις Θεσσ. Metzger (452) points out that this implies a stop
in Apollonia, which is in any case probable; there is no great
difference in meaning. κατῆλθον is not correct, since Amphipolis is
a seaside, Apollonia an inland town. D, with E <a> also has ή
συναγωγή, in agreement with Apollonius’ canon. But here the
omission of the article is legitimate; the meaning is that in Thessalo-
nica there was one of the (many) synagogues of the Jews.

2. For κατά τὸ εἰωθός with dative of possession cf. Lk. 4.16.

According to BDR § 189.1. n. 1. use of the dative emphasizes the
thing possessed, of the genitive the possessor. For Paul’s habitual use
of the synagogue up to this point cf. 13.5, 14; 14.1; (16.13). For τω
Παύλω, D latt syp have ό Παύλος; with this reading τό εἰωθός will
refer to the general custom of the Jews in resorting to the synagogue.
πρὸς αυτούς is an ad sensum construction: Paul went in to those
who might be expected to be found in the synagogue—Jews, with
perhaps a few interested Gentiles.
ἐπί σάββατα τρία. There is an important note on the use of
σάββατον (Sabbath, week) in Begs. 4.202f. For the use of ἐπί with
reference to time cf. Lk. 4.25 (si v.l.); 18.4; Acts 3.1; 4.5; 11.10;
13.31; 16.18; 18.20; 19.8, 10, 34; 27.20. The meaning varies: time
when, time how long, number of occasions. The meaning here could
be for three weeks but is perhaps more probably on three Sabbaths
(cf. 13.27, 42, 44; 15.21; 18.4). Phil. 4.16 (cf. 4.9) suggests very
strongly that Paul stayed in Thessalonica a good deal longer than
three weeks; so does 1 Thessalonians.
810 COMMENTARY ON ACTS

διελέξατο, argued, debated. G. D. Kilpatrick (JTS 11 (1960), 340)

points out that in Lk. διαλογίζεσθαι is used, in Acts διαλέγεσθαι.
He is right in refusing to see here evidence for different authorship,
perhaps right in noting in Acts a change in style in the direction of
Attic Greek or a more literary Koine. Paul’s argument turned (as in
the synagogue might be expected) on the interpretation of the OT
Scriptures. This is true whether ἀπό (D, ἐκ—an improvement) τῶν
γραφών is connected with διελέξατο or with διανοίγων καί
παρατιθέμενος (in the next verse). It makes little difference: Paul’s
argument as a whole is based on Scripture; it is Scripture that he
‘opens’ (expounds). The more general sense is probably better: he
argued on the basis of Scripture, and the argument consisted of
exposition and affirmation. For διελέξατο, διελέγετο (imperfect, as
in D at 18.19) might have been expected, but the continuous content
of the dispute is brought out by the present participles in v. 3, and
BDR § 327.1, n. 2 are probably right in taking διελέξατο as a
constative aorist.
Schmithals (155) makes the point that Paul will not have been able
to enter every fresh synagogue as an unknown; news of him (as a
trouble-maker) will have spread through the Jewish communities.
News from Philippi may have helped to precipitate the opposition
recorded in v. 5.

3. The new verse continues the sentence of v. 2 with an account of

the manner and content of Paul’s synagogue disputation, διανοίγων:
the verb is used in the same sense at Lk. 24.32, but neither LS 405
nor BA 375 is able to cite a parallel earlier than Aeneas Gazaeus
(Theophrastus 5B (MPG 85.877), ... διανοίγειν ... πειράσομαι τά
των παλαιών απόρρητα), of the 5th-6th century AD. The connec-
tion of the word with the Scriptures suggests a Jewish background; it
is however doubtful whether this can be found in the verb which
is commonly employed in exegesis with the meaning ‘to open a
lecture or sermon with a Bible passage’—so e.g. Berakoth 63b (not
63a as in Bacher), cited by W. Bacher, Die exegetische Terminologie
der jüdischen Traditionsliteratur I (1965 (1899)), 162f. This is
presumably the meaning also of Shekalim 5.1 as this is given in the
Babylonian Talmud. Mordecai was given the name Pethahiah
he began his discourses with words (of
Scripture) and proceeded to expound them. In the Palestinian Talmud
however the preposition is wanting and we have simply
and the meaning may be, ‘He opened the words (of Scripture) and
expounded them.’ Danby translates (a ‘fusion of the two types’ of
Mishnah), ‘ ... because he was able to “open” matters and to
expound them’. The additional statement that Mordecai knew sev-
enty languages suggests an interest in verbal explanation. There can
be little doubt of Luke’s general meaning—a biblically based
 44. FROM PHILIPPI TO ATHENS. 17.1-15. 811

argument—but it would be hazardous to assert that his language is

based on a Jewish model. In the Qumran literature CD 5.3 is scarcely
relevant since this refers to the opening of a sealed book; 1QH 18.20
is difficult but might be relevant, to an uncircum-
cised ear a word was opened.
παρατίθεσθαι is not used elsewhere in the NT with an object
clause (but cf. 28.23 A), and this use of the word appears to be late.
Cf. however Plato, Politicus 275b, τόν μῦθον παρεθέμεθα, ἵν’
ἐνδείξαιτο ... Dodd, AS 18, compares this ‘opening up’ and
‘applying’ with Romans 9-11.
No attempt is made here to show which passages in the OT are held
to prove that the Messiah must suffer and die (suffering must include
death in view of the following words; cf. 1.3; 3.18; Lk. 24.26,46) and
rise from the dead. On this fundamental theme of Christian preaching
as represented in Acts see Introduction, pp. lxxxv-lxxxvii. The basic
proposition, that Scripture foretells the death and resurrection of the
Messiah, is followed by the identification of the dying and rising
Messiah with Jesus. The second ὅτι, and perhaps the first, introduces
direct speech, as καταγγέλλω shows. ‘This man is the Messiah (who
thus suffers and rises), namely Jesus, whom I am now proclaiming.’
For a similar change from indirect speech to direct cf. 1.4. On the
assumption that the text of NA26 is correct one would say that it was
the unusual appositional position of ό 'Ιησούς that gave rise to several
variant readings: ὁ Χριστός ὁ Τησοῦς, Β; ό Χριστός ’Ιησοῦς, ΨΜ;
Χριστός ’Ιησοῦς, P74 A D 33 81 pc gig vgst; Τησοῦς Χριστός 614
2495 pc vgcl; Χριστός, E 36 453 pc. But there is something to be said
for the reading of D or of .

4. For πείθεσθαι as a positive response to the Gospel cf. 28.24

(Lk. 16.31; Gal. 5.7); elsewhere in Acts it denotes the acceptance of
some other teaching or opinion (5.36, 37, 39; 21.14; 23.21; 26.26;
27.11). Some of the Thessalonians (Jews or proselytes, since they
were to be found in the synagogue, and others who might have been
there are mentioned separately) accepted the propositions that the
Messiah must suffer and die, and that Jesus was the Messiah. It is
further said that these believers προσεκληρώθησαν τω Π. καὶ τῷ Σ.
The verb is hapax legomenon in the NT. In the passive its normal
meaning would be to be assigned (possibly by lot); so for example
Josephus, War 2.567, of various districts assigned to a military ruler.
Philo, Legatio ad Gaium 68 suggests a more suitable meaning with
τῶν μὲν τούτῳ τών δέ ἐκείνω προσκληρουμένων, though here the
participle could be regarded as middle, whereas Luke’s form must be
passive. One is tempted to translate freely they threw in their lot with
P. and S.; but Luke held a predestinarian view of conversion (13.14;
16.14) and may have thought that those who attached themselves to
Paul and Silas did so because God had allocated them to this end.
812 COMMENTARY ON ACTS

Blass (186) thinks that προσεκληρώθησαν may have been a corrup-

tion of προσεκολλήθησαν (7.34) or of προσεκλίθησαν (5.36). The
word is omitted (with καί by P74, almost certainly by homoeoteleuton
(-θησαν ... –θησαν).
The believing Jews and proselytes were joined by a large crowd of
pious (σεβομένων) Greeks. For the participle σεβόμενος see 13.43,
50; 16.14; 18.7. Luke would hardly use it of those who were devout
in terms of pagan religion, so that it must refer in some sense to the
religion of Judaism. On the question of the precise relation of these
persons to the synagogue see I.499—501; also now Trebilco (145-66)
and an important article by J. Murphy-O’Connor in RB 99 (1992),
418-24. The present verse distinguishes them from proselytes and
indicates that notwithstanding their piety they count as Έλληνες, not
as Jews.
Luke distinguishes a further group as leading (literally, first)
women. The word πρώτος occurs in this sense several times in the
later part of Acts: 13.50; 25.2; 28.7.17. Cf. Josephus, Ant. 11.141,
τούς πρώτους τού λαού. ND 1.72 notes an inscription from Assos
(IAssos 16.3-4, in the time of Augustus, before 2 BC) which refers to
Lollia Antiochis as πρώτη γυναικών, which appears to be an
honorific title. Horsley asks, 'Is the phrase used of women at
Thessalonike in Acts 17.4 also a title, or merely a descriptive way of
referring to leading women?’ An alternative translation is, ‘Wives of
the leading men’; this meaning is required by the Western text, καὶ
γυναίκες τών πρώτων (D lat). For Luke’s interest in the conversion
of women and their share in the church see 1.14; 5.1; 8.3, 12; 9.2;
13.50; 16.1, 13, 14; 17.1, 2, 34; 18.2; 21.5; 22.4; 24.24. For women
in Macedonia see W. W. Tam, Hellenistic Civilisation (1930), 89,
90. The use of τε and the litotes (οὐκ ὀλίγαι) are also characteristic
of Luke.
The text of the verse is in some confusion. Clark (108, 366) on the
basis of D, reconstructs it as follows:

καί τινες ἐξ αυτών ἐπείσθησαν τή διδαχή,

καί προσεκληρώθησαν
τῷ Παύλω καί τω Σιλᾷ
πολλοί τών σεβομένων
καί 'Ελλήνων πλήθος πολύ
καί γυναίκες τών πρώτων οὐκ ολίγαι.

5. ζηλώσαντες: as at 7.9 (the patriarchs envied Joseph); cf. 5.17;

13.45. (ἐπλήσθησαν ζήλου). The Jews feared that they were losing
control of the synagogue and their appeal to religious non-Jews,
and objected to the success of the Christian preachers. Accordingly
they took to themselves (προσλαβόμενοι, middle) certain evil
(πονηρούς) men τών αγοραίων. There is no serious doubt about the
 44. FROM PHILIPPI TO ATHENS. 17.1-15. 813

meaning Luke here ascribes to this word (but cf. 19.38), but there are
linguistic problems behind it which focus on its accentuation. Thus
Μ. 2.57: ‘ἀγοραιος and αγοραίος are differently distinguished by
grammarians. Zonaras [12th century] has αγοραίοι for οἱ ἐν ἀγορᾷ
ἀναστρεφόμενοι άνθρωποι, and ἀγοραιος as ἡ ἡμερα ἐν ᾖ ή αγορά
τελείται. If that is correct, we must write αγοραίοι in Acts 19.38, and
make αγοραίοι the nominative of the noun in 17.5; but Ammonius
[1st or 2nd century AD according to LS], who is eight centuries older
[ten or eleven centuries on LS’s reckoning] than Zonaras, gives an
entirely different distinction,’ which is not quite correctly given by
LS; see Ammonius 11. The distinction given by Ammonius (4(6)) is
ἀγόραιος μεν γάρ ἐστιν ἡ ἡμέρα, αγοραίος δἑ ό Έρμης ό ἐπί τῆς
αγοράς (Winer-Moulton 61; Winer-Schmiedel 69). LS 13 however
write, ‘The disto, ἀγόραίος vulgar, ἀγοραίος public speaker, drawn
by Ammon., etc., is prob. fictitious.’ Fortunately there is no doubt
that we may here follow the meaning given by Zonaras to αγοραίος.
The word was sometimes used in an innocent sense—traders,
but it tended to refer to the lower kind of frequenter of market
places; cf. Aristophanes, Frogs 1015 (1047) (in company with
διαδρασιπολιται, κόβαλοι, and πανούργοι); Plato, Protagoras
347c (τών φαύλων καί αγοραίων ανθρώπων); Theophrastus,
Characters 6(7).2; Herodotus 2.141.4; Xenophon, Hellenica 6.2.23.
Luke means, and would be understood to mean, louts of the worst
kind, ready to make any amount of trouble for a consideration. BA
22f. refer to Plutarch, Aemilius Paulus 38.4 (275) ( ... ανθρώπους
άγεννείς καί δεδουλευκότας, ἀγοραίους δέ καί δυναμένους όχλον
συναγαγεῖν ... ), with the suggestion that the αγοραίοι might be
professional agitators, but the meaning is not supported and is not
needed here. The subject of ὀχλοποιήσαντες ἐθορύβουν must
grammatically be οἱ Ιουδαῖοι, though one suspects that the
αγοραίοι may have been intended. They would without difficulty
gather a crowd and set the city in an uproar (θορυβεῖν; cf. Mt. 9.23;
Mk 5.39; used in a somewhat different sense at Acts 20.10) for which
it would be easy to blame the Christians.
They gathered at, perhaps set upon, the house of Jason. Of this
man we know only what the present passage tells us. The name
occurs at Rom. 16.21 as that of a helper of Paul's who is also a
συγγενής, that is, a fellow Jew; the name Jason might be a Greek
substitute for Joshua (see Josephus, Ant. 12.239, ὁ μὲν οὖν Ιησούς
Ίάσονα αὑτόν μετωνόμασεν). It is not unreasonable to identify the
two, but the identification is quite uncertain. It would imply that
Jason had at some time travelled to join Paul at Cenchreae (Rom.
16.1). One can say no more than that this is not impossible. Begs.
4.205 points to v. 6, Ίάσονα καί τούς αδελφούς. Does the wording
imply that Jason was not a brother, a Christian? Not necessarily.
The intention of the trouble-makers is to bring them (presumably
814 COMMENTARY ON ACTS

Paul and Silas) out to the mob. Ordinarily δήμος would deserve a
better translation, but Luke appears to have in mind not an orderly
assembly but a riotous gathering bent on violence, though not (if we
may judge by v. 6) on lynching. This however is disputed. Hanson
(174) distinguishes the δήμος (parliament) from the ὄχλος (v. 8);
Conzelmann (95) equates them.
In this verse also the text is in some disorder. The text of NA26 is
probably correct: Ζηλώσαντες δέ oἱ ’Ιουδαῖοι καὶ προσλαβόμενοι
τών αγοραίων ἄνδρας τινἀς πονηρούς καί οχλοποιήσαντες
ἐθορύβουν. This is the reading of P74 A B 33 81 945 1175 1739 al vg
(sy) co. Ψ share this text, reversing the order of ἀνδρ. τ. 0120 614
1241 2495 add after Ίουδ., οἱ ἀπειθούντες; E adds after πον.,
ἀπειθήσαντες. These MSS have probably been affected by D which
reads οἱ δέ ἀπειθούντες Ίουδ. συστρέψαντες τιν. ἀνθ. τῶν ἀγ. πον.
ἐθορυβούσαν; this has affected which has προσλ. δἐ οι Ίουδ. οι
ἀπειθ. τῶν ἀγ, τιν. ἄνδ. πον. κ. ὀχ. ἐθορ.

6. Paul and Silas were not to be found, whether because they were
not in Jason’s house or because they were too well hidden. Other
victims, however, would serve equally well to satisfy the desire of
the πονηροί for violence, and almost equally well as a warning
object-lesson for Christians. Ίάσονα καί τινας ἀδελφούς is very
probably to be taken to mean, Jason and certain other Christian
brothers; that is, Jason himself is a Christian. See however the
reference to Begs. 4.205 on v. 5. Probably Jason’s name was
remembered because he was a member of the Christian group; so e.g.
Weiser (449). Luke evidently thinks it possible that three Sabbaths
might suffice to gather men into the new brotherhood; having been
nurtured in the synagogue they would probably need little instruction
beyond the simple identification of the Messiah with Jesus.
The rioters set about dragging (ἔσυρον, imperfect; the word at 8.3;
14.19) Jason and his fellow Christians ἐπἰ τούς πολιτάρχας. Cf.
Lucian, Lexiphanes 10, Δεινίαν σύρουσιν ἄγδην ἐπί τήν ἀρχήν,
ἔγκλημα ἐπάγοντες ... πολιτάρχης ‘is mainly if not exclusively a
Macedonian title for the non-Roman magistrates of a city. It is found
in inscriptions ranging from the second century B.C. to the third
century A.D.. . . .It would appear that the Macedonian cities had
several politarchs, the number varying with their importance.
Amphipolis had five, Pella only two, Thessalonica had five in the
time of Augustus, but afterwards six’ (Begs. 4.205). See Sherwin-
White (96) with his reference to C. Schuler, Classical Philology 55
(1960), 90-100; also Tajra (34) and Hemer (115). There is further
bibliography in Bruce (2.324) and an important note with much fresh
evidence in ND 2.34f. There is no doubt that Luke has here used the
correct term; naturally, this does not vindicate the accuracy of his
story as a whole, and we shall see that there are obscurities in it, but it
 44. FROM PHILIPPI TO ATHENS. 17.1-15. 815

suggests that someone involved in some way in the report was not
unfamiliar with Macedonia.
βοῶντες, a Lucan word; 7 times in Luke-Acts, 6 times in the rest
of the NT.
ἀναστατοϋν: the word is used in a different sense at Gal. 5.12, in a
fundamentally similar sense at Acts 21.38—the Egyptian rebel led
an armed terrorist revolt against Rome. It is of something similar that
the Christians are here accused, as the next lines make clear. They are
acting (it is alleged) contrary to imperial decrees and attempting to
set up a rival emperor (v. 7). The charge is much more dangerous
than ‘They have upset everyone’ or even than ‘they have turned the
world upside down’. Cf. 24.5. Conzelmann (95) quotes from the Acta
Isidori (Musurillo 23): ἐνκ[αλῶ αυτοῖς] [ὅτι κ]αί ὅλην τήν
οικουμένην [θέλουσι] [ταράσ]σειν. This charge is brought against
the Jews.
οικουμένη is a Lucan word: in Luke-Acts 8 times; in the rest of
the NT 7 times (of which 3 are in Revelation). Luke uses it in the
accepted sense: the whole civilized world.
πάρεισιν (cf. the cognate noun, παρουσία) is a present tense used
(see BDR § 322.2, n. 2) for the perfect of a different verb—better,
perhaps, for a perfect and present combined: they have come, and
here they are.
After βοώντες, D gig w unnecessarily add καί λέγοντες. D* Ψ gig
w change the construction slightly by adding εἰσιν after οὖτοι.

7. Luke characteristically continues the sentence with a relative:

οὓς ὑποδέδεκται. ὑποδέχεσθαι (deponent) is used from the time of
Homer for ‘to welcome into one’s house’; examples in BA 1682.
Approval of the guests and of their mission is implied; Jason if not
fully a Christian is at least a sympathizer.
οὖτοι πάντες: Jason; the brothers (v. 6); Paul and Silas; any other
Christian converts. The phrase is vague, but precision is hardly to be
looked for in the circumstances. Similarly it is difficult to give a
precise meaning to the δόγματα Καίσαρος. In any case, Thessalo-
nica was a civitas libera, and the decrees (Vg has decreta; contrast
Lk. 2.1, where δόγμα is edictum) of Caesar were thus not binding on
the magistrates; Sherwin-White (96) observes that for this reason
‘the city magistrates were not compelled to take serious action’.
Ehrhardt (Acts 96) asks, ‘What decrees were they? The most likely
answer is that they were the very ones by which the Jews had been
banished from Rome because of their rioting impulsore Chresto ...
The mob thus regarded the differences between St Paul and the
synagogal Jews at Thessalonica as an internal quarrel of the Jews,
and was determined to side with that party which was loyal to the
Emperor.’ It may be possible to do better than this. E. A. Judge saw
here ‘reference to edicts against predictions, especially of the death
816 COMMENTARY ON ACTS

or change of rulers, first promulgated by the aged Augustus in AD 11

(Dio 56.25.5-6) and enforced through the local administration of
oaths of loyalty’ (Hemer 167, summarizing Judge in RTR 30 (1971),
1-7). K. P. Donfried (NTS 31 (1985), 342-6) asks why the politarchs
are appealed to rather than the proconsul; his answer is that it was the
politarchs who were responsible for administering the oath of loyalty
to the Emperor.
λέγοντες is probably to be taken as explanatory of πράσσουσιν:
they act against the decrees of Caesar by proclaiming a rival emperor
(βασιλεύς). There may not have been a decree specifically to this
effect, but it was action that would hardly be encouraged by the
reigning Emperor. The charge was one that could readily be used
against Christians; the term βασιλεία τού θεού runs deep into the
gospel tradition and must have found its way from time to time into
Christian preaching, especially in the synagogue. The preachers
could hardly deny that they were proclaiming Jesus as βασιλεύς; Lk.
23.2 shows how dangerous this could be and Jn 18.36 may reflect
explanations that Christians found it necessary to give.
Schille (351) finds in vv. 6, 7 two different accusations (causing
disturbance and proclaiming a different emperor), and concludes,
‘Die Dublette beweist vollends die Bearbeitung eines älteren Textes.’
But it would be quite reasonable to allege that the Christians were
setting the world in turmoil precisely by making claims for a rival
Emperor, and thus inciting civil war.

8. ἐτάραξαν τόν όχλον picks up ὀχλοποιήσαντες rather than

δήμος in v. 5, especially if that verse describes the intention to bring
Paul before the δήμος as a duly constituted court; on this see
however the note on v. 5.
The Western text (D gig syp) reverses the order of όχλον and
πολιτάρχας (for this word see on v. 6), putting the authorities first.
The common text makes better sense: the crowd heard the allegation
about the proclamation of another Emperor first, and thus it came to
the ears of the magistrates. The better sense, however, is not
necessarily the better text.

9. λαβόντες τὸ ἱκανόν is a Latinism, cum satis accepissent, see Μ.

1.20; BDR § 5.4; Moule, IB 192. The Greeks adopted the legal usage
from Rome; ‘What is happening to Jason is clear enough: he is
giving security for the good behaviour of his guests, and hence
hastens to dispatch Paul and Silas out of the way to Beroea, where
the jurisdiction of the magistrates of Thessalonica was not valid’
(Sherwin-White 95f.). This comment does not quite do justice to καἰ
τών λοιπών. Not only Jason but also the brothers (v. 6) were thus
cautioned, and no doubt they were required to give security for their
own behaviour as well as that of Paul and Silas before being
 44. FROM PHILIPPI TO ATHENS. 17.1-15. 817

dismissed (the magistrates απέλυσαν αυτούς)—indeed it had not

been proved that Paul and Silas were guests with Jason (v. 6).
The verse begins οἱ μἐν ουν πολιτάρχαι in syhmg. As Clark (xxxvi)
points out, in the text of all other authorities, if it is strictly read, it
was the rioters who took security and dismissed the persons
charged.
10. For διἀ νυκτός see on 16.9.
οἱ αδελφοί are here not the ‘certain brothers’ of v. 6 but the newly
founded Christian community as a whole, who acted in defence and
protection of the missionaries; 1 Thessalonians (e.g. 1.4 for ἀδελφοί)
is sufficient proof that after a short ministry Paul left a church in
Thessalonica; so also is 20.4. They sent Paul and Silas to Beroea, a
good place for flight, oppidum devium (Cicero, In Pisonem 36(89)) in
that it lay off the Via Egnatia which Paul had followed as far as
Thessalonica. It lay 45 miles WSW of Thessalonica in the district of
Bottiaea (not, as sometimes is said, of Emathia—C. Μ. Danoff, in
KP 1.869), the first town to give itself up to the Romans after the
battle of Pydna. It was μεγάλη καί πολυάνθρωπος (Lucian, Lucius
34—cf. the reference to Lucius 46 on v. 1). The existence of Jews
there is confirmed by inscriptions (NS 3.67, 68). It is possible that
there was only one synagogue (τήν συναγωγήν), but the definite
article can hardly be pressed so far.
ἄπειμι occurs here only in the NT (ἔξειμι only at 13.42; 17.15;
20.7; 27.43; εἶμι not at all, unless, improbably, at Jn 7.34, 36).
ἀπήεσαν is to be taken with εἰς τήν συναγωγήν: Having arrived
(παραγενόμενοι; in the town), they went into the synagogue. The
verb does not imply that they went out of the town to a synagogue
situated outside the walls (Zerwick § 133).
11. The decision to go at once to the synagogue in Beroea,
notwithstanding indifferent success at Thessalonica, was rewarded.
In Thessalonica, though some Jews had believed, ‘the Jews’ were
opposed to Paul; in Beroea ‘the Jews’ were more favourable. They
were εὐγενέστεροι (D p*, εὐγενεῖς). εὐγενής (Lk. 19.12; 1 Cor.
1.26) refers originally to noble birth, but it came naturally to be
applied to noble behaviour (cf. Josephus, Ant. 12.255, οι δέ
δοκιμώτατοι καί τάς ψυχἀς εύγενεῖς οὐκ ἐφρόντισαν αύτού—the
threatening Antiochus Epiphanes; Chrysostom, in Cramer, Catena
282, τουτέστιν επιεικέστεροι). See also Philo, Moses 1.18. the infant
Moses was εὐγενῆ καί ἀστεῖον. Luke means that the Beroean Jews
allowed no prejudice to prevent them from giving Paul a fair
hearing.
Moule (IB 124) thinks this one of the passages in Acts (see also
10.47) where οἵτινες may be used not as a simple substitute for oἵ
(contrast v. 10). It should perhaps be rendered, ‘ ... seeing that they
received the word ...’. For receiving the word see 2.41. μετά πάσης
818 COMMENTARY ON ACTS

προθυμίας, with all eagerness; not an uncommon expression. See

e.g. Herodotus 7.6.2, πάσαν προθυμίην παρεχόμενοι; Philo, Abra-
ham 246, μετά προθυμίας πάσης.
καθ’ ἡμέραν is a common expression in Acts (2.46, 47; 3.2; 16.5;
19.9; cf. 17.17. In none of these passages is τό prefixed, even as a
variant; this is in favour of its inclusion here (it is read by B H L P 6
1175 pm, omitted by P45 P75 D Ε Ψ 0120 33 81 323 614 945 1739
249k5 pm). There is no evident reason why if it was original a
copyist should omit it.
Paul’s fundamental assertion in Beroea, as in Thessalonica, was
understood by Luke to have been (a) that the OT Scriptures affirm
the coming of a Messiah who will suffer and rise from the dead, and
(b) that Jesus was this Messiah. This was a novel interpretation of the
OT and the Beroean Jews wished, before committing themselves to
the new faith, to see if it was true. ἀνακρίνειν is nowhere else in the
NT used of the study of Scripture; it suggests rather the legal
examination of witnesses (or of an accused person)—see Acts 4.9;
12.19; 24.8; 28.18—and this is in fact the sense in which it is used
here. Paul has set up the Scriptures as witnesses: does their testi-
mony, when tested, prove his case?
εἰ ἔχοι (D* E (0120) 36 453 2495 al have έχει) ταύτα ούτως is
virtually an indirect question (Μ. 3.127). Cf. Josephus, Ant. 12.197,
εἰ δύναιτο πρὀς τόν βασιλέα βαδίσαι... ἀνέκρινεν. Cf. the direct
question of 7.1, εἰ ταῦτα ούτως έχει. After ούτως, 614 pc gig syh**
add καθώς Παύλος απαγγέλλει.

12. For the very frequent use of μἐν σὖν in Acts see on 1.6. It is
less common in the middle of a paragraph than at the beginning or
the end; it does however sometimes, as here, introduce the result of
preaching: 2.41; (8.25); 9.31; 16.5. The result of Paul’s work in the
synagogue, and of the open-minded study by the Beroeans, was that
many believed (ἐπίστευσαν; cf. v. 4, έπείσθησαν)—many ἐξ αυτών
(the use of a preposition where earlier Greek would have been
content with a partitive genitive was growing in the Hellenistic
period), that is of the Beroean Jews; also not a few (for the
characteristic litotes see v. 4) of the Greek (that is, non-Jewish)
women of good standing (εὐσχημόνων; for the word see 13.50), and
men. For Luke’s frequent references to women converts see v. 4.
The text of this verse appears in a somewhat different form in D
and some of its allies. Combining the readings of D*, D, and 614 we
obtain as possibly a primitive form of the Western text: τινὲς μὲν συν
αυτών [without ἐξ] ἐπίστευσαν, τινὲς δὲ ἠπίστησαν, καἰ τών
Ελλήνων καί τών εὐσχημόνων ἄνδρες καί γυναίκες ικανοί
ἐπίστευσαν. The effect of this is to reduce the response of the
Beroean Jews, to distinguish between Greeks and εὐσχημόνες, and
slightly to take the emphasis off the women converts—D can
 44. FROM PHILIPPI TO ATHENS. 17.1-15. 819

scarcely be labelled anti-feminist here (see the reference in Epp, 75,

n. 3; also Ramsay (Church 161f.), who sees here a mark of ‘Catholic’
editing), but the women take second rather than a special place. The
reading of d is curious: ... et Grecorum et non placentium et viri et
mulieris [sic—a slip for mulieres] pleres crediderunt.
13. The result of the preaching in Beroea was naturally unpleasing
in Thessalonica, and the Jews of that city (ἀπό as at Jn 21.2—place
of origin, and in this case of normal residence also) made the journey
to Beroea to stir up trouble there.
ὁ λόγος τοῦ θεοῦ is one of Luke’s commonest terms for the
Christian message as preached by the apostles and others.
The object of the Jews was to stir up trouble among the crowds; if
Paul and his colleagues could be represented as the source of unrest
and as likely to cause riots they were sure to be turned out even if the
magistrates could find no fault with what they actually said.
σαλεύειν is a Lucan word (Lk., 4 times; Acts, 4 times; rest of the
NT, 7 times). the word is more often used of literal, physical
disturbance (but cf. 2.25 = Ps. 16.8); it was probably felt to double
with καὶ ταράσσοντες, which was accordingly omitted by P45 E 0120
<a>. In the Old Uncial text the participles σαλεύοντες, ταράσσοντες,
are adverbial to ᾖλθον; they are differently constructed by D (cf. syp),
which adds οὐ διελίμπανον: they came, and did not cease ...
14. The main cause of the trouble in Beroea was no doubt Paul,
and it was therefore important to get him away quickly. τότε is
intended to underline εὐθέως: immediately, at that very time. The
Western text (see below) avoids the conjunction of the two adverbs.
οι αδελφοί, the Christian brotherhood (cf. v. 10), sent Paul off
(ἐξαπέστειλαν in this context means facilitating his departure rather
than dismissing him) to go (πορεύεσθαι) in the direction of the sea.
This is expressed in the majority of MSS (Ψ 0120 <a> syh) by the
phrase ὡς ἐπί τήν θάλασσαν. This use of ὡς ἐπί is a standard
Hellenistic construction meaning towards, sometimes against. See
Μ. 3.321 and BDR § 453.4, n. 7, but especially Field (79): ‘Π. ὡς ἐπί
“to go in the direction of ” a place, whether the person arrives there
or not, is an excellent Greek idiom.’ As Field observes, Wettstein
gives examples, and Field adds Pausanias 2.11.2, καταβαίνουσι δέ
ὡς ἐπί τὀ πεδίον; 3.20.3, ἰούσιν εὐθεῖαν ὡς ἐπί θάλασσαν; and
other passages, ὡς is omitted by the Western text (D 049 pc gig syp),
and in P74 A B E 33 81 323 945 1175 1739 al lat is replaced by
έως. The short text (ἐπί alone) is quite intelligible: Paul is to go to the
sea. But there is no good reason why ὡς ἐπί, if it had stood originally
in the text, should have been disturbed, unless, as Field continues, the
‘excellent Greek idiom’ may not have been familiar to those scribes
who changed ὡς into έως. Bruce (1.330) thinks that Paul’s com-
panions acted ‘as if to conduct Paul to the sea’; that is, they were
820 COMMENTARY ON ACTS

trying to put possible pursuers off the trail. έως may be an

orthographical variant (or error) for ὡς, but it may be original and
mean (in English as awkward as the Greek would be) that Paul was
accompanied as far as to the sea; that is, his companions saw him on
board (cf. 20.38). He would reach the sea probably at Pydna, and sea
travel would take him into a different jurisdiction (Hemer 116). The
reference to the sea must almost certainly mean (unless we follow
Bruce) that Paul, to avoid further trouble in northern Greece, went
from Beroea to Athens by ship, avoiding the land journey through
Thessaly (though for θάλασσαν Markland conjectures Θεσσαλίαν;
see also the textual note on v. 15). Clark (366f.) argues that if Paul
went by sea one would expect the port, Methone or Pydna, to be
mentioned; and οἱ δέ καθιστῶντες... ἤγαγον (ν. 15) suggests a land
journey. ‘There was a Roman road with stations and organized
services from Pydna to Athens by way of Dium, Larisa, Demetrias,
Opus, Chalcis, Thebes, and Oropus, the total distance being 222
miles.’
The Western text of the opening part of the verse runs: τόν μὲν
οὖν Παύλον οἱ αδελφοί ἐξαπέστείλαν ἀπελθεῖν ἐπί... It is hard to
see any motive for the change, apart from a possible distaste for
ευθέως τότε; either we recognize two early recensions of the text, or
have an example of the free copying of a text that lacks full canonical
authority. See Introduction, pp. xxi, lxviii.
When Paul left Beroea for Athens, Silas and Timothy stayed
behind (this is how Luke understands the verb ὑπομένειν; with him it
does not mean to endure). So far in chs. 16 and 17 we have heard
only of the movements of Paul and Silas; Timothy has not been
mentioned since 16.1-3, and 1 Thess. 3.2 might suggest that
Timothy was not known to the Thessalonians till Paul sent him from
Athens. This however is not the necessary meaning of the verse, and,
as Haenchen (490) observes. Luke tends not to mention unimportant
secondary figures. A prima facie reading of Acts 16 and 17 suggests
that he had accompanied Paul, perhaps as a subordinate member of
the party, from Lystra to Beroea. 1 Thess. 3.2 should perhaps be
understood as an assurance that this hitherto slightly regarded
assistant was in fact a trusted lieutenant.

15. καθιστάνειν is a late form of καθιστάναι; on the tendency to

replace –μι verbs with simpler forms in –ω see Μ. 1.33, 38, 55f. In
meaning the word duplicates ἥγαγον; cf. Thucydides 4.78.5, οἱ δέ
Περαιβοί αυτόν ... κατέστησαν ἐς Δῖον. The brothers who were
seeing Paul on his way brought him as far as Athens (see on v. 16), a
considerable journey, whether by land or sea. They returned with
instructions (εντολήν) that Silas and Timothy, left behind in Beroea
(v. 14), were to join Paul as quickly as possible—ώς τάχιστα, a
literary expression; see BDR § 60.2, n. 3; 244.1, n. 2; § 453.4, n. 7.
 44. FROM PHILIPPI TO ATHENS. 17.1-15. 821

According to Acts (18.5) they caught up with him at Corinth. 1

Thess. 3.2 suggests that Silas and Timothy, or at least Timothy, had
joined Paul in Athens. ‘Das Befehl zum Nachkommen gehört zu den
typisch redaktionellen Massnahmen’ (Schille 352). It is not clear
why this should be so. Cf. 1 Thess. 2.14.
With ἐξήεσαν cf. ἀπήεσαν, v. 10.
In the first half of the verse the Western text introduces fresh
features: παρήλθεν δὲ τήν Θεσσαλίαν ἐκωλύθη γάρ εις αυτούς (the
inhabitants of Thessaly) κηρύξαι τόν λόγον (D). It is hard to think
that these interesting observations stood in the original text and were
omitted by an editor. Their addition is easier to understand. A reviser
who missed the hint in v. 14 that Paul travelled by sea and noted the
long stretch of Thessaly between Beroea and Attica (Athens) might
well have asked himself why Paul left the area unevangelized. The
example of 16.6f. suggested an answer which it was easy to frame in
Lucan language. The addition after ἐντολήν of παρά Παύλου (D) or
ἀπ’ αυτού (Ε vg syp sa) was a natural and no doubt correct, though
scarcely necessary, piece of interpretation. It is however by no means
clear why ὅπως ἐν τάχει should be substituted for ἵνα ὡς τάχιστα. It
is probable that we see again the free rewriting of a text not yet fully
canonical. It seems worthwhile to give the text of D for vv. 14, 15 in
full.
τὸν μὲν ουν Παῦλον οἱ αδελφοί ἐξαπέστειλαν ἀπελθειν ἐπί
τήν θάλασσαν ὑπέμεινεν δέ ὁ Σείλας καί ό Τιμόθεος ἐκεῖ. οἱ
δέ καταστάνοντες τόν Παύλον ἤγαγον έως ’Αθηνών,
παρῆλθεν δέ τήν Θεσσαλίαν, ἐκωλύθη γάρ εἰς αυτούς
κηρύξαι τόν λόγον, λαβόντες δὲ ἐντολήν παρά Παύλου πρὀς
τόν Σείλαν καί Τιμόθεον ὃπως ἐν τάχει ἔλθωσιν πρὸς αυτόν
έξήεσαν.
On παρ. δέ τήν Θεσσ. Weiser (452) comments, ‘Die einfügung lässt
erkennen, dass man den Weg ans Meer als Täuschung verstand ...’.
This understanding would agree with the opinion of Bruce given
above.
 45. PAUL AT ATHENS 17.16-34

(16) While Paul in Athens was waiting for Silas and Timothy1 his spirit was
vexed within him as he saw that the city was overgrown with idols. (17) So
in the synagogue he disputed with the Jews and the devout persons and in the
Agora every day with those he chanced to meet. (18) Some of the Stoic and
Epicurean philosophers argued with him, and some said, ‘What does this
third-rate journalist want to tell us?’ Others said, ‘He seems to be a preacher
of foreign gods.’ For he was preaching the message of Jesus and the
resurrection. (19) They got hold of him and brought him to the Areopagus,
saying, ‘May we know what this new teaching, spoken by you, is? (20) For
you are bringing strange things to our ears. We should like to know therefore
what these things mean.’ (21) For all the Athenians and the resident
foreigners had leisure for nothing but to say or hear some novelty.
(22) Paul stood up in the midst of the Areopagus, and said, ‘Gentlemen of
Athens, I see that you make a great display of piety. (23) For as I passed by
and looked at your objects and instruments of worship I saw among other
things an altar on which was inscribed, To an unknown god. What therefore
you worship in ignorance, that I proclaim to you. (24) The God who made
the world and all the things that are in it, since he is Lord of heaven and earth
does not live in shrines made by human hands, (25) nor is he served by
human hands as though he were in need of anything, since he himself gives
to all life and breath and all things. (26) He made of one origin every race of
men to dwell upon the whole face of the earth, having appointed fore-
ordained seasons and the boundaries of their habitation, (27) with a view to
their seeking God, though indeed his being is2 not far from each one of us.
(28) For in him we live and move and are; as indeed some of your own poets
have said: “We too are his family.” (29) Since then we have our being3 as
God’s family, we ought not. to suppose that the divine being is like gold or
silver or stone, an object carved by man’s art or imagination. (30) So the
times of ignorance God has overlooked, but now he is commanding men that
they should all everywhere repent, (31) inasmuch as he has set a day on
which he will judge the whole world in righteousness by a man whom he has
appointed. Of this appointment he has provided proof in that he raised him
from the dead.’
(32) When they heard of the resurrection of the dead some mocked, but
others said, ‘We will hear you on this topic again.’ (33) So Paul went out of
their midst. (34) But some men adhered to him and believed, among whom
were Dionysius the Areopagite, and a woman, Damaris by name, and others
with them.

1Greek, them.
2RSV, NEB, NJB, he is; Greek perhaps, he exists.
3See n. 2.
822
 45. PAUL AT ATHENS. 17.16-34. 823

Bibliography
P. Auffret, NovT 20 (1978), 185-202.
T. D. Barnes, JTS 20 (1969), 407-19.
C. K. Barrett, FS Sawyerr, 69-77.
C. K. Barrett, ZThK 86 (1989), 18-32 (ET, Jesus and the Word (1995),
149-62).
G. Bornkamm, 2.119-37; 4.149-61.
C. Burchard, ZNW 61 (1970), 159f.
F. C. Burkitt, JTS 15 (1914), 455-64.
J. Calloud, RechScR 69 (1981), 209-48.
P. Colaclides, VigCh 27 (1973), 161-4.
H. Conzelmann, FS Schubert, 217-30.
F. G. Downing, NTS 28 (1982), 546-59.
A. Μ. Dubarle, RScPhTh 57 (1973), 576-610.
J. Dupont, Études 157-60.
J. Dupont, Bib 60 (1979), 530-46 (= Nouvelles Études 380-423).
Μ. J. Edwards, ZNW 83 (1992), 266-9.
W. Ehester, FS Bultmann (1954), 202-27.
W. Ehester, NTS 3 (1957), 93-114.
B. Gärtner, The Areopagus Speech and Natural Revelation (1955).
D. J. Geagan, ANRW 2.7.1 (1975), 371-437.
S. Halstead, FS Lake (1937), 139-43.
C. J. Hemer, NTS 20 (1974), 341-50.
H. Hommel, ZNW 46 (1955), 145-78; 48 (1957), 193-200.
P. W. van der Horst, NovT 16 (1974), 309.
P. W. van der Horst, in Knowledge of God in the Graeco-Roman World ed.
R. van den Broek, T. Baarda, J. Mansfield (1988), 19-42.
P. W. van der Horst, ANRW 2.18.2 (1989), 1426-56.
H. Külling, ThZ 36 (1980), 65-83.
J. C. Lebram, ZNW 55 (1964), 221-43.
L. Legrand, in Coppens, BEThL 41 (1976), 337-50.
R. S. Mackenzie, JBL 104 (1985), 637-50.
A. J. Mattili, CBQ 34 (1972), 276-93.
W. Nauck, ZThK 53 (1956), 11-52.
A. D. Nock, Gnomon 25 (1953), 506.
E. Norden, Agnostos Theos (41956).
R. F. O’Toole, RB 89 (1982), 185-97.
Η. P. Owen, NTS 5 (1959), 133-43.
É. des Places, Bib 46 (1965), 219-22.
É. des Places, Bib 52 (1971), 526-34.
Μ. Pohlenz, ZNW42 (1949), 69-104.
K. O. Sandnes, JSNT 50 (1993), 13-26.
824 COMMENTARY ON ACTS

R. E. Wycherley, JTS 19 (1968), 619-20.

D. Zweck, NTS 35 (1989), 94-103.

Commentary
Escaping from Beroea, Paul reached Athens alone, where he awaited
Silas and Timothy. Shocked by the idolatry of the city he discoursed
in the synagogue and the Agora. His reception was mixed; and when
he took the opportunity of addressing the Areopagus court the
response was again divided. A few disciples however were made.
That Paul visited Athens is confirmed by 1 Thessalonians (see 3.1,
and cf. the references to Achaea in 1.7,8). That Luke’s account of the
visit is at least incomplete is shown by 1 Thess. 3.2, in which Paul
sends Timothy back from Athens to Thessalonica; Timothy’s pres-
ence in Athens is not mentioned in Acts. There is in this section no
use of the first person plural, nothing to suggest that the writer was
present—nothing, that is, apart from the content of the narrative,
which shows some awareness of what Athens was like: the forest of
idols (or, as otherwise expressed, of priceless works of art); the
curiosity of the population; the presence of philosophers, notably
Epicureans and Stoics; the existence of a court known as the
Areopagus (or, much less probably though equally valid as showing
acquaintance with the city, of the Areopagus hill); a particular altar,
dedicated to an Unknown God; the fact that Paul made a speech; the
names of the two converts, Damaris and Dionysius. To these
observations may be added a few touches which show some
knowledge of the most famous of all Athenian citizens, Socrates.
These features may constitute a case for believing that the writer was
personally familiar with the city, whether or not he had been there in
company with Paul. It is not however a wholly convincing case; there
is little in it to prove first-hand knowledge. The notable features of
Athens were very widely known and the picture given in this chapter
may be derivative (see on v. 21). There is however nothing in vv.
16-21, 34 that is in any way incredible. If we know anything at all
about Paul it is that in whatever city he found himself he would use
every opportunity—synagogue and market place—to commend the
Gospel, and though the writer has probably gone out of his way to
hint at an analogy with Socrates, the summary of Paul’s teaching in
v. 18 (He was preaching as good news Jesus and the resurrection)
bears remarkable resemblance to the word of faith quoted by Paul in
Rom. 10.9 (If with your mouth you confess Jesus as Lord, and in
your heart believe that God raised him from the dead, you will be
saved). The speech attributed to Paul (vv. 22-31), however, is a
different matter.
Weiser (458f.) rightly says that the preaching in synagogue and
agora, and the conversion of Dionysius and Damaris, may be taken
 45. PAUL AT ATHENS. 17.16- 34. 825

from a pre-Lucan itinerary, but the Areopagus scene and speech must
be viewed as ‘Produkt der luk Gestaltungskunst und nicht als
Detailwiedergabe eines historischen Einzelereignisses’. Pesch
(2.131) takes vv. 16-20, 22, 23 to come from the pre-Lucan source,
v. 21 to be Lucan redaction. Weiser’s rhetorical analysis is instruc-
tive, though possibly too formal; he takes (457) vv. 22f. as Captatio
benevolentiaen 24-26 as Narration 27-29a as Argumentation 29bc as
Reprehensio. These terms are suitable for the analysis of a full-length
speech, less so for a sketch of a dozen or so lines. Luke has probably
given as short a summary as he could of the kind of address to
Gentiles that Hellenistic Christians had inherited from Hellenistic
Jews. This being so, it is correct to say (again with Weiser, 479),
‘Methodisch ist es erforderlich, den Text weder einseitig atl.-biblisch
noch einseitig stoisch auszulegen, sondern gemäss des neuen
Kontextes.’
Luke no doubt supposed that this kind of address was what Paul
would have said to the Areopagus Court; it was ‘... not what
happened but Luke’s idealized version of what ought to have
happened’ (Johnson 318). It might be a little better to say that Luke
was not in a position to recount something that he had himself heard
but used what had come to be the accepted Christian approach to
Gentiles. It is very doubtful whether he was correct in ascribing this
approach to Paul. There is an admirable brief summary of the
contrast between Acts 17.22-31 and Rom. 1.18-31 in Maddox
(83f.); see also (in addition to commentaries and other works listed in
the bibliography on pp. 823f.) Wilson (Gentiles 196-218). Haenchen
(508) explains the difference in terms of the two different lines that
may be detected in Jewish-Hellenistic mission propaganda: the mild
line, taken for example by Aristobulus (who, like Paul in Acts,
quotes Aratus) and the strict line taken in the Sibylline fragments. Of
these the Areopagus speech takes the former (as the Pastorals did—
Wilson, Pastorals 31), Paul (in Romans 1) the latter. This is an
observation worth noting, but it is inadequate as an analysis of the
difference between the two passages. The same may be said of
Schneider (2.234), who emphasizes—questionably—that the Areo-
pagus speech is distinctively Christian. Paul’s approach to the
Gentile world is essentially Christocentric (cf. e.g. 1 Cor. 2.2), and
his criticism of Gentile society, though Jewish in its line of argument,
develops the attack on idolatry found in the Wisdom of Solomon in a
way that is ultimately Christologically determined. The Areopagus
speech lacks this determining Christological factor. It is again not
adequate to say that the language of the speech is not Pauline.
‘Though the content of the speech is far from incompatible with
Paul’s ideas, the language is wholly un-Pauline ... The speech was
certainly not composed by Paul; it was composed by Luke ... But the
doctrine which it conveys is not the product of Luke’s fancy, but of
826 COMMENTARY ON ACTS

the mind of the early church. In as far as this is true, it may represent
in content, though not in expression, the sort of message which Paul
used to deliver to Gentile audiences’ (Hanson 182f.). That Paul
sought to turn his Gentile, idolatrous, hearers from their idols to the
living and true God is proved by 1 Thess. 1.9, 10; that he could do
this by means of his preaching of Christ crucified is shown in ZThK
86 (1989), 18-32 (see bibliography). That he sometimes engaged in
polemical speech against materialism, idolatry, and polytheism may
well be true; these ways of thinking are open to philosophical attack,
but, if we may judge from the epistles, such negative themes did not
constitute the basis and prime content of his message. The world’s
monotheistic and noetic wisdom, no less than its idolatrous wisdom,
had failed to lead it to the power and wisdom of God, which were
uniquely revealed in Christ crucified.
To say this is not to disparage Luke’s work or the preaching of his
contemporaries. That the chapter was of great importance to him is
certain. ‘Athen wird durch die Szene (ähnlich wie Jerusalem in Kap.
15 zum ekklesiologischen) zum geistig-kulturellen Mittelpunkt der
Gesamtdarstellung’ (Schille 354). It had in Luke’s time become clear
(as it had not in Paul’s) that Christianity was to spend some time in a
world that had an important ‘geistig-kulturell’ element, and it was
necessary that Christians should be able and willing to converse with
those who represented this element (just as it had long been
necessary for Jews to do so). There is much truth in an observation of
Bauemfeind’s (215): ‘So wenig Paulus (Rm 9-11) seine Hoffnung
auf die endliche Errettung Israels aufgeben kann, so wenig kann
deshalb Lk seine Hoffnung auf die Rettung jener unerschöpflich
reich beschenkten Welt aufgeben, die der Name Athen umfasst. Gott
hat zu viel an Athen getan, um es wieder los zu lassen.’ It was
however Luke not Paul who perceived this, and Paul would have
insisted on what Luke does in fact with some success achieve: the
restriction of the use of philosophy to those themes which it shares
with the OT. ‘Die entscheidende Frage für die Beurteilung der Rede
ist, ob darin tatsächlich die nicht-biblischen Gedanken und Wendun-
gen das Übergewicht haben oder aber doch die biblischen’ (Stählin
230f.). Johnson’s observation (319), that Luke recognizes Greek
philosophy ‘as a legitimate conversation partner in the approach to
God’, needs this qualification.

16. Έν δέ ταις Άθήναις. Athens was no longer as exciting a

place as it had been in, say, the days of Pericles; but it was exciting
enough, and the chapter shows that Luke was not unaware of this,
though he may have known it only second-hand. There is no first
person plural here. Sulla, besieging the city in the Mithridatic war
and approached by its representatives with appeals based on its
distinguished past, replied that he had not come to study history but
 45. PAUL AT ATHENS. 17.16-34. 827

to put down rebels (ού φιλομαθήσων ... ἀλλὰ τούς ἀφισταμένους

καταστρεψόμενος, Plutarch, Sulla 13 (460)); but Rome was aware
of the history and respected it as history; Athens was well treated,
and continued to be well treated and respected in its role of notable
university city. Luke’s narrative in this chapter shows that the story
of Socrates was not forgotten and that more recent philosophers were
still at work. ‘Adsunt Athenienses, unde humanitas, doctrina, religio,
fruges, jura, leges ortae, atque in omnes terras distributae putantur’
(Cicero, Pro Flacco 26 (62)).
ἐκδεχομένου αυτούς: but (as far as Luke informs us) in vain, for
Silas and Timothy did not come, only joining Paul later in Corinth
(18.5), when they came from Macedonia, with no suggestion that
they had left Macedonia for Athens and then returned. But 1 Thess.
3.1f. shows that Timothy (at least) joined Paul in Athens. Perhaps he
had come but not Silas, since when Paul sent Timothy back to
Thessalonica he was left alone. This however is not certain for Paul
uses the plural—εύδοκήσαμεν καταλειφθηναι ἐν Άθήναις μόνοι.
This may be a genuine plural (we = Paul and Silas) or an epistolary
plural. The fact that Timothy and Silas are not said to have joined
Paul in Athens may account for the variants (for αύτούς τού
Παύλου: αὐτοῦ, *; αὐτοῦ τού Παύλου, D*). This verse (but see
on ν. 17) suggests that at first Paul occupied himself simply in
observation of the city in which he found himself and did not
immediately begin his mission.
His spirit (πνεύμα—here simply his inward life, thought and
feeling) was vexed. For παροξύνεσθαι cf. the use of the cognate
noun παροξυσμός in 15.39. Paul had not ceased to be a Jew, whose
primary theological conviction was that there was but one God, and
that he was to be worshipped without the use of images (Exod.
20.2-5; Deut. 5.6-9). He was vexed to find the city a ‘veritable
forest of idols’. The meaning of κατείδωλος is discussed by R. E.
Wycherley (JTS 19 (1968), 619f.), who draws attention to the fact
that κατά in composition often refers to an abundance of something,
but that ‘a very large proportion of these words refer to luxuriant
vegetation’ (619); ‘the association with trees and plants is predomi-
nant’ (620). He draws attention to the Henns, which were partic-
ularly numerous. ‘We are told of a particularly large and important
accumulation which stood (we now know) at the north-west comer
of the agora, between the Poikile (Painted) Stoa and the Basileios
(Royal) Stoa; in fact the figures so dominated the scene that the place
was called simply “the Herms”. This was the main approach to the
agora, by which Paul would probably enter as he came up from
Peiraeus. Moreover, these stoas, and above all the Poikile, were the
favourite resort of philosophers, and one need not doubt that Paul
frequented them as Socrates, Krates, Zeno, and many others had
done before him. The adjacent Herms more than anything else would
828 COMMENTARY ON ACTS

make him feel that at Athens idols were like trees in a wood’ (620).
Wycherley probably makes rather too much of this. He himself cites
a number of examples of κατά in combination with words that have
nothing to do with vegetation. What is important is that Paul was
impressed by Athens not as a city of art but as a city of false religion.
See two columns of quotation in Wettstein (2.562f.). Contrast Livy
45.27, ... simulacra deorum hominumque, omni genere et materiae
et artium insignia.
For θεωροῦντος ( A B E 33 81 323 (614) 945 1175 1739 al), D
Ψ 0120 <a> have θεωροῦντι, assimilating to the immediately preced-
ing αὐτφ rather than the slightly more remote genitive. It is possible
to defend each case, or rather to attack each case; neither is elegant. It
would have been better to recast the sentence.
οὖσαν τήν πόλιν, accusative and participle after a verb of
perceiving. Here Luke’s style is correct.

17. As in Thessalonica (17.1f.) and Beroea (17.10) Paul visited the

synagogue and used it as an opportunity of evangelism. On this
occasion, however, the synagogue is dismissed in a sentence. There
were many synagogues but there was only one Athens, and the
notable scene in Athens was the Agora.
For the synagogue as an institution see on 6.8. For Jewish-Greek
inscriptions at Athens see NS 3.65 (CIJ 712-15). None refers
explicitly to a synagogue, though there is evidence that there were
Jews in the city from the second century BC, and it is very probable
that a long-established community would have a building, μὲν συν is
a common expression in Acts; see on 1.6. It often introduces a
development in the story, here the beginning of Paul’s public
ministry. Conzelmann (96) takes διελέγετο (17.2; 18.4, 19; 19.8f.;
20.7,9) to mean preached; it is more likely (especially in view of the
fact that the one verb controls the whole verse and both scenes of
Paul’s activity) that it means that in the synagogue Paul discussed
and argued with the Jews, also with the σεβόμενοι, who presumably
were not Jews either by birth or by proselytization. On the question
whether such persons constituted a distinct and recognized class of
‘God-fearers’ see on 10.2.
Haenchen (496) thinks that Paul used the synagogue on the
Sabbath; on other days he was at work elsewhere, διαλέγεσθαι is a
word that could be appropriately used of discussion in the synagogue
but it was especially suitable to Paul’s activity ἐν τῆ αγορά. It is used
only in the later part of Acts (see above; in the rest of the NT only 3
times). ἀγορά occurs in Acts only here and at 16.19. Both verb and
noun recall the archetypal philosophical figure of Socrates, who was
always available for discussion in the public places of Athens. In
Xenophon, Memorabilia 1.1.10, αγορά is used in the first instance to
denote time: πληθούσης αγοράς εκεί φανερός ἧν, When the market
 45. PAUL AT ATHENS. 17.16-34. 829

was in full swing (i.e. in the morning) he (Socrates) was to be seen

there (i.e. in the market); cf. the preceding clause, πρωί τε γάρ εἰς
τούς περιπάτους καί τά γυμνάσια ἤει. For διαλέγεσθαι see e.g.
Plato, Apology 33a, ουδέ χρήματα μἐν λαμβάνων διαλέγομαι; 19d,
εμού ... ἀκηκόατε διαλεγομένου. Cf. 38a. For the dialectical
method (which grew out of market-place arguments), Republic 454a,
οὐκ ἐρίζειν αλλά διαλέγεσθαι.
πρός τούς παρατυγχάνοντας recalls Socrates’ readiness to con-
verse with anyone willing to converse with him—δτω ἄν ἁεί
ἐντυγχάνω υμών (Plato, Apology 29d). The present participle
παρατυγχάνοντας expresses better than the aorist (D*,
παρατυχόντος) the fact that this was Paul’s habit.
For topography and archaeology see the Princeton publication,
The Athenian Agora, 1953 onwards.

18. Among the strangers whom Paul chanced to meet were

philosophers. Luke mentions Epicureans and Stoics. There are good
summary articles, with bibliographies, in OCD; see also NT Back-
ground 65-77, 78-81. It may be asked why these two groups are
singled out. Rackham (303) remarks that both were practical rather
than speculative philosophers, but the answer may be that their views
are alluded to, and indeed used, in the Areopagus speech that
follows. It is of course possible to put this differently and say that
Paul included Epicurean and Stoic material in his speech because he
happened to have met Epicurean and Stoic philosophers. The speech
includes a criticism of popular religion: If we are to conceive a god
who created the universe (and the Epicureans were not necessarily
atheist, though they believed that any gods that existed were, in all
senses, too remote to be taken into practical account) then it is
ridiculous to bring him a pig or a sheep, as if he were a hungry
beggar, or to suppose that he could be localized in a building. This
was an Epicurean criticism of popular religion. The Stoics believed
that the human race was one, proceeding as it did from a single point
of origin, that there was a divine being (note the neuter τὀ θειον in v.
29), conceived in pantheistic rather than personal terms (in him we
live and move and have our being, v. 28), and that it was man’s duty
to seek and to live in accordance with this indwelling god. The
speech (whatever we make of its origin; see above, pp. 825f.) may
from one point of view be looked on as an attempt to see how far a
Christian preacher can go in company with Greek philosophy.
The Stoics were a popular group, whose beliefs the Romans found
naturally attractive; the Epicureans were not comparably numerous,
and the use of Epicurean doctrine in the speech is the best explana-
tion of the reference to them.
The philosophers συνέβαλλον αὐτῷ, argued with him; the word is
Lucan (3 times in Lk, 4 times in Acts; nowhere else in the NT). Here
830 COMMENTARY ON ACTS

it means that they argued with him; not all of them politely, though
the impolite are made to express themselves in (for the NT)
unusually stylish Greek—unless, that is, Page (193) is right in the
belief that in καί τινες ἐλεγον the subject is ‘probably not the
philosophers but generally some of those who heard Paul’. τί ἄν
θέλοι.. . ;ἄν with the potential optative ‘was old-fashioned in the NT
age’ (Μ. 3.123). For its infrequency in papyri, LXX, and Apostolic
Fathers see BDR § 385.1, η. 1. θέλειν λέγειν recalls the French
vouloir dire, to mean, but it is not quite so firm an idiom. ‘What does
he want to say?’ Perhaps, ‘What is he getting at?’ See LS 479, ἐθέλω
II 3; e.g. Herodotus 4.131.2,... γνῶναι τὸ θέλει τά δώρα λέγειν.
Used of a person present αὖτoς is often disparaging; so BDR §
290.1, η. 1, adding Lk. 15.30; 18.11. σπερμολόγος is certainly
disparaging. It is used by Demosthenes (18.127(269)) along with
περίτριμμ’ αγοράς, ὄλεθρος γραμματεύς. Literally it means picking
up seeds, and so came to be used of an inferior speaker or writer who
picks up and uses as his own ideas that he has found in others. It is
again connected with περίτριμμα by Philo (Legatio ad Gaium 203),
where Colson (LCL) translates ‘scrap retailer, piece of riff-raff’. BA
(1522) have Schwätzer, which does not seem quite right; it fails to
take up the first cumbersome part of ‘one who picks up and retails
scraps of knowledge, an idle babbler, gossip’ (LS 1627). Hanson
(176) has ‘an intellectual magpie’. Begs, 4.211 gives a long quotation
from Eustathius of Thessalonica on Odyssey 5.490, suggesting cock-
sparrow, which does not suggest (to me) the right image. ‘Journalist’
is an admirable suggestion provided it is understood that an inferior
and not a superior journalist is intended. ‘What does this third-rate
journalist want to tell us?’ will perhaps do.
The hints at the figure of Socrates noted in v. 17 now become more
definite in what is nearly an explicit quotation. According to
Xenophon, Memorabilia 1.1.1 the charge against Socrates was (in
addition to the corruption of the youth), ’Αδικεῖ Σωκράτης οὕς μεν
ή πόλις θεούς οὐ νομίζων, ἔτερα δέ καινά δαιμόνια εἰσφέρων. For
Athenian objections to strange gods and strange religions see
Josephus, Apion 2.265-267, and cf. Euripides, Bacchae 255f., τόνδ’
αὖ θέλεις τόν δαίμον’ ἁνθρώποισιν εἰσφέρων νέον. In the charge
δαιμόνιον is of course used in a sense different from that of the
gospels (e.g. Mk 1.34); it means a deity, though clearly not in a
monotheistic sense; cf. the special δαιμόνιον of Socrates (e.g.
Xenophon, Memorabilia 1.1.2) and the wording of the charge in
Plato, Apology 24bc, which includes ἔτερα δέ δαιμόνια καινά. Of
such foreign deities it was alleged that Paul was a καταγγελεύς. The
noun does not occur elsewhere in the NT, but καταγγέλλειν is
characteristic of Acts (11 times, including 17.3, 13, 23; rest of the
NT, 7 times); preacher, or herald (as at the games). The objection to
introducing new deities may have been partly political; Dio Cassius
 45. PAUL AT ATHENS. 17.16-34. 831

52.36.1f. records that Maecenas advised Augustus to hate and punish

those who bring in new ideas about τό θεῖον.
The impression Paul gave was based on the fact that he τόν
Ίησοῦν και τήν ἁνάστασιν εὐηγγελίζετο (the verb is a near enough
synonym—but a Christian synonym—of καταγγέλλειν). It can be
readily understood that Paul’s preaching of Jesus could be regarded
as the proclamation of a new δαιμόνιον. ἁνάστασις as commonly
understood will hardly serve as a second personal deity (the plural
ξένων δαιμoνιων, taken literally, implies at least two). For this
reason it has been suggested (e.g. Bultmann, Theologie 80) that
Άνάστασις was taken to be a female deity corresponding to the male
deity Jesus. The suggestion is superficially attractive but could be
maintained only if there were reason to think that Paul in his
preaching constantly referred in quasi-personal terms to ’Ιησούς καί
ἁνάστασις. This is unlikely; it is much more probable that he used
verbs: Χριστός ἁπέθανεν ... ὅν ὁ θεός ἁνέστησεν (or ἤγειρεν).
Luke gives a brief summary of what Paul said, less than perfectly
clear; cf. 4.2. The Athenian comment is cast in a form intended to
recall the story of Socrates, and means no more than, This is a
strange new religion, with all this talk about a man called Jesus and a
resurrection. ‘Jesus and the resurrection’ is Pauline enough, and
indeed probably pre-Pauline; see Rom. 10.9 (Romans 186f.). But the
formulation in Acts is different, and far less clear and trenchant.
The last clause in the verse, ὅτι... εύηγγελίζετο, is omitted by D
gig, possibly because it did not seem consistent with the plural ξένων
δαιμoνιων, possibly (Metzger 455) because the writer did not wish
to class Jesus among the δαιμόνια.
Taylor (5.304-6) reconstructs the place where the whole incident
(vv. 18-34) took place—the Στοά βασίλειος, at the NW comer of
the Agora.

19. ἐπιλαβόμενοι is an ambiguous word. It may mean an arrest,

with or without violence (16.19; 18.17; 21.30, 33) or a well
intentioned attachment (9.27; 23.19). It is a Lucan word; in addition
to the seven occurrences in Acts there are five in Luke, six in the rest
of the NT. The meaning here will depend on the view that is taken of
the proceedings as a whole: Was Paul arrested and brought before the
Council of the Areios Pagos to be tried by them, or was he conducted
to the Areios Pagos hill in order that he might in a convenient place
(Haenchen’s statement (498) that the hill was too small for a meeting
is rightly contradicted by Marshall (285)) address a number of
interested inquirers? Or is this statement of alternatives fallacious?
The history of the Areopagus court (ή βουλή ή ἐξ Άρείου πάγου,
Demosthenes 18.133 (271); ή β. ή έν Άρείῳ πάγῳ, 59.80 (1372)) in
the first century, indeed in the Roman period generally, is obscure. It
has however been shown, especially by T. D. Barnes (JTS 20 (1969),
832 COMMENTARY ON ACTS

407-19; useful bibliography), that it remained in this period the

supreme authority in Athens. ‘The Areopagus seems to be the
effective government of Roman Athens and its chief court. As such,
like the imperial Senate in Rome, it could interfere in any aspect of
corporate life—education, philosophical lectures, public morality,
foreign cults’ (413). It could try crimes of any kind, and probably had
authority to inflict capital punishment (412). It is incorrect to say that
it no longer met on the Areopagus Hill but only in the Stoa Basileios
(409; see however Hemer, NTS 20 (1974), 341-50). Barnes con-
cludes, ‘The obvious meaning of the words in Acts should be
accepted: Paul was taken before the Areopagus, i.e. before the
council sitting on the hill’ (410). This perhaps goes too far. The
words ἐπί τὸν Άρειον πάγον ἤγαγον may mean, They brought him
before the Areopagus council (wherever it met), or They brought him
to the Areopagus hill (for whatever purpose). Luke may have meant
both, but if he intended to say both he should have done so explicitly.
According to Schneider (2.237) he is using a ‘literarischer Topos’;
one may compare the way in which the Areopagus council is brought
in by Cicero, De Natura Deorum 2.29 (74) (Si quis dicat, Athe-
niensium rempublicam consilio regi, desit illud, Ariopagi). Weiser
(466) notes (1) that there is a change of place from the Agora, (2) that
more are present than the members of the council; ‘Lukas schwebte
eine ideale Szene vor.’ Barnes concludes that, as in many other
Hellenistic cities, Paul was brought before the local authority,
accused of introducing a new religion. He seems to be mistaken in
the view that ἐπιλαμβάνεσθαι must imply a violent arrest; see above.
Begs. 4.213 had already observed that here Luke uses the local name,
Areopagus, as he does at Philippi (στρατηγοί), Thessalonica
(πολιτάρχαι), and Corinth (ανθύπατος)—and gets it right every
time.
’Άρειον πάγον is correctly so written, divisim; this is confirmed by
many inscriptions. The compound Άρειόπαγος ‘never had any real
existence’, though Αρεοπαγίτης (v. 34) was formed from it as if it
existed; at a late period it came into existence as a back-formation
from ’Αρεοπαγίτης (Μ. 2.151, 277).
δυνάμεθα in this sentence is more like the English ‘May we ... ?’
than ‘Can we ... ?’, ‘Are we able ...?’; that is, it looks more like a
polite request than a demand for an explanation, such as a court
might have made.
In ή καινή αὕτη ή ὑπό σοῦ λαλουμένη διδαχή the second article
is superfluous (see Robertson 701, 785); not surprisingly it is omitted
by B D pc. For λαλουμένη, D syp have καταγγελλομένη (v. 18) and
E 81 have λεγομένη. Instead of τε, and suggesting an adversative
element in the sentence, B Ψ 33 36 81 453 1241 al have δέ. More
interesting than these variants is the fact that D (614 pc syh**) rewrite
the beginning of the verse: μετά δὲ ἡμέρας τινὰς ἐπιλαβόμενοι
 45. PAUL AT ATHENS. 17.16-34. 833

αὐτοῦ ἤγαγον αυτόν ἐπἰ Άριον πάγον, πυνθανόμενοι καί

λέγοντες. The insertion of certain days gives Paul a period of
undisturbed activity, and inquiring suggests interest rather than an
attempt to set in motion a legal process.
20. This verse is a duplicate of vv. 18,19. The ξένα δαιμόνια have
now become simply ξενίζοντα, by which Luke presumably means
things (statements, beliefs) that are strange, surprising, though at
every other place in Acts where the verb is used it means to receive
as a guest (or, in the passive, to lodge); so 10.6, 18, 23, 32; 21.16;
28.7. Elsewhere in the NT it is used only at Heb. 13.2 (to entertain)
and 1 Pet. 4.4, 12 (both, passive, to be surprised). ‘You are bringing
into our ears (for ἀκοαί as ears, rather than the faculty of hearing, cf.
Mk 7.35) surprising things.’ For εισφέρεις cf. Euripides, Bacchae
650, τούς λόγους γάρ ἐσφέρεις καινούς αεί. Cf. also Sophocles,
Ajax 148f., τοιούσδε λόγους ψιθύρους πλάσσων εις ωτα φέρει
πάσιν Όδυσσεύς. Instead of εισφέρεις, Ε has ρήματα εισφέρεις, D
co have φέρεις ρήματα, and (Ψ) pc have εισφέρει.
The questioners want to know what these surprising things mean,
θέλει εἶναι (cf. the similar but not identical θέλειν λέγειν in v. 18).
In τίνα θέλει ταῦτα εἶναι BDR § 299.1, n. 1 take τίνα as neuter
plural subject of θέλει, and translate, ... wie es sich damit verhält.
Instead of τίνα θέλει (P74 A Β Ψ 33 36 81 945 1175 1739 al), D Ε
<a> have τί ἂν θέλοι (cf. the optative in ν. 18).
The verse suggests nothing more than a desire for information and
enlightenment, but the severe attitude reported by Josephus, Apion
2.267 (... τιμωρία κατά τῶν ξένον εἰσαγόντων θεόν ὥριστο
θάνατος) should be borne in mind.
21. The Athenians have expressed a desire to hear the new
message that Paul appears to have brought; Luke adds a note that
explains this and at the same time shows an acquaintance with
Athenian life and characteristics. This acquaintance is not necessarily
first-hand; Athenian curiosity was widely recognized and com-
mented on. So e.g. Demosthenes 4.10(43), ἢ βούλεσθ’, είπέ μοι,
περιίοντες αυτῶν πυνθάνεσθαι, 'λέγεται τι καινόν;’ γένοιτο γάρ
ἄν τι καινότερον ή ... Cf. Thucydides 3.38.4. The proverbial
characterization of the Athenians (inter alia) led A. D. Nock to the
judgement (Gnomon 25 (1953), 506), ‘brilliant as is the picture of
Athens [in Acts 17], it makes on me the impression of being based on
literature, which was easy to find, rather than on personal observa-
tion’. The fact that others had observed Athenian curiosity does not
in itself prove that Luke had not himself observed it; but it makes it
more difficult to prove, or to feel confident, that he had.
In this curiosity native Athenians were joined by οι ἐπιδημούντες
ξένοι. For έπιδημεῖν cf. 2.10 (and 18.27 P38vid D). For the com-
bination with ξένοι cf. Lysias 12.35, ὄσοι ξένοι ἐπιδημούσιν.
834 COMMENTARY ON ACTS

Many visitors came to Athens, some as serious students, some as

tourists. It is not clear why D should add εις αυτούς after
ἐπιδημούντες. The preposition may be intended to bring out the fact
that the foreigners had entered Athenian society and so had come to
share its ways.
For εὐκαιρεῖν cf. Mk 6.31 (with an infinitive, as here); 1 Cor.
16.12. For the use of the infinitive see BDR § 392.3, n. 11.
The Athenians desire ἤ λέγειν τι ἢ ἀκούειν τι καινότερον (the
second τι is omitted by D E <a>). The comparative adjective is
probably used for the positive; so e.g. Μ. 3.30, adding ‘but possibly
an Atticistic refinement: newer'. It is not clear what this means;
possibly ‘newer than the last thing we heard’. Zerwick (§ 150) agrees
that the positive is the probable meaning, ‘nisi quis mavult hunc
comparativum sumere pro superlativo: “recentissimum quodque” ’.
BDR § 244.2 after saying that καινότερον can ‘mit dem Positiv
wiedergegeben werden’ rather surprisingly translate ‘etwas ganz
Neues = das Allerneuste’, which looks more like a superlative.
For the whole picture cf. Demosthenes 11.17 (156), ήμεῖς δέ
(εἰρήσεται γὰρ τἀληθή) οὐδέν ποιοῦντες ενθάδε καθήμεθα,
μέλλοντες αεί καί ψηφιζόμενοι καί πυνθανόμενοι κατά τήν
αγοράν εἴ τι λέγεται νεώτερον.
This verse had probably been read by the author of the Acts of
Philip 7(2) (L.-B. 2.2.4), εἰ δέ τι καινότερον ἐχεις, ω ξένε, ἐπίδειξον
ἡμῖν ἀφθόνως μετά παρρησίας· οὐδενός γάρ άλλου χρείαν ἔχομεν
ή μόνον ἀκούειν τι καινότερον. 8(3) ... Ώ ἄνδρες φιλόσοφοι τής
"Ελλαδος, εἰ βούλεσθε καινοτέρου πράγματος ἀκοῦσαι καί ἐστε
ποθοῦντές τι καινότερον ...

22. Σταθείς: Paul stood. This would be expected in Athens, as it

would not in the synagogue at Pisidian Antioch (13.16). He stood έν
μέσω τού Άρείου πάγου. For Areopagus see on v. 19; ἐν μέσω (cf.
1.15; 2.22; 4.7; 27.21) suggests in the midst of a group of people
rather than at the mid-point of an area.
Άνδρες ’Αθηναίοι. The form of address is common in Acts (cf.
Άνδρες Γαλιλαῖοι, 1.11, and 20 other examples), but here, with
'Αθηναίοι, it adds to the reminiscence of Socrates (e.g. Plato,
Apology 17a).
The interpretation of this verse has been much discussed: is it an
accusation (‘unter schwelligen Spott’, Schille 356), or a captatio
benevolentiae (forbidden before the Areopagus court, according to
Lucian, De Gymnasiis 19)? The answer must to a great extent turn on
the interpretation of ὡς δεισιδαιμονεστέρους, discussed below.
Luke has however prepared the way for the understanding of the
address by recording Paul’s observation of the city as κατείδωλον (v.
16). The Athenians practise religion, but their religion is expressed in
a profusion of idols. For δεισιδαίμων cf. in Acts itself δεισιδαιμονία
 45. PAUL AT ATHENS. 17.16-34. 835

at 25.19. Festus uses this word when speaking of Judaism to Agrippa

II, whom, as a Jew, he would not deliberately insult by calling his
religion a superstition.
LS (375) define δεισιδαιμονέω as 'have superstitious fears ... ,
rare in good sense, to be religious'; δεισιδαιμονία as fear of the
gods, religious feeling ... in bad sense, superstition'; δεισιδαίμων as
f' earing the gods, 1. in good sense, pious, religious ... 2. in bad
sense, superstitious'. Examples can be found for both senses. Yet the
δεισιδαίμων described by Theophrastus (16(17)) is not an admirable
character, ‘δεισιδαιμονία is cowardice in face of the supernatural
(δειλία πρός τό δαιμόνιον).’ ‘The δεισιδαίμων is one who, if
anything dirty touches him, will wash his hands, sprinkle himself
with water from a holy fountain, and walk about all day with
his mouth full of bay leaves.’ In De Superstitione ( = περί
δεισιδαιμονίας) Plutarch places atheism and δεισιδαιμονία at oppo-
site extremes (1 (164E)). The atheist is unmoved by any thought of
the divine (τὸ θειον); the δεισιδαίμων is moved, but in a perverse
way (2(165BC)). Atheists acknowledge no gods at all; δεισιδαίμονες
take the gods to be evil, holding what is kind to be fearful, what is
fatherly to be tyrannous, what is gentle to be hurtful, what is kindly
to be savage (6 (167DEF)). Neither Theophrastus (4th/3rd century
BC) nor Plutarch (1st/2nd century AD) would have been pleased to
be described as ὡς δεισιδαιμονέστερος. The word could indeed be
used in a good sense; so e.g. Xenophon, Cyropaedia 3.3.58 (oἱ
δεισιδαίμονες ἧττον τούς ανθρώπους φοβούνται); Agesilaus 11.8
(αεί δέ δεισιδαίμων ἦν, νομίζων τούς μὲν καλώς ζῶντας οὔπω
εὐδαίμονας, τούς δ’ εὐκλεῶς τετελευτηκότας ήδη μακαρίους). An
inscription describes the person commemorated as φίλος θνητoῖς εις
τάθανάτους δεισιδαίμων (IG 14.1683). The simple fact is that
δεισιδαίμων means religious. To the sceptic, this means super-
stitious; to the religious, judgement depends on whether the religion
is one he shares, or at least approves, or is one that he rejects. We
shall see that Paul’s attitude (as represented in the speech) is not
simple; he has however noted with disapproval, one might almost say
with arrogance, that the city is full of idols.
The word was discussed by Hatch (Essays 43-5), who argued that
‘in the first century and a half of the Christian era the words had
come to have in ordinary Greek a bad or at least a depreciatory sense’
(45). This should perhaps be qualified by the addition of ‘especially
among Jews’. Hatch quotes Philo, De Sacrificiis 15 (...
δεισιδαιμονίαν δέ πράγμα αδελφόν ἀσεβεία ... ); Quod Deus 164
(the mean between δεισιδαιμονία and ασέβεια is ευσέβεια); De
Gigantibus 16; De Plantatione 107 (δεισιδαιμονία a parasitic growth
(παραναπέφυκεν) on true religion); De Specialibus Legibus 4.147
(γεννήσει γἀρ ή μἐν πρόσθεσις [to ευσέβεια] δεισιδαιμονίαν, ή δ’
ἀφαίρεσις ασέβειαν); Josephus, Ant. 15.227 (but here Marcus and
836 COMMENTARY ON ACTS

Wikgren (LCL edition) translate religious scruples, though they

recognize superstition as a possible alternative); and Plutarch, as
above. Also Justin, 1 Apology 2: those addressed are not ‘under the
dominion of prejudice or a desire to gratify superstitious persons
(ἀνθρωπαρεσκεία τῇ δεισιδαιμόνων)’. Conzelmann’s argument
(97) that δεισ. must be understood sensu bono because it occurs in a
captatio benevolentiae is invalid: Paul’s words are to be understood
as a captatio benevolentiae only if we know that δεισ. is intended
sensu bono.
ὡς δεισιδαιμονεστέρους. Opinions differ not only in regard to the
meaning of the word δεισ. but also in regard to the comparative form
of the adjective and the use of ὡς. Μ. 3.30 comments on the
comparative: ‘... not class. rather but probably popular elative
extremely god-fearing (strengthened by ώς)’. Similarly Zerwick
(§ 148): ‘Paulus Athenienses laudat agnoscens eos ὡς
δεισιδαιμονεστέρους “superstitiosiores”, quod nihil aliud est quam
“valde pios” (comparativus pro superlativo “elativo”)’. Zerwick
has no note on ὡς. It is hard to understand the alternatives in BDR
§ 224.1, n. 2 ‘... ist zweideutig: “ungewöhnlich götterfürchtig”
(klass.) oder “sehr götterfürchtig’”. Perhaps this corresponds to
Robertson (665): ‘... more religious ... than ordinary or than I had
supposed’. A comparative may still be elative even if truly compar-
ative. So far as one may think of δεισιδαιμονεστέρους as standing
for a superlative we may perhaps compare the familiar construction
of ὡς with a superlative—as ὡς τάχιστα, as quickly as possible. But
in the long article on ὡς BA 1789-1793 do not mention Acts 17.22;
under δεισιδαίμων (347) they offer the translation, ‘ich nehme wahr,
dass ihr durch und durch religiöse Leute seid.’ It may be that ὡς here
has something of the ‘as if’ sense to be found in Rom. 9.32 (ὡς ἐξ
ἔργων); Athens presents a show of (idolatrous) piety, but it is an
unreal, uninformed piety directed towards a deity who must remain
unknown (v. 23).
BDR § 416.1, n. 3 note the absence of a participle (contrast v. 16).
Cf. Radermacher (170).
The piety of Athenians was well known (and need not therefore
have been observed at first hand by Luke). The following are among
passages often quoted. Pausanias 1.17.1: The Athenians θεούς
εὐσεβοῦσιν άλλων πλέον. Sophocles, Oedipus Coloneus 260: τὰς γ’
’Αθήνας φασί θεοσεβεστάτας. Josephus, Apion 2.130: ... τούς δέ
ευσεβέστατους τών Ελλήνων ἄπαντες λέγουσιν.
The meaning of this in some ways ambiguous verse can be found
only in the light of ἀγνώστω, ἀγνοοῦντες, in v. 23. See further
below.

23. διερχόμενος here does not have the meaning that διέρχεσθαι
sometimes (see on 8.4) has in Acts. Paul was simply making his way
 45. PAUL AT ATHENS. 17.16-34. 837

through the city; as he went, however, he was looking carefully at

religious objects. ἀναθεωρεῖν is a stronger word than θεωρεῖν (v.
16); διϊστορεῖν (D*) is stronger still. Idols struck the eye; Paul
looked more closely at the σεβάσματα. The word is derived from
σέβας, reverential awe (LS 1587): something viewed with such awe;
broadly, any object related to cultus. At Wisd. 14.20; 15.17;
Josephus, Ant. 18.344 the word is used of objects of idolatrous
worship, and so it is here, though one such object will be found to
point to, or rather to suggest, the true God. εὗρον does not
necessarily imply that Paul was looking for what he found—he came
across. Among various religious objects, σεβάσματα, a βωμός is
almost certainly an altar, though the base of a statue (Homer,
Odyssey 7.100) is, in the context, not impossible. The statue would
be an image of the unknown god. The altar, or base, was inscribed.
There is no difference between ἐπεγέγραπτο and the periphrastic
pluperfect (ἦν γεγραμμένον) of D; the variant, like many other
Western variants, points to a time of greater textual flexiblity than
obtained later. Luke writes καἰ βωμόν, also an altar, that is, in
addition to other things, an altar.
The inscription was Άγνώστῳ θεῷ; on this see especially Begs.
5.240-6; also P. W. van der Horst in ANRWII 18.2 (1989), 1426-56
(with bibliography). No such inscription has been uncovered in
Athens by archaeologists. This of course does not prove that no such
inscription ever existed; countless inscriptions must have perished or
become indecipherable. But already Jerome (Ad Titum 1.12) wrote,
Tnscriptio autem arae non ita erat, ut Paulus asseruit, “ignoto deo”,
sed ita, “diis Asiae et Europae et Africae, diis ignotis et peregrinis”.’
The plural form of the dedication is firmly attested, notably by
Pausanias 1.1.4 (βωμοὶ δέ θεών τε όνομαζομένων αγνώστων καί
ἡρώων καί παίδων τών Θησέως καί Φαλήρου—these are at
Munychaia, a harbour of Athens) and Philostratus, Apollonius 6.3
(... Άθήνησιν, ου καί αγνώστων δαιμόνων βωμοί ἵδρυνται). See
also Tertullian, Ad Nationes 2.9, Nam et Athenis ara est inscripta:
ignotis deis. colit ergo quis quod ignoret? But Tertullian had
presumably read Acts. Pausanias’s reference to heroes lends some
support to the suggestion of R. E. Wycherley (JTS 19 (1968), 620f.).
In the establishment of the agora and adjacent buildings many earlier
burials were disturbed; Athenian δεισιδαιμονία would lead to the
foundation of cults to placate the dead. ‘A legend and a name might
grow and attach itself to the spot, but not necessarily so; the cult
might remain truly the cult of an unknown god. Strictly speaking it
would be a hero cult, and the shrine a heröon, but the distinction is
not at all clear, and even an obscure local hero could be called theos'
(621). This is no more than an interesting possibility. The balance of
probability remains with Jerome, but it is by no means impossible
that Paul saw what he describes—or that there was in Athens for him
838 COMMENTARY ON ACTS

and for others to see what Luke says that Paul says that he saw. Van
der Horst (op. cit.) and Taylor (290) draw attention to a Pergamene
inscription (Die Arbeiten zu Pergamon 1908-1909, II (Mitteilungen
des kaiserlich deutschen archäologischen Instituts, Athenische Abtei-
lung 35), 1910, pp. 454-7), restored, not certainly but with some
probability as θεοις ἀγνώστοις. There could have been a similar
dedication in Athens, due to anxiety lest any gods should be
inadvertently left without due honour. It is worth while to add,
though they furnish no more precise information, Diogenes Laertius,
Epimenides 3 (10), βωμούς ανωνύμους and Tertullian, Adversus
Marcionem 1.9, Invenio plane ignotis deis aras prostitutas, sed Attica
idololatria est. More important is the use that is made of the alleged
inscription. This, even in the singular, implies polytheism; the
speaker makes it monotheist.
Surprisingly, the masculine θεός is taken up as if it were neuter: ὃ
οὖν ἀγνοούντες εὐσεβεῖτε, τοῦτο ἐγώ καταγγέλλω ὑμῖν. This is
the reading of P74 A* B D (81) 1175 pc lat; masculine pronouns
(ὃν ... τούτον) appear in X Ac E Ψ <a> sy Clement. It is likely that
the neuters are original; there was a double reason for changing them,
the grammatical reason that the antecedent was θεός, the theological
reason that Paul was understood to proclaim a personal, not an
impersonal, deity (but cf. τό θειον in v. 29).
The sentence taken as a whole makes two statements, one about
the Athenians and one about Paul. The subjects are different, the
verbs are different, but the objects are the same, as the relative
construction shows (ὄ ... τούτο). The Athenians reverence a certain
object, Paul proclaims it. εὐσεβεῖν (often with a preposition, act
piously towards, rather than an object accusative) denotes the
appropriate attitude towards divine beings, the practice of religion.
καταγγέλλειν in Greek generally is to declare or proclaim anything,
in the NT to proclaim the Gospel (or an aspect of it), or the being
whose person and work constitute or determine the Gospel.
On this passage Bultmann (Theologie 470) writes: ‘Auch insofern
ordnet der Verf. der Act das Christentum als Religion in die
Weltgeschichte ein, als er in der Areopagrede den Paulus an die
heidnische Frömmigkeit anknüpfen lässt durch Bezugnahme auf die
athenische Altar-Inschrift und auf den stoischen Gottesglauben
(17.23, 28). Dadurch wird “die heidnische Geschichte, Kultur- und
Religionswelt als Vorgeschichte des Christentums reklamiert” (Viel-
hauer), und das entspricht der Auffassung der Act vom Verhältnis
des Christentums zum Judentum: die paulinische Gesetzeslehre ist
nicht mehr verstanden, und die jüdische Geschichte ist einfach zur
Vorgeschichte des Christentums geworden?
There is much in this that is well said, but it is important not to
give too heavy a theological treatment to Paul’s (Luke’s) sentence; it
must be understood as a preacher’s ad hoc way of introducing his
 45. PAUL AT ATHENS. 17.16-34. 839

theme, and it would be unfair to hold him bound to all the theological
implications of his illustration. The Athenians (those of them who
were religiously rather than sceptically disposed) reverenced a
considerable number of gods. The preacher could have made a note
of many other σεβάσματα bearing the names of particular gods; he
picked out this god, whose name was not given because it was not
known, as the one whom, to the exclusion of all the others, he
intended to proclaim. The Athenians worshipped ἀγνοούντες; this
picks up the adjective άγνωστος (and perhaps we should add the
δυνάμεθα (implying θέλομεν) γνῶναι in v. 19), but it does so in a
general rather than a particular way, and must not be used as the basis
of inferences about the knowability, or unknowability, of God. Paul
declares: You are religious, but your religion is uninstructed (cf.—
from a different angle—Rom. 10.2: the Jews have a zeal for God but
it is not κατ’ ἐπίγνωσιν; Jn 4.22 is more remote); I am now
proclaiming what you must know if your religion is to be real. ὅ
should be taken rather as the object of ἀγνοοῦωτες than of
εὐσεβεῖτε: What in the practice of your religion you do not know
(namely, the true God; see 1 Thess. 4.5; Gal. 4.8), that I proclaim.
‘Die Welt des Heidentums gilt als in άγνοια und πλάνη versun-
ken’ (Bultmann, Theologie 70). Yet after this beginning Paul goes on
to say much that religious Athenians would find familiar, and never
reaches what in his Gospel was truly (to the heathen world)
unknown. On the other hand, to say that Luke means to show that
Paul is not preaching something that ‘vollkommen ausserhalb des
athenischen Erwartungshorizontes lag’ (Weiser 468) is in danger of
going too far in the other direction; and to claim that the Gentiles live
in a simultaneous Yes and No to the true God (Haenchen 501) makes
Luke too much of a theologian. For the language cf. Isa. 45.15: σὺ
γὰρ εἶ ό θεός, καί οὐκ ἤδειμεν

24. The truth that Paul has to communicate to the ignorant

Athenians is the truth about God and his relation to the world and
men. The first part of this truth, given as a predicate of the opening
words, ὁ θεός, is that God is the creator: ὁ ποιήσας τόν κόσμον καί
πάντα τά ἐν αὐτῷ. This is a truth frequently asserted in the OT, not
only in the creation narratives but in other summary statements
which the present statement resembles without actually reproducing
any of them. The following may serve as examples. Gen. 1.1, ἐν
ἀρχῆ ἐποίησεν ό θεός τόν ουρανόν καί τήν γην; cf. Exod. 20.11.
Isa. 42.5, κύριος ό θεός ό ποιήσας τόν ούρανὸν ... ό στερεώσας
τήν γην καί τά ἐν αὐτῆ; Wisd. 9.1, ὁ ποιήσας τά πάντα, 9, ἐποίεις
τόν κόσμον. Cf. 4 Macc. 5.25, and in the same tradition 1 Clement
19.2; Barnabas 21.5. Philo not merely borrowed and re-expressed the
language of the OT but wrote περί τής κατά Μωυσέα κοσμοποιίας
(de Opificio Mundi). Josephus began the Antiquities with creation,
840 COMMENTARY ON ACTS

altering one word in Gen. 1.1, ἐν αρχῆ ἔκτισεν ό θεός. But the
concept of God as the maker of the universe was Greek as well as
Jewish. Plato’s Timaeus is an account of creation very different from
that of Genesis except in the notion of divine causation. Cf. Epictetus
4.7.6: ὁ θεός πάντα πεποίηκεν τά ἐν τῷ κόσμῳ καί αυτόν τόν
κόσμον. Corpus Hermeticum 4.1: τόν πάντα κόσμον ἐποίησεν ό
δημιουργός. The word κόσμος was used by Greek philosophers
(Schneider 2.239), but, as appears above, it was used by Jewish
writers too. ‘All the things that are in it’ is paralleled also, and in
biblical sources; we may add 1QH 1.13-15, Thou hast created the
earth by thy power, seas and floods [...] thou hast estab-
lished in thy wisdom, and all that is in them thou hast
determined by thy counsel. In Acts the theme of God as creator has
already appeared at 4.24, where the language is even closer to that of
theOT.
God thus described, in terms that would be familiar to both Jews
and Greeks, is naturally κύριος of the things that he has made, that is,
of heaven and earth, the totality of creation, including of course men.
As such a universal creator and lord he is not to be confined within a
space circumscribed by human invention and manufacture. Cf.
7.48-50, where again the language is more scriptural than here,
though in the present verse χειροποίητος, used in secular Greek (e.g.
Herodotus 1.195.1, σκῆπτρον χειροποίητον), cannot fail to recall its
frequent use in OT denunciations of idolatry. The question of the
Jerusalem Temple is not raised here, as it is in ch. 7. Cf. Isa. 2.18;
16.12; Lev. 26.1 (LXX).
Conzelmann (98) points out that Judaism used such polemic
against heathen temples, referring to Josephus, Ant. 8.227ff. (which
does not seem to make his point) and Oracula Sibyllina 4.8-11. But
Greeks could say this too. Plutarch (De Repugnantiis Stoicis 6
(1034B)) quotes a δόγμα Ζήνωνος as follows: ιερά θεών μή
οἰκοδομεῖν ιερόν γάρ μή πολλοῦ ἄξιον καί άγιον οὐκ ἔστιν
οικοδόμων δ’ἔργον καί βαναύσων οὐδέν ἐστι πολλοῦ ἅξιον. SVF
1.264 (pp. 61f.) quotes versions of the same saying from Clement of
Alexandria, Theodoret, and Epiphanius.
Η. P. Owen (NTS 5 (1958), 133-43 (cf. Hanson 178)), rightly
observes that Luke is not presenting the Hellenistic-Jewish argument
that God can be known from creation; he is known only by the word,
or Gospel; God’s creatorship is known by faith.

25. The form of this verse is parallel to that of v. 24, with διδούς
occupying the place of υπάρχων (since he is ... since he gives ...).
θεραπεύειν is often in the NT (and elsewhere) used of the service
done by men to men, especially in the healing of diseases. It is
ridiculous to suppose that God needs service by human hands (ὑπό
χειρῶν ανθρωπίνων; cf. χειροποίητος in ν. 24). There is an implied
 45. PAUL AT ATHENS. 17.16-34. 841

contrast between ανθρώπινος and θεῖος; cf. an Athenian inscription

of AD 174/175 quoted in ND 4.83f. (περί τε τά θεία καί τά
ανθρώπινα). He made everything; how can he need anything? He
not only created all things in the beginning but continues to give
(διδούς, present participle) to all life, breath, and all things, that is,
all the things men need for their human existence. It is scarcely
possible to distinguish between ζωή and πνοή (take away men’s
breath and they die), and τά πάντα is deliberately imprecise. In Gen.
2.7 when God formed man of the dust of the earth he breathed into
his face πνοήν ζωής; as a result man became ψυχῂ ζῶσα. Cf. 2
Macc. 7.22, 23.
For τινος, αὐτός διδούς, D(c) has ὅτι οὗτος ὁ δούς, for it is he who
gave. The absence of τινος makes no difference to the sense; the use
of the aorist participle points to the act of creation.
There are many parallels to this verse, biblical, Jewish, non-
Jewish. Isa. 42.5 is close; it is interesting to note the differences.
Isaiah’s πνεῦμα is omitted, because the Spirit is given not to all but
to believers; τω λαῷ, which suggests the chosen people, is changed
into all; all things is added. Cf. Ps. 50.12f.; 2 Macc. 14.35; Tobit
7.17; 3 Macc. 2.9; Mt. 11.25 = Lk. 10.21.
Among other Jewish writings: Josephus, Ant. 8.111, ἀπροσδεές
γάρ τό θεῖον απάντων (it is interesting to note the Stoic treatment of
Solomon); among non-Jewish texts Calvin notes Persius, Satire
2.69-71. See also Euripides, Hercules Furens 1345f., δεῖται γάρ ό
θεός, εἴπερ ἐστ’ ὀρθῶς [? όντως] θεός, οὐδενός; Seneca, Epistle
95.47, non quaerit ministros deus, quidni? ipse humano generi minis-
trat... ; Corpus Hermeticum 5.10, πάντα δέ ἐν σοί, πάντα ἀπό σου,
πάντα δίδως καί οὐδὲν λαμβάνεις. Cf. Corpus Hermeticum 2.16.
It is interesting to note that Luther (perhaps remembering Seneca)
takes τίνος to be masculine (ymands). Chrysostom (Homily 38.2)
takes up a fresh (and probably unintended) point λέγων δέ, μή υπό
χειρῶν ανθρωπίνων θεραπεύεσθαι τόν θεόν, αἰνίττεται ὅτι
διανοία καί νῷ θεραπεύεται.

26. ἐποίησεν resumes the ποιήσας of ν. 24, a fact which probably

determines the way in which the word is to be taken (Haenchen 502).
It may be taken either as an equivalent of created (as ἐποίησεν
translates in Gen 1.1) or understood modally with κατοικεῖv, or
with ζητεῖν (v. 27), or with both. The former view is taken by
Dibelius (35f.), the latter by Pohlenz (ZNW 42 (1949), 84f.); there is
a reply by Dibelius (153, n. 37). The occurrence of ἐξ ἐνός as an
adverbial description of the circumstances of creation seems decisive
in favour of Dibelius’ view; κατοικεῖν and ζητεῖν follow epexeget-
ically. God made them of one (on this see below), with a view to their
dwelling ... seeking. It must be emphasized that the whole sentence,
comprising vv. 26, 27, is a unit of which the various parts belong
842 COMMENTARY ON ACTS

together; one consequence of this is that the method of making (ἐξ

ἐνός) must be seen in relation to the dwelling and seeking.
After ενός (P74 A B 33 81 323 1175 1739 pc vg co Clement),
αίματος is added by D E <a> gig sy. It has been maintained that this
longer way of expressing the original unity of mankind was changed
and αἵματος dropped because it was recalled that God made men not
out of blood but out of the dust of the earth (Gen. 2.7; Preuschen
109). More probably the short text was original and referred to the
creation of the one man, Adam, the father of all; there was no clear
parallel to this in Greek thought and mythology, and blood was
added as a different way of expressing the unity of all races. Instead
of αἵματος, Ψ has στόματος; if this is not a simple negligent error
(homoeoteleuton) it must refer to mankind before the confusion of
languages at Babel (Gen. 11.1 –9; note in 11.9, ἐπί πρόσωπον πάσης
τῆζ γῆς and the words later in the present verse).
Beginning from unity God made πᾶν ἔθνος ανθρώπων. ἐνός and
παν are juxtaposed in sharp contrast. Again there is a measure of
ambiguity in the Greek: every race of men, or the whole race of men.
The absence of an article before πᾶν might suggest the former, but is
in fact not decisive; cf. in this verse ἐπὶ πόντος προσώπου τής γῆς
(on the text see below). The meaning is in fact determined by the
words that follow. See Dibelius 29-34, and below; on the grammat-
ical point, Moule (IB 94f.); Zerwick (§ 191).
The infinitive κατοικεῖν is epexegetical rather than strictly pur-
posive, and πόντος προσώπου recalls πᾶν έθνος. The human race
as a whole (based on its unitary origin) is to occupy the whole earth.
Of ἐπί προσώπου Μ. 2.466, noting Jer. 32.12 (LXX; = Hebrew
25.26) rightly says, ‘the influence of the Greek of the LXX is
unmistakable’. Only in this sense is the expression a Semitism, but it
is significant that Luke should (in this Areopagus speech) use the
language of the Bible. His criticism of popular polytheism and
idolatry is at once philosophical and biblical, and his statement of
God’s intention in creation also is twofold, though the use of ποιεῖv,
of ἐξ ἐνός, and of πρόσωπον is biblical. Instead of πόντος
προσώπου, E Ψ <a> have πᾶν τό πρόσωπον, either an accidental
error, or a correction.
The next clause (cf. Corpus Hermeticum 5.4) is introduced by the
aorist participle όρίσας. This has been taken (Rackham 184) as an
example of an aorist participle of subsequent action, but Μ. 1.133
rightly observes that the truth is the opposite of this: ‘the determina-
tion of man’s home preceded his creation, in the Divine plan’;
similarly Bruce (1.337). God determined προστεταγμένους (D* pc
bo have προτεταγμένους, with no significant difference—if the
times were appointed at all they were appointed in advance) καιρούς
καί τάς οροθεσίας (D* Irlat have κατά τήν οροθεσίαν) τής
κατοικίας αἐἀυτών. The last words simply take up the verb κατοικε tv,
 45. PAUL AT ATHENS. 17.16-34. 843

but the meaning of καιρούς and οροθεσίας is disputed, with

Dibelius and Pohlenz again taking opposite sides. The main possibil-
ities are:

1 (a) God has ordained the various areas in which the races live, and
the periods in history of their dominance.
(b) As 1(a) but the periods are those not of history but of
apocalyptic; that is, they belong to the future.
2 The areas are the different zones of the earth, and the καιροί are
the seasons of the year.

Pohlenz (87f.), who took πᾶν ἔθνος to refer to the various races of
men, takes interpretation 1(a); they have their assigned areas (cf.
Deut. 32.8) and God controls their ‘zeitliche und räumliche Entwick-
lung’ (87). Dibelius (29-32), who took πᾶν ἔθνος to refer to the
whole human race, observes that the readers or hearers of the speech
would not have the historical knowledge to appreciate 1(a), and
argues that 14.17 points to the meaning seasons for καιροί. The
ὁροθεσίαι refer to the zones of the earth, of which only two, the
temperate zones, were fit for and assigned to human habitation. As
for 1(b), Dibelius points out that the passage is not apocalyptic-
eschatological.
Passages from the Qumran literature have been invoked on either
side. In favour of 1(a), 1QH 1.16f.: In their times thou hast
appointed their service for all their generations and
judgement for their appointed times but read In
favour of 2, 1QM 10.12-16: He who created the earth and the laws
of its divisions for desert and steppe and all that it brings
forth with [...], the circle of the seas and the containers of the
rivers and the divisions of the primal floods ... the divisions
of the peoples and the dwelling of the tribes and the
inheritance of the lands ... sacred festivals and the
cycles of the years and eternal times See also
1QM 2.6-15; 1QS 1.14, 15. Braun (ThR 29 (1963), 171f.) observes
that since in Qumran times may refer to both Geschichtsabschnitte
and Feste it may be a mistake to regard the different interpretations
of Acts as mutually exclusive.
The parallel with 14.17 suggests that καιροί refers to seasons of
the year, though as Hanson(179f.) emphasizes, the classical word for
seasons is not καιροί but ὧραι; on this however see Ehester (NTS 3
(1957), 100f.) who agrees with Dibelius and sees allusions to Gen.
1.14,9. Begs. 4.216 argues that Lk. 21.24 and Dan. 8.10 [this must be
a mistake; it seems possible though not certain that 8.19 is intended]
point to periods rather than seasons, but also notes lines 7-9 of
Aratus (quoted in v. 28): ‘He telleth it, when the clod is best for oxen
and for mattocks; he telleth it when the seasons are favourable.’
844 COMMENTARY ON ACTS

Corpus Hermeticum 5.4 may also be cited and Roloff (362) refers to
Ps. 74(73). 12-17; Jer. 31.35. The parallel with ch. 14 would be
important only if both speeches were composed by the same speaker
(Luke?). In any case, the point made is that all the affairs of men and
nations are in the hand of God.
Schmithals (163) draws attention to similar thoughts expressed at
about the same time as Acts: 1 Clement 19.3; 20.6, 9; cf. Slavonic
Enoch 19.4; Apostolic Constitutions 8.12.

27. The human race was disposed in areas of the earth’s surface,
and under climatic conditions, calculated to make human life
possible; but physical existence was not the final purpose for which
men were made. They were intended ζητεῖν τον θεόν, to seek God,
whom (it is implied) they would know only if they sought him; the
search itself had value and was willed by God. ζητεῖν τόν κύριον (E
<a>, if κύριον refers to Christ, misses the point; Christ is the self-
motivated self-revelation of God, which is not yet being considered.
Seeking God is a theme of the OT: among many examples the
following will suffice. Isa. 51.1, οἱ διώκοντες τὀ δίκαιον καί
ζητοῦντες τόν κύριον 55.6, ζητήσατε τόν κύριον (ν.1.,
θεόν) Dibelius rightly observes (32) that in the OT
seeking the Lord is ‘a matter of the will’, but it is perhaps not so
exclusively so as he suggests. As with γινώσκειν (see R. Bultmann,
TWNT 1.688-719, especially 697; J. Bergmann and G. J. Botter-
weck, TWAT 3.479-512, especially 500-510) there is also an
intellectual element, and, under Greek influence, this is brought out
in Hellenistic Judaism (as it is here). Thus e.g. Philo, De Specialibus
Legibus 1.36, ἄμεινον γὰρ ούδέν τού ζητεῖν τόν αληθή θεόν, κἃν ή
εὕρεσις αὐτοῦ διαφεύγη δύναμιν ἀνθρωπίνην. In Wisd. 13.6 there
is an unfavourable view of the Gentile search for God, but the
language is strikingly akin to Paul’s (Luke’s): θεόν ζητοῦντες καί
θέλοντες εύρεϊν.
Paul (Luke) does not regard the search for God as an impossible
one, though he does not represent its successful conclusion as
certain. Confidence in the possibility of successful seeking is
expressed at the end of the next clause by the use of the verb
εὑρίσκειν, uncertainty by the form of the clause, introduced by εἰ
ἄρα γε and containing verbs, ψηλαφήσειαν and εὑροιεν, in the
optative mood. Μ. 3.127 writes that ‘the other instances [17.27;
27.12] of εἰ c. opt. are not so much real conditions as final clauses’,
but the very fact that they are expressed in this way shows that the
purpose of the final clauses will not be unconditionally achieved.
This use of the optative is in the NT peculiar to Acts. The forms of
the verbs should be noted. D has in each case the ending –σαν:
ψηλαφήσαισαν and εὕροισαν. This form is common the LXX
(Thackeray (215) gives many examples); Μ. 2.196 (cf. 211) thinks
 45. PAUL AT ATHENS. 17.16-34. 845

that the occurrences in D should probably be linked with those in the

LXX, but this does not explain why the form should occur only in D.
See also Μ. 1.56.
As the infinitive is modified by the uncertainty of εἰ and the
optative, so the uncertainty is modified by the next clause, where
again there is textual variation, καί γε is the reading of B D Ψ 0120
<a>; καίτοι is read by P74 A E 945 1739 1891 pc Clement, καίτοι γε
by 323 pc. καίτοι (γε) followed by the participle υπάρχοντα is
straightforward: though he is not far ... This is closely parallel to
14.17, in construction and in sense. καί however can equally well be
used in a concessive sense, though other meanings could not be
excluded, γε adds little (see BDR § 439.2, n. 3); according to
Rademacher (29) it was added to avoid the clash between αι and the
following ου.
God is near; notwithstanding Philo (above) the search is not
hopeless. Cf. Josephus, Ant. 8.108: ουδέ ... απολείπεις του πᾶσιν
ἔγγιστα εἶναι. (Solomon). God’s nearness is not that intended in
Acta Andreae et Matthiae 26 (L.-B. 2.1.103): ἐγώ γἀρ γινώσκω,
κύριε, δτι ὅύκ εἶ μακράν από τών σῶν δούλων), which, though the
author may have had our verse in mind, refers to God’s nearness to
his own people who acknowledge and serve him. The present verse
refers to God’s nearness to men as such, to all men. This was a Stoic
belief; cf. e.g. Seneca, Epistle 41.1: prope est a te deus, tecum est,
intus est... 2: in unoquoque virorum bonorum ... habitat deus. The
OT is nearer to Andreae et Matthiae; e.g. Ps. 144(145).18, ἐγγὐς
κύριος πᾶσιν τοῖς ἐπικαλουμένοις αυτόν ... ἐν ἀληθεία; cf. Deut.
4.7.
The primary sense of ψηλαφᾶν is well illustrated by Homer,
Odyssey 9.416, where the blinded Cyclops is described as χερσἰ
ψηλαφόων, as he tried to lay hands on Odysseus and his men. That
he was unsuccessful was due to the ruse employed by his prisoners.
Cf. Plato, Phaedo 99b, δ δή μοι φαίνονται ψηλαφῶντες οι πολλοί
ὥσπερ ἐν σκότῳ ... This leads to the transferred sense of the word
(which does not imply a hopeless quest); see Philo, De Mutatione
Nominum 126, an allusion to Moses as ψηλαφῶντος ... καί διά
χειρός ἔχοντος αεί τά θεῖα. See Gärtner, 156-61, who however
emphasizes too much the difficulty of the quest and the unlikelihood
of success. Luke does represent the outcome as uncertain, but (a)
God intends that men should seek and, presumably, find, and (b) he is
not far from us. The Western text, which at the beginning of the verse
had τὸ θεῖον instead of τόν θεόν, continues with the neuter gender.
For αὐτόν, D* (gig) Irlat have αυτό; and for από ἐνός έκαστου ἡμῶν
υπάρχοντα, D has ὂν (neuter participle, agreeing with θεῖον) ἀφ’
ἐνός ἑκάστου ἡμῶν. In the opening clause D has not only τό θειον
but an unwanted έστιν. This is probably due to the influence of the
Latin, quaerere quod divinum est.
846 COMMENTARY ON ACTS

Commentators differ in their understanding of what Luke means by

this verse, Calvin, who comes near to rewriting the address, says
(2.119), unconvincingly, Of course Paul is not speaking here about
the ability of men, but he is only warning that they are inexcusable,
when they are blind in such a clear light.’ Bengel (459) is better: ‘Via
patet; Deus inveniri paratus est: sed hominem non cogit. Ita liberum
eum esse voluit, ut, quum homo Deum quaerit et invenit, id respectu
DEI sit quodammodo quasi contingens quiddam.’ Modern writers
note the difference between seeking and knowing God in the OT and
with the philosophers; e.g. Roloff (262f.): in Greek philosophy seek-
ing is rational (Plato, Apology 19b; 23b; Gorgias 457d); in the OT it is
wider and more existential (Amos 8.12, Ps. 14(13).2; 53(52).3; Rom.
3.11; 10.20; et al.). Given the distinction, opinions differ. Thus Stählin
235: The seeking required is not ‘ein verstandesmässiges Aufsuchen
und wissenschaftliches Untersuchen wie in der Philosophie der neu-
testamentlichen Zeit, sondern eine Sache des Willens, der Dankbar-
keit und der Ehrfurcht, das Verlangen des ganzen Menschen nach
Gott...’. On the other hand, Conzelmann (100): ‘Das Suchen ist hier
nicht wie im AT (Deut. 4.29; Amos 5.6: Jes. 55.6 usw) Sache des
Willens, sondern des Erkennens (Dibelius Aufs. 33ff. [= ET 32]);
doch fehlt die systematische Reflexion, zB die Unterscheidung der
Fragestellung εἰ ἔστι τό θεῖον und τί ἐστι κατ’ ουσίαν (Philo, Spec
Leg 1.32; ähnlich Cicero De Natura Deorum 2.4(12)).’ This seems not
far from being a contradiction. Weiser (472) sees in Luke’s words a
combination of the two lines of thought: ‘Wie schon mehrfach in der
Rede, so verbindet Lukas mit dem Wort “suchen” (zētein) biblische
und griechisch-philosophische Inhalte.’ There is something like this, a
small rapprochement of two ways of thinking, in Seneca, Epistle
95.47, Deum colit, qui novit. Cf. Cicero, De Natura Deorum
1.61(153), Cognitio dei, e qua oritur pietas [So Weiser, but Cicero
wrote ‘... cognitionem deorum...’.].
But we do well to ask whether Luke had seen as clearly as modern
students the difference between the biblical and the philosophical
search. All used (more or less) the same words; must they not mean
the same? To analyse the distinctions too sharply may mean missing
Luke’s point.

28. This verse is intended to supply the basis (γάρ) for the
statement in v. 27, or at least for the last clause in that verse. God is
not far from each one of us, for it is in fact in him that we exist. So
Chrysostom (Homily 38.3): τί λέγω μακράν; οὕτως ἐγγύς έστιν, ὡς
χωρίς αυτού μῆ ζῆν. ἐν αὐτω γάρ ζῶμεν κτλ... . καἰ οὐκ εἶπε, δι’
αυτού, άλλ’ δ ἐγγύτερον ἦν, ἐν αἐτω. This is put in a threefold
proposition, using the three verbs, ζῶμεν, κινούμεθα, ἐσμέν. Gärt-
ner’s conclusion (195) seems to be correct: ‘The triad ... is used to
bring out all sides of man’s absolute dependence on God for life. It is
 45. PAUL AT ATHENS. 17.16-34. 847

not a veiled poetic quotation, but a combination that must be ascribed

to Paul or Luke.’ The positive part of this conclusion rests upon the
negative, which excludes other possibilities, and on the observation
that though the Stoics connected life with movement (the Prime
Mover being God) and movement with being (‘κίνησις transforms
that which only has ἕξις into οὐσία’) they did not put the three
together. This is true—as far as we know; but it is somewhat
precarious to suggest that no Stoic ever combined the three. We may
say that Paul (Luke) was responsible for the use of the triad in the
speech, but may have heard it used and borrowed it. Paul omits the
preposition ἐν at Rom. 11.36. On it Begs. 4.217 says, ‘The ἐν is an
obvious example of the meaning “in the power of’: cf. Sophocles,
Oedipus Coloneus 1443, ταῦτα δ’ἐν τῷ δαίμονι, and other examples
given by Liddell and Scott.’ Begs. translates, ‘By him we live and
move and are.’ It is true that the reference to the poets may refer
backwards rather than (or as well as) forwards, and it has been
maintained (see Begs. 5.246-51; but see Williams 205; Bruce 2.339,
n. 75) that the triad is a quotation from Epimenides. This view
however finds its strongest support in a passage of Isho‘dad of Merv,
and it has been shown by Pohlenz (ZNW 42 (1949), 101-4, an
appendix to his article ‘Paulus und die Stoa’) that the passage in
Isho‘dad is not to be used in this way. H. Hommel (two articles, ZNW
46 (1955), 145-78; 48 (1957), 193-200; for this expression espe-
cially the latter; also Schneider 2.241) makes Posidonius the starting-
point for this sentence, though there is a background in Plato. The
conclusion is best summed up in Hommel’s own words. ‘Also
dürfen wir jetzt wohl auch unsere ZNW 46.165ff. an Hand zahl-
reicher Belege auf den platonisierenden Poseidonios zurückgeführte
Stelle Act 17.28a ἐν αὐτῷ γὰρ ζῶμεν καὶ κινούμεθα καί ἐσμέν,
indem wir das dort gefundene Ergebnis abrunden, so erklären, dass
in ζῶμεν das physische, in ἐσμέν dagegen das seelische-geistige
Leben steckt, während das Dritte, das κινούμεθα, beides ins
Kosmische überhöht, d.h. solche leiblich-geistige Existenz als
Abbild der himmlischen περίοδοι zu begreifen lehrt. Dass Pos-
eidonios die ersten beiden Glieder—Leben und Bewegung—einfach
aus dem “Timaios” zu übernehmen brauchte, ist seinerzeit von uns
ausführlich gezeigt worden. Nun sehen wir, dass er aus intimer
Kenntnis des ganzen Platon auch noch das im Sinn des geistigen
Seins zugefügt hat, was die Reihenfolge der Glieder erklären mag’
(198f.).
So far this says more about Posidonius than about Paul and Luke.
In Luke’s mind, ἐν αύτῷ must be understood in terms of v. 27c, ού
μακράν ... God is not remote but accessible, so near as to constitute
the environment in which we live, but in a personal sense. In
the Greek philosophical background the words will have had a
pantheistic meaning, God being hardly anything other than our
848 COMMENTARY ON ACTS

environment. The change is likely to have been made already in

Jewish-Hellenistic use.
The same observation will apply to the quotation that follows. It is
ascribed to τινες ... ποιητών. This may be a device for concealing
ignorance of the actual author, or for suggesting that the words wre
used (as in fact they were) by more than one (Begs. 4.218). Dibelius
50,51 (n. 76) contradicts the argument that ‘it was not in accordance
with literary convention that the word τινες should be used to give a
veiled reference to one only’. He cites however only two passages
from Philo (De Specialibus Legibus 1.48, 74); Pohlenz (104, n. 77)
adds Pseudo-Aristotle, De Mundo 397b, 16 [? 6]. Three passages
hardly amount to a convention (Conzelmann 101 also speaks of
a ‘literarische Konvention’), but they are enough to indicate a
possibility.
καθ’ υμάς is a little more than equivalent to a possessive pronoun
(BA 828). In 18.15 νόμου τοῦ καθ’ ὐμᾶς is a little more than ‘the
law you happen to have’; it means the law that you regard as
authoritative. Here as a translation ‘Your own poets’ suffices, with
the sense, Poets that you ought to be prepared to listen to. For καθ’
ὑμᾶς, P74 B 049 326 614 pc have καθ’ ἡμᾶς, which sets Paul among
‘us Greeks’; this is in the context unlikely, even though Aratus was
like Paul a Cilician—‘us Cilicians’. η and υ were often confused.
καθ’ ὑμᾶς can hardly be taken as ‘poets of your own (Stoic) party’
(so Blass 192).
The words quoted are found in Aratus, Phaenomena 5, as was
already observed by Clement of Alexandria (Stromateis 1.19). There
is a similar half line in the Hymn to Zeus of Cleanthes: ἐκ σοῦ γάρ
γένος εἴσι (SVF 1.537 (line 4), p. 121). It is possible (see above) that
more than one source is in mind, but the quotation is definitely from
Aratus. For him the words are pantheistic (1.4: πάντη δὲ Διὸς
κεχρήμεθα πάντες), and to be distinguished from such passages as
Vergil, Aeneid 1.250 (Nos, tua progenies), which are mythical. Luke
(Paul) uses the quotation, as the context (γένος υπάρχοντες, ν. 29)
shows, in the attack on idolatry which has continued, more or less,
from the beginning of the speech. The words are of course in
themselves capable of a purely Christian meaning: through the Holy
Spirit we are born again as children of God (Jn 3.5). But they lose
their point here if they are not used in a sense different from this: Not
the regenerate but human beings as such are the children of God—
you cannot deny it for we have it on the authority of your own
poets.
Between καὶ ἐσμέν and εἰρήκασιν, D (gig) Irlat have το καθ’
ημέραν, ώσπερ καί τών καθ’ ὑμᾶς τινές. Daily adds little to living,
moving, and being, and there is little to be said for the suggestion that
it is a misplaced marginal note (from v. 11). There seems to be little
point in omitting poets; D may perhaps be right here. ‘It would be
 45. PAUL AT ATHENS. 17.16-34. 849

difficult to find a more typical example of a gloss than the addition of

ποιητῶν’ (Clark 367). This however is not the only variation, and it
is probably correct to observe that the reading of D reflects a period
of freedom in regard to the transmission of the text; the general sense
sufficed. For the scansion of the quoted line see BDR § 487.1.
τοῦ is worth noting as the only example in the NT of the use of the
article as a demonstrative pronoun (otherwise than with μέν and
δέ)—unless in Rev. 19.20 τῆς stands for ταύτης or αυτής (BDR §
249; § 423.4, n. 10).
Here as elsewhere in the speech it is easy to interpret the material
one-sidedly. The Greek side is unmistakable; a Greek poet is quoted,
and the thought is not only Stoic but pre-Stoic; see e.g. Orphic
Fragment 164:
έστιν δὴ πάντων αρχή Ζεύς. ζήν γάρ ἔδωκεν
ζῷά τ’ ἐγέννησεν, καἰ Ζῆν’ αυτόν καλέουσιν
και Δία τῆδ’, ὅτι δὴ διά τούτον άπαντα τέτυκται.
εις δέ πατήρ οὖτος πάντων, Θηρών τε βροτῶν τε.
Stählin (236) takes it that Luke is thinking of the creation of man in
the image of God, as in the OT. He continues: ‘Die Gottesver-
wandschaft des Menschen als Geschöpf ist aber etwas anders als die
Gotteskindschaft aller Menschen; diese kennt das Neue Testament
nicht (ausser vielleicht der Epheserbrief, vgl 4.6; 3.15).’ See also
Bauemfeind (219): The first part of the speech ends at vv. 28,29, and
the end corresponds with the beginning. ‘Dort: In allem Heidentum
dennoch die Altarinschrift, die über das Heidentum hinausfuhren
sollte; hier: ein heidnisches Dichterwort (v. 28), dessen Konsequenz
im Grunde auch Sprengung der heidnischen Abgötterei ist.’

29. The new sentence follows verbally and argumentatively upon

the quotation from Aratus in v. 28: γένος οὖν υπάρχοντες—since
we are, as has just been stated, God’s offspring. The argument runs
back from men to God: since we are the thinking and feeling persons
that we are, we ought not to suppose that the divine being (τό θειον,
rather than τόν θεόν; on this see below) is made of metal, even
precious metal, or of wood. Luke might have balanced θειον with
ὰνθρώπειον. If human nature is what we know it to be, and if we
who have human nature are God’s children, the divine nature will be
of no lower order. We deny our own proper being if we identify our
progenitor with material objects.
For χρυσῷ, P41 P74 A E 104 326 pc have χρυσίῳ; for ἀργύρω,
P41 P74 A E 36 104 453 pc have ἀργυρίφ. The forms in –ἱov are much
more common in the OT.
χάραγμα is a carved object, and the genitives τεχνῆς and ἐνθυμ-
ήσεως are subjective, an object carved by man’s art and imagination.
One would expect imagination, conception, to precede art, the
850 COMMENTARY ON ACTS

realization of the concept (Bengel 459: artis extemae, cogitationis

intemae), and this might add weight to the omission of καί ἐνθυμ-
ήσεως ανθρώπου (Ρ74), but the attestation is too slight to be credible
since there is no serious problem, only at most inappropriateness, in
the word order.
Idolatry was condemned by Jews and by Christians, also by some
pagan philosophers, though, as Gärtner (224) says, ‘The polemic is
not often levied direct at the images, since the general opinion
obviously was that an identification of the image and the god was too
absurd to require treatment.’ The prophetic attack on image-worship
is sufficiently familiar (Isa. 40.18-20; 41.5-7, 29; 42.17; 44.9-20;
46.1-7; 48.5; 57.13; Jer. 1.16; 2.26-28; 10.2-5, 8,9); it was echoed
in Judaism of all kinds. It suffices to mention the Mishnah tractate
Abodah Zarah (though here we have regulations for the avoidance of
idolatry rather than polemic against idolatry; see however 3.4).
Wisdom takes up the polemic in Greek; see chs. 13 and 14
(especially 13.10-19); so also Philo, De Vita Contemplativa 7 (τὰ
ξόανα καί αγάλματα); De Decalogo 66 (τών ξύλα καί λίθους
άργυρόν τε καί χρυσόν καί τὰς παραπλήσἱους ὕλας
μορφωσάντων); De Specialibus Legibus 1.21f. Among Gentile
writers see for example Seneca, Epistle 31.11, ‘Finges autem non
auro vel argento; non potest ex hac materia imago Deo exprimi
similis’; Lucretius 1.63-80; Plutarch, De Superstitione 6 (167DEF).
For the neuter τὸ θεῖον (ND 3.68 supplements examples in MM
285f.) cf. ν. 23 (ὃ ... τούτο) and v. 27 (D). Here it is less remarkable.
We have just heard of the relation between God and men, so that
there is no question of the personality of God; τό θεῖον refers to the
property of being divine, divinity contrasted with an implicit τό
ἀνθρώπειον, that which is human, and τό υλικόν, that which is
material.

30. The sentence is introduced by μὲν οὖν; for this characteristic

expression see on 1.6. There is no answering δέ; according to
Knowling (377) this is because the opening clause has a participle,
ὑπεριδών, not a finite verb; we may note also the adversative force
of τά νυν.
It is not quite correct to say, as Conzelmann (101) does, that the
more specifically Christian material is now introduced into the
speech without any transition. Verse 29 complains of the ignorant
error of supposing the Divine to be identifiable with material objects.
To make this identification—and according to Luke the Athenians
were doing it all over the city (v. 16)—was to act in ignorance,
άγνοια. This takes up the ἀγνοούντες of v. 23. This ignorance,
which perverts the ευσέβεια that accompanies it into δεισιδαιμονία,
has been going on for a long time; the story of Athens is a record of
χρόνοι τῆς άγνοιας. From nature the Greeks have evolved not
 45. PAUL AT ATHENS. 17.16-34. 851

natural theology but natural idolatry. That this should have been
permitted was a mark of God’s forbearance (cf. 14.16; also and
especially Rom. 3.26). God did not will or approve this ignorant
idolatrous worship, but he did not suppress it; he overlooked it,
ὑπεριδών. ὑπερορᾶν is occasionally used literally (BA 1867 quote
Herodotus 7.326.5, ... ἵνα μή φοβῆται τά υποζύγια τήν θάλασσαν
ὑπερορώντα), more often metaphorically, sometimes in the sense of
looking down upon a person, despising him (Herodotus 5.69.1,
ὑπεριδών 'Ιωνας), sometimes, as here, in the sense of looking down
and not at, thus ignoring, overlooking (e.g. Aeschines, 1 (= Adv.
Timarchum) 116, τήν ὕβριν ... ὑπερεώρακε; also Gen. 42.21; Deut.
21.1; Ps. 55.1 (54.2); 78(77).62).
It was not God’s intention that men should continue permanently
in this ignorance of his true being and worship in ignorant idolatry.
Now (τά νῦν simply = νυν; so LS (1185), referring to Herodotus
7.104.2; Euripides, Heraclidae 641) he is taking steps to end this
situation. It is ὁ τής μετανοιας καιρός (Clementine Homilies 1.7.6).
He does this with a command (παραγγέλλει: 14 (15) times in Luke-
Acts out of 30 in the whole NT, but the word is too common to be
regarded as a Lucan characteristic) that all everywhere should repent.
Emphasis is given to the command by the paronomasia of πάντες
πανταχοῦ. 21.28 is similar (πάντας πανταχῆ) and there is another
example in the next verse (πίστιν παρασχὠν πᾶσιν). These suggest
original composition in Greek, with some rhetorical refinement.
The summons to repentance may be regarded as parallel to the
demonstration of God’s righteousness in Rom. 3.26; in each case
there is an implicit charge that God’s righteousness is impugned;
does he not care what his creatures do? See Romans 75f. There is
however a marked difference in that in Romans God takes the
initiative by revealing his saving righteousness whereas in Acts man
must initiate the process by repentance. It is only in this sense that
Pesch (2.132) is right in saying that the present verse corresponds to
the good news (εὐαγγελίζεσθαι) of Jesus and the resurrection (v.
18). On repentance see on 2.38. Here it is clear that repentance will
mean in the first instance turning from the false gods with which
Athens abounds. It is also true however that since the call is for
repentance the defect of Greek religion is not simply intellectual but
existential. Man is guilty of having withdrawn from fellowship with
the Creator. Cf. Weiser, who points out (475) ‘dass die heidnische
Haltung nicht nur als ein Erkenntnismangel, sondern auch als eine
zum Teil schuldhafte Verweigerung und religiöse Verirrung angese-
hen werden muss’. Cf. also 479f. ‘ ... die Unkenntnis wird nicht
durch Erkenntnis, sondern durch existentielle Umkehr aufgehoben
(ν. 30)’.
For ὑπεριδών, D(c) vg have ταύτης παριδών, which Metzger
(458) (against Epp 48-50) regards as an ‘innocent heightening’.
852 COMMENTARY ON ACTS

Haenchen (505), following Zahn, refers to Sirach 28.7, πάριδε

άγνοιαν. See also ND 1.62.
For παραγγέλλει, in P41 P74 (c) A D E Ψ <a> syh Ath Cyr, B have
απαγγέλλει, a weaker verb.

31. The requirement of repentance is supported by the sanction of

judgement. καθότι, is a Lucan word (Lk. twice; Acts 4 times;
nowhere else in the NT), inasmuch as (= καθ’ δ τι). ἱστάναι is said
not to be used for fixing a day for a special purpose (but see
Dionysius of Halicarnassus 6.48.3, ἡμέραν έστησαν); it is in any
case readily understandable and there is no need to seek any other
translation. It is however not impossible that, through its association
with the day ofjudgement, ήμερα had come to bear occasionally (as
at 1 Cor. 4.3) the meaning court, institution for dispensing justice.
God has established an agency of justice. But day is better. It is
implied that the day is near, otherwise the warning would carry little
force.
ἐν ἦ μέλλει is omitted by D Irlat Speculum (partly affected by the
Latin (iudicaturus est); this leaves κρίνειν as a quasi-final infinitive
(Μ. 1.240f.): God has appointed a day for judging. There is little or
no difference in meaning; again there seems to be a readiness to vary
the wording of the text which suggests something less than a belief in
its canonical inspiration.
οικουμένη also is Lucan (Lk. 3 times; Acts, 5 times; rest of the
NT, 7 times). For God’s judging the οικουμένη see Ps. 9.9; 95.13;
97.9 (LXX). Nothing different is intended from judging the living
and the dead (10.42—also with ὁρίζειν). ἐν δικαιοσύνη is equiva-
lent to δικαίως: a sort of instrumental ἐν. Cf. (for Lucan style) 5.23,
ἐν πάση άσφαλείᾳ, and see LS (552), s.v. ἐν II 3.
Not only has God determined judgement day (ἔστησεν), he has
appointed (ὥρισεν, cf. 10.42) the judge. The use of ἐν to introduce
the agent of judgement is classical according to BDR § 219.1, n. 1,
but this is hardly proved by the third-century Delphi inscription
(Dittenberger, Syll. 850.8, κριθέντω ἐν ἄνδροις τρίοις [sic; this is
oddly described as classical!]) alluded to by BDR and quoted by Μ.
1.107. The only example of ἐν as introducing a personal instrument
given by LS (552) is Mt. 9.34. MM 209 cite only the inscription
given above (dating it 173-172 BC). BA 525 add Synesius, Epistle
91, p. 231B [in MPG the reference appears to be 90,231A]. Bruce
1.340 quotes 1 Cor. 6.2 and Dittenberger, Syll. 147.57f. (4th century
BC), κρινέσθω ἐν Άθην[αίο]ις καί τ[οῖς] συμμάχοις.
It is of course clear that the Judge whom Paul (Luke) proclaims,
here in what had been the court of the Erinyes (Stählin 238), is Jesus.
He is however simply introduced in the words ἐν ἀνδρί. For ἐν
ἀνδρί, D Irlat have ἀνδρὶ ’Ιησοῦ. The omission of ἐν is probably
accidental; it was thought well to identify the judge explicitly. But if
 45. PAUL AT ATHENS. 17.16-34. 853

’Ιησοῦ had been original it would hardly have been so widely

omitted. All that the absence of the name means is that, at this stage,
the speaker is more interested in the theme of judgement than in the
details of the process. The next clause effects the identification—for
the reader. Luke has not forgotten that Jesus is the man who is also
Lord and Christ. (2.22, 36). Begs. 4.219 writes; ‘This is “Son of
man” eschatology, and if the custom of the gospels had been
followed the underlying “bar nasha” would have been rendered by
υἱός τού ανθρώπου instead of by ἀνδρί.’ This is important and
worthy of consideration, but it is too simple. Where—in Athens—
was the underlying Aramaic?
That one who has been a man should be exalted to the role of
universal judge is unheard of and needs proof. God provided this by
raising him from the dead. It is evidently implied that the man had
died, but nothing is said of the manner or significance of his death.
The proof was provided for all; for the paronomasia of πίστιν
παρασχών πᾶσιν see ν. 30 and the note. Nowhere else in the NT
does πίστις have the meaning proof. The meaning appears regularly
in Aristotle (where πίστις is distinguished from ἀπόδειξις, a demon-
strative proof—Rhetoric 3.13 [1414ab]) and elsewhere (e.g. Plato,
Phaedo 70b), though it is not the most common meaning of the word.
At Josephus, Ant. 15.260 it is combined as here with παρέχειν. The
suggestion that the word does have its usual NT meaning, and that
the clause means that, in the resurrection, God was offering to all (the
opportunity of) faith, is unconvincing. Bornkamm (4.158) is right
therefore to conclude, ‘Von einem Zugleich von Anknüpfung und
Widerspruch lässt sich hier streng genommen nicht mehr reden,
während es für Paulus selbst allerdings zu behaupten ist. Die
Areopagrede als solche intendiert vielmehr eine Anknüpfung ohne
Widerspruch [dem entspricht die Einführung der Auferstehung Jesu
als “Beweis” (πίστιν παρέχειν)]—nämlich an die religiösen Vor-
aussetzungen ihrer Hörer, um gerade so deren faktischen Wider-
spruch als Verstocktheit erscheinen zu lassen.’
For παρασχών, D gig have according to ΝΑ26 παρεσχειν, accord-
ing to Ropes (Begs. 3.171) παρασχεῖν. This is presumably Ropes’s
correction, his conjecture of what D meant (so Dibelius 62). WW
(3.156) have παρεσχειν for D, exibere (sic) for d gig. The reading of
D (as intended) may be taken to suggest the usual, Christian meaning
of πίστις.
LS 1338 (s.v. παρέχω B II) give the meaning 'bring forward
witnesses or proofs’. This seems to be the technical usage, but the
active means furnish, supply, in a variety of contexts, and there is no
problem in its use here.
Calvin (2.125f.) again wishes to rewrite, or at least expand, the
speech. ‘There is no doubt that Paul said a good deal more about
Christ, so that the Athenians might know that he is the Son of God,
854 COMMENTARY ON ACTS

by whom salvation had been brought to the world, and to whom all
power in heaven and on earth had been given. Otherwise the speech,
such as we read it here, would have been powerless to persuade.’
This comment, by a theologian, on the theological content of the
speech as it stands, is not without interest.

32. The speech is ended. Haenchen (506), following Dibelius, is

probably right when he says that it is not to be thought of as a
fragment, in need of a supplement; ‘sie ist innerlich völlig ges-
chlossen’. Schneider (2.246) understands the framework in too
historicistic a way when he says that it was the speaker’s intention
that the hearers should respond (to v. 31) with the question, Who is
this man? Luke represents the resurrection of the dead as more than
the Athenians, or some of them, can stand. Cf. v. 18. He knows the
Greek belief that the dead do not stand up (ἀνιστάναι). Cf. Homer,
Iliad 24.551, ουδέ μv ἀνστήσέις; Aeschylus, Eumenides 647f.,
ἀνδρός δ’ ἐπειδάν αἶμ’ ἀνασπάση κόνις ἅπαξ θανόντος, οὔτις ἔστ’
ἀνάστασις; Agamemnon 1360f.; Sophocles, Electra 137-9, οὔτοι
τόν γ’ ἐξ Άῖδα παγκοινου λίμνας πατέρα θρήνοις ούτε λιταῖσιν
ανατάσεις. For Celsus’s reaction to (what he understands to be)
Christian teaching see Origen, c. Celsum 5.14. Since it is clear that
Paul is speaking absurdities some of his hearers (οί μέν) are content
to mock, ἐχλεύαζον. It is often supposed that the words of the second
group of hearers (οί δέ) come to the same thing—We will hear you
again; that is, we will not listen to you now, and we shall be careful
not to fix a date for a second session; we say that we will hear you
καὶ πάλιν, but we do not really intend to do so. This is probably a
mistaken understanding of the text. There is nothing in the narrative
as a whole to suggest that Luke wished to represent the mission in
Athens as a complete failure; and the use of the οἱ μέν ... οἱ δέ
construction seems decisive. It is impossible that Luke should have
meant, On the one hand there were those who mocked; on the other
hand, there were those who made a mocking reply. Certainly, the
latter group did not make the immediate and whole-hearted response
that the evangelist would have desired, but it is probably part of
Luke’s attempt to portray a philosophical scene that they should say,
This is interesting; we do not come to decisions quickly, but there is a
prima facie case and we should like to hear the argument again.

33. ούτως, so, receiving this response, εκ μέσου: cf. 23.10, but
here the words, with αυτών, correspond to v. 22, ἐν μέσῳ τού
Άρείου πάγου, and confirm the view (see also v. 34) that the
Areopagus is a company of people, not a locality.
Paul has said what he had to say, and there is no more to do. Cf.
Lk. 4.30 (διά μέσου αυτών) for a comparable event in the story of
Jesus; but there is no suggestion of a violent attack on Paul.
 45. PAUL AT ATHENS. 17.16-34. 855

34. The general meaning of this verse goes with that of v. 32. The
Athenian environment made Paul’s work exceptionally difficult, but
it was nevertheless not without fruit. Two converts can be named,
and there were others.
For κολληθέντες cf. 5.13, where there is a difficulty that does not
occur here, where κολληθέντες is linked with ἐπίστευσαν; those in
question definitely joined the Christian group. See C. Burchard
(ZNW 61 (1970), 159f.) There is no mention of baptism. Dionysius
the Areopagite is otherwise unknown. According to Dionysius,
bishop of Corinth in the second century, he was the first bishop of
Athens (Eusebius, HE 3.4.10; 4.23.3). He was sometimes confused
with Dionysius of Paris (c. AD 250) and with the author (Pseudo-
Dionysius) of a number of mystical writings (c. AD 500). His
designation as ’Αρεοπαγίτης is correct; the Council was ή ἐξ
Άρείου πάγου βουλή, but the compound form for an individual
member is usual in Athenian inscriptions, e.g. IG 3.1.704 (Hemer,
119). The term confirms that, in Luke’s view, the Areopagus was a
body of men, not a place. As an Areopagite, Dionysius must have
served as an archon; a not undistinguished person. It is possible that
the traditional name gave rise to the Areopagus story. Damaris also is
unknown; Luke’s interest in the part played by women is familiar; cf.
e.g. 17.4,12. On the name there is a detailed note in Hemer (232). No
other example of it is known, but it may be a variant of Damalis.
Lüdemann (202) takes the view that Dionysius and Damaris are
historical figures, but cannot have been converted on the occasion of
Paul’s first preaching in Athens, since Stephanas, Fortunatus, and
Achaicus were the ‘first fruits of Achaea’ (1 Cor. 16.15).
D spoils the sentence by turning the participle κολληθέντες into an
indicative, ἐκολλήθησαν, and omits Damaris, adding after
’Αρεοπαγίτης, ευσχήμων. Ramsay (Church 161f.) thinks that the
omission of Damaris was due to catholic depreciation of women, but
it was probably accidental, because at the end of the verse D retains
συν αὐτοῖς (plural). E describes Damaris as γυνή τιμία ονόματι Δ.
τιμία may (by way of e’s honesta) have led to D’s ευσχήμων, ὁ
(before ’Αρ.) is omitted by B and is replaced in D by τις.
 46. PAUL AT CORINTH, WITH RETURN TO PALESTINE
18.1-23

(1) After these things Paul1 left Athens and came to Corinth, (2) and found2 a
Jew, Aquila by name, of Pontus in origin, who had recently come from Italy,
with Priscilla his wife, because Claudius had issued an edict that all the Jews
should leave Rome. Paul1 approached them,3 (3) and because they were of
the same trade he4 stayed with them and worked; for by trade they were
tentmakers. (4) Every Sabbath he argued in the synagogue5 and sought to
persuade Jews and Greeks; (5) but when Silas and Timothy came down from
Macedonia Paul6 was constrained by the word, testifying to the Jews that7 the
Christ was Jesus. (6) When they opposed him and8 blasphemed Paul1 shook
out his clothes and said to them, ‘Your blood be upon your own heads;9 I am
clean. Henceforth I shall go to the Gentiles.’ (7) He moved away from there
and entered the house of Titius Justus,10 one who reverenced God, whose
house was adjacent to the synagogue. (8) But Crispus, the Archisynagogue,
became a believer in the Lord, with all his household, and many of the
Corinthians when they heard this11 believed and were baptized. (9) In the
night the Lord said to Paul in a vision, ‘Do not be afraid but continue to
speak and do not fall silent; (10) for I am with you and no one will set upon
you so as to harm you, for I have a people, a large people, in this city.’ (11)
He stayed a year and six months teaching12 the word of God among them.
(12) When Gallio was proconsul of Achaea the Jews with one accord set
upon Paul and brought him to the place of judgement, (13) saying, ‘This man
is persuading people to worship God in a manner contrary to the law.’ (14)
As Paul was about to open his mouth, Gallio said to the Jews, ‘If there had
been some matter of13 injury or wicked deceit, you Jews, I should14 of course

1Greek, he.
2NEB, fell in with.
3NJB, went to visit them.
4NEB, made his home with them.
5NJB, synagogues.
6RSV, occupied with preaching; NEB, devoted himself entirely to preaching; NJB,
devoted all his time to preaching.
7NJB, that Jesus was the Christ.
8RSV, reviled him; NEB, resorted to abuse; NJB, started to insult him.
9NEB, my conscience is clear; NJB, from now on I will go to the Gentiles with a
clear conscience.
10RSV, NEB, NJB, a worship(p)er of God.
11RSV, hearing Paul; NEB, listened; NJB, had heard him. In Greek the verb has no
object.
12NJB, preaching.
13RSV, wrongdoing or vicious crime; NEB, crime or grave misdemeanour; NJB,
misdemeanour or a crime.
14RSV, have reason to bear with you.
856
 46. PAUL AT CORINTH, WITH RETURN TO PALESTINE. 18.1-23. 857

have been forbearing with you; (15) but if they are disputes15 about talk and
words'6 and the law that you observe, you will have to see to it yourselves; I
have no wish to be a judge of these things.’ (16) And he drove them from the
place of judgement. (17) They all took hold of Sosthenes the Archisyna-
gogue and beat him before the place of judgement; and none of these things
troubled Gallio.17
(18) Paul stayed on a number of days, took his leave, and sailed away to
Syria. With him were Priscilla and Aquila. He had shaved his head in
Cenchreae, for he had a vow. (19) They reached Ephesus and there he left
them; he himself went into the synagogue and argued with the Jews. (20)
When they asked him to stay a longer time he did not consent, (21) but took
his leave with the words, T will come back to you, God willing.’ Then he set
sail from Ephesus (22) and landed at Caesarea. He went up and greeted the
church and came down to Antioch.
(23) Having spent some time there he left, passing through, in order, the
Galatian territory and Phrygia, strengthening all the disciples.

Bibliography
S. Benko, ThZ 25 (1969), 406-18.
Μ. Black, in FS Nida, 119-31.
E. Dinkier, Signum Crucis, 118-33.
E. J. Goodspeed, JBL 69 (1950), 382, 383; 110 (1991), 439-49.
K. Haacker, BZ 16 (1972), 252-5.
C. J. Hemer, JTS 28 (1977), 99-101.
C. J. Hemer, in FS Bruce (1980), 3-18.
R. F. Hock, BL 97 (1978), 555-64.
R. F. Hock, CBQ 41 (1979), 438-50.
G. H. R. Horsley, NovT 34 (1992), 105-68.
J. Jeremias, ZNW 30 (1931), 299.
P. Lampe, BZ 31 (1987), 256-61.
G. Lüdemann, in FS Schneider, 289-98.
J. Murphy-O’Connor, St Paul's Corinth (1983).
J. Murphy-O’Connor, JBL 112 (1993), 315-17.
J. L. North, NTS 29 (1983), 264-6.
R. Riesner, Die Frühzeit des Apostels Paulus, WUNT 71 (1994), 139-80;
180-89.
J. Μ. Ross, NovT 34 (1992), 247-9.
D. Slingerland, JBL 110 (1991), 439-49.
D. Slingerland, JQR 83 (1992), 127-44.
D. E. Smith, HThR 70 (1977), 201-31.
C. H. Talbert, NovT 9 (1967), 26-40.
J. Wiseman, ANRW 2.7.1 (1979), 438-548.
15NEB, bickering; NJB, quibbles.
16NEB, NJB, words and names.
17RSV, but Gallio paid no attention to this; NJB, Gallio refused to take any notice at
all.
858 COMMENTARY ON ACTS

Commentary

This paragraph unites a striking number of Lucan themes. Paul

continues his travels and visits two notable cities, Corinth and
Ephesus, both of which are to become important centres of early
Christianity. He is assisted by colleagues: among those who are
named are new ones, and in introducing them Luke finds another
opportunity of relating his story to secular history. Paul begins his
work in the synagogue, but finds it necessary to move to a different
centre, and the Jews (with some exceptions) prove to be opponents
rather than allies. Their opposition has the effect of bringing Paul a
second time (cf. 16.19-23) before a Roman authority, and the
reference to Gallio (v. 12) provides the modern historian with the
means of dating the event with fair accuracy and provides Luke with
an example of Rome’s sense of justice and unwillingness to take
action against the Christians. After the encounter with Gallio Luke
shows Paul again on his travels and is able to represent him as a good
Jew, who takes a vow upon himself (unless v. 18 means that Aquila
took the vow). Paul visits Palestine, perhaps Jerusalem (v. 22), then
sets out again towards the old mission fields.
All this Luke will have enjoyed writing. Every piece contributed to
the picture of Paul that he wished to convey. To infer from this that it
was all Luke’s own invention would be mistaken. That Paul was
acquainted with and assisted by a married couple, Aquila and
Priscilla, is confirmed by the epistles (Rom. 16.3-5; 1 Cor. 16.19: cf.
2 Tim. 4.19). Expulsion of the Jews from Rome raises historical
problems (see on v. 2) as regards its date but hardly as regards the
fact. There is nothing in the NT more certain than that Paul founded a
church in Corinth and continued to have a complex and sometimes
stormy relation with it; the name of Crispus (v. 8) recurs in 1 Cor.
1.14. Gallio was proconsul of Achaea in the 50s of the first century.
The epistles also bear witness to the fact that Paul visited Jerusalem
from time to time.
Lüdemann (203-12) finds in vv. 1-17 a mixture of tradition and
redaction; Schneider (2.247) agrees with Haenchen (516) that Luke
wrote out the whole Corinthian story, not interpolating pieces here
and there into a source. It seems probable that both of these views are
true and must be combined. Schmithals (166) rightly observes that in
the whole section there are too many concrete details, which in
themselves show no special Lucan tendency, for the paragraph not to
have been drawn from a Paul-source. The name for example of Titius
Justus (v. 7) serves no purpose beyond its evident narrative use; Luke
would gain nothing by inventing it. But it is equally true that the
whole section shows Lucan editorial management. There is no first
person plural in it; Luke probably wrote up in his best style pieces of
information derived from the Pauline circle and, perhaps, collected in
 46. PAUL AT CORINTH, WITH RETURN TO PALESTINE. 18.1-23. 859

Corinth itself. Apart from the name Crispus there is nothing to

suggest knowledge of the Corinthian epistles; these were not Luke’s
source of information.
There are two points at which the historicity of Luke’s story may
be seriously questioned. The first turns on the date of the expulsion of
the Roman Jews by Claudius. The question is discussed on v. 2. If the
view is taken that the expulsion took place, and resulted in the arrival
in Corinth of Aquila and Priscilla, in AD 41, we must suppose that
Paul reached Corinth for the first time in that year, or soon after. This
throws out the whole of Pauline chronology as this is usually
understood, and also means that the order of Acts is distorted; Paul
reached Corinth before the famine (11.28) and before the Council
(15.6-29). We must of course be prepared, if the facts require it, to
review the whole question, but the view taken here is that though
Claudius did act against the Jews early in his reign he did not expel
them till much later.
More serious questions are raised at the end of the paragraph.
According to v. 18 Paul (but Aquila may be intended—see the note)
had a vow, usually taken to be a Nazirite vow. Is it credible that Paul
would accept so much of Judaism? For this, unlike (for example)
keeping the Sabbath, was not an obligatory act; one could be a very
good Jew without ever taking a vow. This therefore was not a matter
of being to the Jews (as if he were) a Jew (1 Cor. 9.20), though it
could be part of Luke’s picture of Paul as the good Jew. The vow is
mentioned in the account of a journey from Corinth to Syria, perhaps
(see the note on v. 22) to Jerusalem. No events are described as
taking place in Palestine, and at least since Loisy (704) it has been
maintained that we have here a doublet of the journey from Greece
(Corinth?) to the east (εἰς τήν Συρίαν in both 18.18 and 20.3) which
introduces the last stage in Paul’s career, the legal contests with both
Jews and Romans. The suggestion is possible but further questions
arise: What purpose can Luke have had in introducing an additional
journey to Palestine? And why, if he himself is responsible for it,
does he omit to say that Paul reached Jerusalem? The first question
can be answered by the existence of the next two pieces, which Luke
probably found in existence: the enlistment of Apollos in the Pauline
circle (18.24-28) and Paul’s treatment of disciples of John the
Baptist in Ephesus (19.1-7). The first of these made it necessary to
get Paul out of the way (the long journey from Ephesus to Palestine
and back would leave Aquila and Priscilla a free hand to conduct
affairs in Ephesus), and the brief halt in Ephesus (vv. 19-21) would
give Paul a foothold and enable him to solve a problem there. All this
it is impossible to disprove, but it is also impossible to disprove that
Paul made a hurried journey back to Jerusalem. He had heard of the
threatened destruction of his work in Galatia by men who, rightly or
wrongly, assumed the support of Jerusalem; did he hurry back to ask
860 COMMENTARY ON ACTS

his Jerusalem colleagues what they meant, then make haste back to
the Galatian territory and Phrygia (v. 23) to set things right? To this
the vow could be Luke’s counterpoise: Paul was the adversary of
Judaizers, but he remained a true Jew.
According to Taylor (5.307f.), vv. 1 (TO), 4a, 5b, 6a (TA), 9, 10a
(TO), 12-14a, 15b-17, come from Act I; Act II provides additions
and changes in vv. 2-3, 4b, 5a, 7-8, 11, 14b-15a.

1. Μετά ταῦτα, after the Athenian episode was over. As a

connecting link the phrase is common in the gospels, not in Acts; in
addition to this passage 7.7; 13.20; 15.16, all in OT material, quoted
or summarized. Paul left (χωρίζεοθαι; 1.4; 18.2) Athens; see on
17.16. ‘Άθήναζε, Άθήνηθεν, Άττικώς· εἰς ’Αθήνας, ἐξ ’Αθήνων,
Έλληνικῶς’ (Moeris). He came to Corinth; in full, Colonia Laus
Julia Corinthiensis. For some account of the city see 1 Corinthians
1-3,2 Corinthians 1, 2; more fully J. Wiseman (ANRW II 7.1 (1979),
438-548) and J. Murphy-O’Connor (St Paul's Corinth: Texts and
Archaeology (1983)). The old city had been wealthy and influential,
but ‘in 146 BC a sharp line is drawn through the history of Corinth,
when Rome brought the Achaean League to an end. After the
decisive engagement at Leucopetra, on the isthmus, the consul
Lucius Mummius was able to occupy Corinth without a blow. The
citizens were killed, or sold into slavery; the city itself was levelled
with the ground ... After 100 years of desolation Corinth was
refounded by Julius Caesar as a Roman colony’ (1 Corinthians 1f.).
Its natural advantages enabled it to regain its prosperity; it probably
did little to shake off the reputation for immorality that the old city
had acquired. From Athens Paul would have about 37 miles to travel,
by way of Eleusis and Megara.
For μετά ταῦτα χωρισθεὶς ἐκ, D (cf. h) has ἀναχωρήσας δέ από,
without change of meaning, but does not add (as A E Ψ <a> sy(p) boms
do) ὁ Παῦλος. Metzger (460) is no doubt right in the view that the
name was added when the passage was used as an ecclesiastical
lection. The earliest MSS do not suggest this use.

2. ευρών: εύρίσκειν may mean finding as the result of chance (e.g.

9.33), or finding as the result of search (e.g. 9.2). A combination of
the two may be seen here: Paul would look for a fellow Jew with
whom he might settle; he would be glad to come across a fellow
σκηνοποιός, surprised to find a fellow Christian (if Aquila and
Priscilla were already Christians, as seems probable; see below); and
he did not know that the person he would find would be Aquila.
’Ιουδαῖον. There was a Jewish community in Corinth; see in
addition to Acts itself and the note on v. 1 NS 3.64-66; for a
fragmentary inscription from a Corinthian synagogue, Background
53; also E. Dinkler (Signum Crucis 131) for fragments found near the
 46. PAUL AT CORINTH, WITH RETURN TO PALESTINE. 18.1-23. 861

theatre, ‘ein Säulenende mit drei Menorot und dazwischen Lulab und
Etrog’.
Άκύλας. See on 18.18, 26; also Rom. 16.3; 1 Cor. 16.19; 2 Tim.
4.19. It is clear that Aquila and his wife (see below) played an
important part in early Christianity. From the passages cited it
appears that they entertained Christians in their home; travelled with
Paul; instructed others. Aquila came from Pontus (see 2.9; also 1 Pet.
1.1); τῷ γένει would normally mean by race, but Aquila has already
been described as a Jew (cf. Barnabas, 4.36). His family must have
settled in Pontus, which was united with Bithynia (see 16.7) to form
a Roman province. He was not a Roman citizen, or he would not
have been affected by the decree mentioned in this verse. Christianity
spread rapidly in Pontus (Pliny, Epistles 10.96.5, 9, 10). This may
suggest that there was already a good population of Jews, but of
further evidence for this NS 3.35f. can supply only a reference to
Philo, Legatio ad Gaium 281. At some point Aquila had moved to
Rome. For interesting, but not really important, references that
connected the name Aquila with both Rome and the Pontus region
see Hemer (232f.).
Aquila had recently (προσφάτως) come from Italy (ἐληλυθότα).
Μ. 3.85 says that Classical Greek would probably have used the
aorist rather than the perfect participle; whether it is true that ‘there
was a distinct tendency in the Hellenistic period to connect very
closely a past action with its present consequences’, so that they
thought, ‘he has been here since coming from Rome recently’ is
another question. The choice of participle may be no more than a
literary fad. Luke will mention in a moment the reason why Aquila
had left Italy (Rome).
Aquila is unique in the NT in that whenever he is named his wife
is named too. In Acts she is Priscilla; in the epistles she is Prisca. It is
hardly open to doubt that the same woman is intended, καὶ
Πρίσκιλλαν γυναίκα αὐτοῦ is somewhat awkwardly attached to the
sentence but this is an insufficient reason for regarding the words as
an insertion, especially since to omit them here would leave Priscil-
la’s presence in 18.18,26 unexplained. It is most improbable that she
would have been mentioned so frequently by name if she had not
been an outstanding person in her own right. The two were in Corinth
διὰ το διατεταχέναι Κλαύδιον χωρίζεσθαι πάντας τούς Ιουδαίους
από τής 'Ρώμης. διατάσοειν means to issue a διάταγμα, Latin
edictum. The simple form, τεταχέναι, read by D E L P 6 33 104
323 (1175 1241) pm, must be understood in the same way. Luke
means that the Jews had been banished from Rome by imperial edict.
It would make an important contribution to the chronology of Acts if
this edict could be dated. Suetonius, Claudius 25, writes: ludaeos,
impulsore Chresto assidue tumultuantes, Roma expulit. This act
occurs in a series of what may be described as police measures, in
862 COMMENTARY ON ACTS

regard to one of which the term edictum is used: Viatores ne per

Italiae oppida, nisi aut pedibus, aut sella, aut lectica transirent,
monuit edicto. There is a similar expression in which the word
edictum is used by Suetonius in Vitellius 14: ... edictum suum, quo
iubebat intra Kalendas Octobris Urbe Italiaque mathematici exce-
derent. Whether Chrestus is (in Claudius 25) the name of a person
otherwise unknown, an error for Christus, or a corruption of Christus
in the MS tradition, is disputed and cannot be settled with certainty.
It is known from Romans (as well as from Acts—28.15) that
Christianity had reached Rome before Paul did, and the suggestion is
at least plausible that the Jewish community had become intolerably
agitated by Christian preaching. Suetonius supplies no evidence for
the date of the expulsion. Orosius (5th century), Historiae contra
Paganos 7.6.15f., says, Anno eiusdem nono expulsos per Claudium
urbe ludaeos Josephus refert. There is no such reference in the works
of Josephus as now extant, and the source of Orosius’ remark is
unknown. It seems quite probable that he had in mind a confused
recollection of Josephus, Ant. 18.81-84, which tells the story of four
Jews who persuaded the proselytess Fulvia to send large gifts to the
Jerusalem Temple, which they themselves pocketed. As a result
Tiberius (not Claudius) κελεύει πᾶν τό ’Ιουδαϊκόν τής 'Ρώμης
ἀπελθεῖν. No reliance can be placed upon Orosius (on whom see
e.g. B. Altaner, Patrologie, § 59.II.9 (with bibliography)), though in
fact he may be right about the date. The ninth year of Claudius
corresponds almost exactly with AD 49 (the accession of Claudius
took place on 25 January 41). If Paul arrived in Corinth in or just after
this year, Aquila and Priscilla having only recently (προσφάτως)
come from Rome, the date would harmonize with what may be
deduced from the reference in v. 12 (see the note) to Gallio. Claudius’
relations with Jews deteriorated after the death of Herod Agrippa I
(see I.590-2); see A. Momigliano (Claudius (ET 1934), 32).
The date (AD 49) is disputed on the ground of a reference by
Cassius Dio to anti-Jewish action taken by Claudius in AD 41.
Lüdemann (209) takes this to be the date of the expulsion mentioned
by Suetonius and deals with the Pauline chronology on this basis. But
Dio (60.6.6) does not speak of an expulsion; his words are οὐκ
ἐξήλασε μέν ... ἐκέλευσε μή συναναθροίζεσθαι. This statement
may in fact have been intended to point forward to an occasion when
Claudius did expel the Jews, described in a later—lost—book of
Dio’s history. Tajra (52-4), Momigliano (loc. cit.), and Jewett
(Dating Paul’s Life (1979), 36-8) are very probably right in thinking
that Claudius twice took steps against the Jews. The date 41 is
accepted by Taylor (5.312-14); but see also Hemer (167f.); NS
3.77f.; D. Slingerland (JBL 109 (1990), 686-90); J. Murphy-
O’Connor (JBL 112 (1993), 315-17).
The addition after 'Ρώμης of οι κατῷκησαν εις τήν Άχαῖαν (D),
 46. PAUL AT CORINTH, WITH RETURN TO PALESTINE. 18.1-23. 863

and the further addition (h syhmg), ‘Paulus autem agnitus est Aquilae’,
may come from tradition of some kind; the substitution of προσηλθεν
αὐτῷ ό Παῦλος (D(c)) for προσηλθεν αυτοῖς may possibly represent
an (anti-feminist) interest of a Western editor; for the reconstruction
of the Western text at this point see Ropes (Begs. 3.170f.) and Clark
367f. προσῆλθεν suggests that the initiative was Paul’s; whether it
was due in the first instance to the necessity of earning a living, to the
natural desire to seek the company of fellow Jews, or to evangelistic
enterprise cannot be determined; they may all be true.
Brandon (Fall 145, 147) suggests that the encounter with Aquila
and Priscilla, with the attendant fact that the mixed Roman church
had lost its large Jewish element, led to the writing of Romans. The
epistle certainly seems to have been written from Corinth or
thereabouts.

3. Whether or not Paul sought out Aquila and Priscilla in order to

find work he stayed with them (ἔμενεν) because he was ομότεχνον;
he and Aquila were both σκηνοποιοί. This is probably the right way
to take the plural noun; but Taylor (5.315) (cf. ND 2.17) thinks it
refers to husband and wife. For a short note on the social status of the
craft see W. A. Meeks (The First Urban Christians (1983), 59). The
natural rendering of σκηνοποιός is tent-maker, and this should in
fact be accepted, though Begs. 4.223 (cf. J. Jeremias, ZNW 30 (1931),
299) strongly supports leather-worker, the strongest evidence for this
is that h renders the word lectarius, maker of beds, that is presumably
of the leather-covered cushions used as mattresses, and that the
Peshitto has which, according to Begs. 4.223 ‘merely trans-
literates the Latin lorarius, a maker of leather thongs’. It is however
at least arguable that lectarius arose as the corruption (in Greek) of
σκηνοποιός into κλινοποιός, and the meaning of lulara is not
certain; according to Payne Smith (238) it is a maker of rough cloth
for tents or horsecloths. In any case, it cannot be assumed that h and
pesh understood σκηνοποιός correctly. It is true that Chrysostom
(Cramer 3.302) says that Paul ἐπ'ι σκηνορραφείου ἐστώς δέρματα
ἐρραπτε, but there is nothing in σκηνοποιός to say of what material
the tents were made and skins are as likely as the goat hair that was
woven in Cilicia (Tarsus!) into coarse cloth. It is in the highest
degree unlikely that the other meaning of σκηνοποιός (LS 1608)
should be adopted—the three Jews were not makers of stage
properties! Preuschen (111) draws attention to CIL III 5183b, aeditus
collegi tabemaclariorum, which shows the existence in Rome of a
corporation of tentmakers.
μένειν here approaches the stronger sense of to lodge. Cf. Rom.
16.3-5; 1 Cor. 16.19; not only was there a church κατ’ οἶκον αυτών,
all the Gentile churches had reason to be grateful to Aquila and
Priscilla. It is probable that they were wealthy and entertained not
864 COMMENTARY ON ACTS

only Paul but other Christians too. For παρ’ αὐτοῖς, D 36 453 pc
have πρὸς αυτούς. There is no difference in meaning; the variant
suggests a copyist or editor concerned to give the sense but not
unduly careful about word-for-word accuracy, ἠργάζετο (P74 (c) A
D Ε Ψ <a> lat sy samss) is more probably original than the plural (*
B(2) Samss Paul's readiness to work to support himself was a
theme important to Luke (see 20.34), in itself as a record of Paul’s
manner of life (attested also in the epistles, 1 Cor. 4.12; 2 Cor. 11.7; 1
Thess. 2.9) and also as an example for future generations of ministers
(20.35). The importance of this is perhaps underestimated by Barth
(CD 3.4.4.72: ‘His work is done on the margin of his apostolic
existence, and such exhortations are given only on the fringe of his
apostolic instruction.’
The clause ἦσαν γάρ ... τέχνη is omitted by D gig (which with
other Western MSS will differ more widely from the Old Uncial text
in ν. 4). If this is not simply a by-product of a wholesale rewriting of
the text one must suppose that the Western editor thought tent-
making an unworthy occupation for an apostle.

4. διελέγετο: the word is characteristic of Luke’s accounts of

Paul’s work; see e.g. 17.17. ἐν τῆ συναγωγῆ: for the synagogue and
its role in the early Christian mission see on the same verse; for a
synagogue in Corinth see on v. 2; for a (very dubious) conjectural
restoration (including the word ἀρχισυνάγωγος as well as
συναγωγή) of a ‘very fragmentary’ Corinthian inscription (SEG 29
(1979 [1982], 300) see ND 4.213 (with on the following pages a
valuable discussion of the two words). It is not surprising that a
wealthy and cosmopolitan city such as Corinth should contain a
colony of Jews, or that they should be able to provide themselves
with a building (contrast the Jews at Philippi, 16.13).
κατά πᾶν σάββατον: the weekly service provided a ready-made
congregation; if Paul was not invited to preach (as e.g. at Pisidian
Antioch, 13.15) he would at least have the opportunity of conversing
with those who attended.
ἔπειθεν: conative use of the imperfect. Paul set about the task of
persuading men to become Christians, he endeavoured to persuade
them. For the use of πείθειν cf. 17.4. Paul’s efforts were directed to
both Jews and Greeks; the latter were presumably Gentiles whom he
found in the synagogue. He turns decisively to the Gentiles in v. 6.
The whole of this verse is omitted by the Sixtine Vulgate, and by
WW. It is contained in some Vulgate and other Latin MSS; for
details see WW 3.158. D(c) (to which gig syhmg approximate) has
εἰσπορευόμενος δέ εἰς τήν συναγωγήν κατά παν σάββατον
διελέγετο καί ἐντιθείς τὸ όνομα τού κυρίου ’Ιησοῦ και ἔπειθεν δέ
ού μόνον ’Ιουδαίους αλλὰ καί Έλληνας. See Ropes (Begs. 3.172)
and Metzger 460f. In his translation Metzger drops without comment
 46. PAUL AT CORINTH, WITH RETURN TO PALESTINE. 18.1-23. 865

the καί before ἐντιθείς and translates the participle inserting (so also
Bruce 2.348), which seems an odd piece of English; bringing in, if
slightly colloquial, would be a more natural idiom: Paul explained
the OT, bringing the name of Jesus into his interpretation, καί
ἐντιθείς ... καί ἔπειθεν is incorrect; it cannot be explained here as an
Aramaism. It is best taken as due to careless paraphrasing and
expansion of the Old Uncial text.

5. Temporal ὡς is characteristic of Acts; see on 16.4.

κατῆλθον (but παρεγένοντο δέ in D (h)), down from the moun-
tainous districts in the north to the maritime region of Corinth in
northern Peloponnese. Silas and Timothy had been left in Beroea
(17.14), and when Paul reached Athens he sent back a message that
they were to join him as soon as possible (17.15). In the account of
Paul’s Athenian mission that follows (17.16-34) there is no refer-
ence to Silas and Timothy; it seems however from 1 Thess. 3.2 that
Timothy at least joined Paul in Athens, since Paul, anxious about his
Thessalonian friends, decided that he could put up with loneliness in
Athens and sent Timothy for news. The movements of Paul’s
colleagues were probably more complicated than Luke knew, or saw
fit to describe. He treats Silas and Timothy as a pair; it may be that,
for some reason unknown to us, they decided for a time to separate,
or were given different instructions by Paul. The meagre evidence at
our disposal would be satisfied if (after 17.14)

(1) Paul went to Athens, leaving S. and T. in Beroea.

(2) T. joined Paul in Athens, leaving S. in Beroea (or somewhere in
Macedonia).
(3) T. was sent by Paul from Athens to Thessalonica,
(4) T. returned from Thessalonica (1 Thess. 3.6), possibly accom-
panied by S., since Silvanus joins in the writing of 1 Thessa-
lonians (1.1).
(5) This return may have been to Corinth, if 1 Thessalonians was
written from Corinth. If it was written from Athens, T. and S.
will have made yet another visit to Macedonia, whence they
return at 18.5.

'It is perhaps easier to accept the plain statement of 1 Thessalonians

and assume that the writer of Acts made a mistake in thinking that
Silas and Timothy did not join Paul before he had reached Corinth’
(Begs. 4.224). It is possible that the words ὅ τε Σιλᾶς καί ό Τιμόθεος
lay some emphasis on both Silas and Timothy, implying an earlier
occasion when only one of the two came down from Macedonia.
συνέχεσθαι is a Lucan word (Lk., 6 times; Acts, 3 times; Mt.,
once; Paul, twice), and is used frequently in a bad sense (e.g. 28.8,
πυρετοῖς καί δυσεντερίῳ συνεχόμενον). To this use the main
866 COMMENTARY ON ACTS

exception is Paul (2 Cor. 5.14; Phil. 1.23). In the present passage there
is no difficulty in the reading of the majority of MSS (<a> syhmg),
συνείχετο τῷ πνεύματι: Paul was constrained by the Spirit, who
impelled him to concentrate on his mission. This reading, however,
because it is easy, should probably be regarded as an alleviation of
συνείχετο τω λόγω (the rest). This must be understood in a similar
way. The word of God (conceived as an almost personal force)
constrained Paul. Another possibility is, Paul was confined (confined
himself) to the word; that is, though he had previously spent some of
his time working as a tent-maker, he now, after the arrival of Silas
and Timothy, who were perhaps able to earn enough for three, or had
brought money from Macedonia (2 Cor. 11.8f.—so Haenchen 517),
concentrated on preaching.
The above interpretation may not do justice to Luke’s use of
tenses—the aorist κατήλθον and the imperfect συνείχετο. ‘The
imperfect clearly expresses that when they arrived “they found Paul
wholly occupied with the word” ’ (Page 199).
For the characteristic διαμαρτυρόμενος cf. 2.40. Paul still
addresses Jews and the mission is presumably still based on the
synagogue.
εἶναι τόν χριστόν 'Iησοῦν. τόν χριστόν, having the article, will be
the subject. Paul’s contention was that the Christ (he could assume
that his Jewish hearers knew that there would be a Messiah) was
Jesus. Cf. 17.3. This was the fundamental and distinctive Christian
conviction; cf. 2.36.
Schmithals 168f. finds no problem in 1 Thess. 3.1, 6 since, in his
view, these verses deal with events on the third journey.

6. The Western text (D h (syhmg)) opens this verse with a vivid

account of much talk and biblical discussion (πολλοῦ δέ λόγου
γινομένου καὶ γραφών διερμηνευομένων); the main substance of
the verse begins with the negative response of the Jews to the
testimony of v. 5. The blasphemy (βλασφημούντων—omitted by
P74—could refer to evil speaking against the human bearer of the
message, but in the context probably means blasphemy) consists in
the denying of Paul’s affirmation—Jesus, they say, is not the Christ;
they may have reinforced their denial by saying what they thought he
really was.
Paul’s reaction (see Wilson, Gentiles 225f.; for the language Begs.
5.274f.) is to shake out his clothes (τά ἱμάτια; BDR § 141.5, n. 11
think that the plural may be used for the singular ἱμάτιον, his cloak;
in any case we can hardly suppose that Paul stripped completely, and
the vague English ‘clothes’ probably represents what Luke meant).
The same verb, ἐκτινάσσειν, is used at 13.51, there more naturally
for shaking dust off the feet. It is not clear why Paul should shake out
his clothes, though it is clear enough that he means to break off
 46. PAUL AT CORINTH, WITH RETURN TO PALESTINE. 18.1-23. 867

relations with the Jews against whom he performs the symbolic

action. The uses of ἐκτινάσσειν in MM 199 all seem to be in various
senses literal; BA (496) cite UPZ 6.11, where ἐκτ. is ‘Gebärde d.
Unschuldsbeteuerung’. This would agree with καθαρός ἐγώ
(below). In UPZhowever, Wilcken has ‘schüttelten (? den Kopf)’.
τό αίμα υμών ἐπί τήν κεφαλήν υμών: supply ἐλθέτω and cf. Mt.
23.35, where the verb is necessarily added. BDR § 480.5, n. 8 cite
this parallel, also Mt. 27.25. Neither of these is a perfect parallel
since in them it is another person’s blood that is to come upon ‘you’
(equivalent here to ‘on your head’); that is, you have shed his blood
and you will be held responsible for having done so. In Paul’s words,
the point is that the Jews will be responsible for their own loss in
rejecting the Gospel; Paul is free of responsibility, having faithfully
proclaimed it to them. Cf. 20.26; Ezek. 33.4, 5. Ammonius (Cramer
3.303) comments, ασαφές ἐστι τό ρητόν; that is, presumably, it was
not a Greek idiom. He continues, οἶμαι δέ αυτό τοῦτο λέγειν
ἔκαστος τῶν ἀπιστούντων Χριστώ, ὅς ἐστι ζωή, δοκεῖ εαυτόν
φονεύειν ... Wilcox (65) thinks that there may be an allusion to 2
Kdms 1.16; it is a Septuagintal expression, though not frequent.
It is not clear whether a stop should be placed after καθαρός ἐγώ
or these words should be attached to what follows: either (a) T am
clean; henceforth I shall go to the Gentiles’, or (b) ‘Clean, I shall go
to the Gentiles’. There is a similar ambiguity in the Western variant,
ἀφ’ υμών (D*vid h) for ἀπό του. It is against (a) that it makes the
second sentence start without a connecting particle (though it could
be said that ἀπό του νυν has the force of an adversative particle), but
against (b) that it yields a sentence that is not characteristically Lucan
in form. For Paul’s turning to the Gentiles cf. 13.46; 28.28. Luke
evidently thinks of it as a frequently repeated pattern rather than as a
once-for-all event. So e.g. Blass (196f.) (Hoc est: non amplius in
hanc synagogam intrabo, sed alium locum quaeram ubi praedicem,
Graecis scil. Ceterum ad eos tantum Iud. qui Corinthi erant hoc
pertinere patet ... ) and Weiser (492) (‘Aus dieser Programmatik
folgt für Lukas aber nicht, dass Paulus sich bei der nächsten
Gelegenheit nicht doch wieder an Juden wendet’).

7. Paul transferred his operations (μεταβάς) from there (ἐκειθεν),

that is, probably, from the synagogue, though the Western text took
the words to mean that Paul separated from Aquila (μεταβάς δέ από
Άκύλα, D*vid h; 614 has από του Ά.). This is an improbable way of
taking ἐκεῖθεν (which does not mean from him), and is also
inconsistent with Paul’s references to Aquila in the epistles. It is
possible that the Western editor meant only that Paul no longer lived
with Aquila, but with Titius Justus; the meaning however seems to be
not that Paul lodged with Titius Justus but that he used his house as a
preaching centre; cf. 19.9. It is possible that Luke’s reticent statement
868 COMMENTARY ON ACTS

conceals the fact that Paul was expelled from the synagogue (became
αποσυνάγωγος—Jn 9.22); Wilson (Law 116).
The new host is called Titus Justus by E 36 453 945 1175 1739
1891 pc co; Titius Justus by B* D2 syh; Justus by A B2 D* Ψ <a> p;
Titus by syp. The most probable form of the name is Titius Justus.
Titus and Titus Justus probably arose from a desire to identify the
person whose house Paul used with the colleague mentioned fre-
quently in the epistles (2 Cor. 2.13; and elsewhere). Justus may have
arisen by the omission of Titus on the ground that the Titus of the
epistles was not a Corinthian. It is an interesting but not quite
convincing suggestion that Τιτίου and Τίτου Ίούστου arose from
Ίούστου through repetition of the last letters of ονόματι and the first
letters of Ίούστου. Metzger rejects the suggestion because ὀνόματι
is omitted by A pc h.
'Of all the persons with Latin names only one has two convincing
Roman names, Titius Justus in Acts 18.7’ (Sherwin-White 158). His
praenomen is not given. E. J. Goodspeed, JBL 69 (1950), 382-3, on
the basis of 1 Cor. 1.14; Rom. 16.23, conjectures that he was Gaius
Titius Justus. He was one who reverenced God; see on 10.2. His
house was adjacent (συνομορεῖν is a rare and late word; the simplex
is more frequent and has the same meaning) to the synagogue.
This may have been convenient but would hardly promote good
relations.

8. Κρῖσπος, probably the Crispus whose baptism (not recorded

here unless in the plural at the end of the verse) is mentioned in 1
Cor. 1.14. As ἀρχισυνάγωγος (for this office see 13.15; each Jewish
community had a number of synagogue heads, their number depend-
ing on the size of the community ; for Corinth see on v. 4) he was a
notable convert, though evidently not weighty enough to carry with
him the Jews as a group, or even his colleagues in office. The aorist
ἐπίστευσεν describes his conversion, not the lasting faith he main-
tained as a Christian. It is here constructed with the dative τῷ κυρίῳ.
D h have εις τόν κύριον, though the Western text at the end of the
verse has a dative. The usage in Acts varies.
Crispus was accompanied in his move into the Christian body by
all his household; for the expression, and the question whether Luke
intended the reader to understand that little children and infants were
included, see on 16.15, 31-33. It is to be noted that it is not said that
all the household (or for that matter Crispus himself) were baptized,
but that they believed. They were not the first converts in the locality;
this distinction was held by Stephanas and his household (1 Cor.
16.15). Many others followed, described simply as τών Κορινθίων;
probably we are to think that some were Jews, some Gentiles. The
Gentile element is emphasized by the reading of D (h), which add,
after ἐβαπτίζοντο, πιστεύοντες τῷ θεῷ (contrast the Western εἰς
 46. PAUL AT CORINTH, WITH RETURN TO PALESTINE. 18.1-23. 869

and accusative above) διὰ τοῦ ονόματος τοῦ κυρίου ἡμών ’Ιησοῦ
Χριστου—that is, they came to faith in God only through Christ, not
through the Jewish religion. They believed because they heard,
ἀκούοντες (but P74 L 614 1241 2495 pm have ἀκούσαντες). It is not
clear whether this means because they heard the word of God
preached by Paul or because they heard of the conversion of Crispus.
According to Schneider 2.251 if Luke had meant the former he
would have included αυτού in the sentence; this is not a conclusive
argument.
The imperfect ἐπίστευον (contrast ἐπίστευσεν, above) could be
inceptive—they began to be believers—but the imperfect
ἐβαπτίζοντο cannot be explained in this way. Both imperfects reflect
their subject, one after another became a believer and was baptized.
On baptism in Acts see Introduction, pp. xc-xcii.
9. The divine message of encouragement stands at this point to
mark the transition from work in the synagogue to unrestricted
mission among the Gentiles (v. 6), and to prepare for the attack of v.
12. Relative (though not complete—v. 8) failure in the synagogue
must not hold Paul back; the Jews may be unbelieving but there can
yet be many who will enter the people of God.
εἶπεν δέ ό κύριος. For the direct speech of the Lord (Jesus) see
9.4-6, 10-16; 10.13, 15: 11.7, 9: 22.18, 21: 23.11. It comes at
notable points in Paul’s career.
ἐν νυκτί; cf. 16.9, διὰ νυκτός. No difference is intended between
the two prepositional phrases, διά is probably avoided here because
δι’ οράματος follows. This probably refers to a dream. ὅραμα occurs
at 7.31; 9.10, 12 (si v.1.); 10.3, 17, 19; 11.5; 12.9; 16.9, 10;
elsewhere in the NT only at Mt. 17.9. In 2 Cor. 12.1 Paul refers to
visions (here, ὀπτασίαι), but records only one, which had appeared
to him 14 years previously. There can be little doubt that Luke was
more impressed than Paul by visions and similar phenomena, but it is
not necessary to infer from 2 Corinthians 12 that Paul had had no
visionary experience for 14 years, still less that he would have
disapproved of this one.
μή φοβοῦ, that is, Do not fear those who might wish to harm you
(v. 10). This means that the parallels adduced by Betz (55) are not
true parallels, since they bid the hearer not to fear the supernatural
giver of vision or voice. Deut. 31.6; Josh. 1.6, 9; Isa. 41.10; Jer. 1.8
are better parallels.
λαλεῖ, present imperative, and μή σιωπήσης, aorist subjunctive, fit
well together: Go on speaking, as you have done; do not fall silent.
Acts of Thomas 1 (L.-B. 2.2.100) may show knowledge of this
passage: ὤφθη αὐτω ὁ σωτὴρ διά τής νυκτός, καὶ λέγει αὐτῷ. Μὴ
φοβοῦ, Θωμά, ἄπελθε εις τήν ’Ινδίαν ...
10. Paul is not to fear (v. 9) διότι (3 times in Lk and in no other
870 COMMENTARY ON ACTS

gospel; 5 times in Acts) ἐγώ έιμι μετά σοῦ. The promise recalls OT
passages, e.g. (in addition to those cited on v. 9) Exod. 3.12; Isa.
43.5. ἐπιτίθεσθαι (middle) is used with the dative to mean to attack
(e.g. Aristophanes, Wasps 1029 (1024), ἀνθρώποις ... ἐπιθέσθαι),
less frequently with the genitive of the infinitive, as here; so Gen.
43.18 (with both constructions), ... ἐπιθέσθαι ἡμῖν τού λαβεῖν
ἡμᾶς ... The infinitive is epexegetic (BDR § 400.8, n. 10). Cf. 7.19.
The general sense of the second διότι clause is clear: Paul is to
continue his ministry without fear of opposition, for, as will appear,
there are in Corinth many who are, potentially and by predestination
(cf. 13.48), the Lord’s people, and it is therefore impossible that
Paul’s work should be in vain. Grammatically the sentence is two-
fold. Luke could have written λαός ἐστί μοι, there is to me a people,
I have a people; he incorporates πολύς in a predicative sense: there is
to me a people, and a large one. λαός is here the people of God; and
it includes both Jews and Gentiles.
11. ἐκάθισεν δέ (D h (syp syh**) have καὶ ἐκάθισεν ἐν Κορίνθω):
elsewhere in Luke-Acts (except at Lk. 24.49) the verb when used
intransitively means to sit. The meaning to reside is uncommon in
Greek (see BA 791). Elsewhere this meaning could be due to the
ambiguity of the Hebrew but it is hard to see how Hebrew could
have affected the present passage. The aorist ἐκάθισεν is constative,
Paul’s residence being regarded as a unit (Μ. 3.72; Zerwick § 253;
BDR § 332.1, n. 2). Cf. 28.30.
ἐνιαυτόν ( sy add ένα, unnecessarily) καὶ μήνας ἕξ. For the
dating of Paul’s residence in Corinth see above on v. 2 and below on
ν. 12; it seems to be implied (though it is not explicitly stated) that if
not the whole at least the greater part of the 18 months fell before the
attempt to bring Paul before Gallio. The ἡμέραι ἱκαναί of ν. 18 are
probably to be taken as additional. It is unlikely that Luke intended or
was in a position to give anything more than an approximation to the
dates of Paul’s movements.
διδάσκων, as v. 13 shows (but Schneider disagrees), must include
mission preaching as well as the instruction of the church. τόν λόγον
τού θεού is one of Luke’s usual ways of summarizing the message of
the Christian preachers. For ἐν αὐτοῖς D has the second accusative
αυτούς, without any substantial difference in meaning.
12. Gallio ( ... dulcem ... Gallionem: Statius, Sylvae 2.7.32),
originally L. Annaeus Novatus, eldest son of Seneca the Orator and
thus brother of the famous writer and philosopher Seneca, was
adopted by Junius Gallio and became L. Junius Gallio Annaeanus.
He had a sufficiently distinguished career (consul suffectus between
53 and 55) but eventually committed suicide. The date of his
proconsulship in Achaea can be determined with some exactness
from an inscription (see Begs. 5.460-4; Background 51f.) found at
 46. PAUL AT CORINTH, WITH RETURN TO PALESTINE. 18.1-23. 871

Delphi and dated by a reference to the 26th acclamation of Claudius

(as imperator). The precise date of this is not known, but it fell
probably in the first half of Claudius’ twelfth year, between 25
January and 1 August 52, possibly at the close of the eleventh year.
This means that Gallio probably became proconsul of Achaea in 51
(summer), possibly, if, as was less usual, he held office for two years,
in 50.
ανθύπατος was the regular Greek equivalent of proconsul (on
proconsul see a note in Tajra 46). Achaea had been a Roman
province since 27 BC and included central Greece, with the islands,
Thessaly, Aetolia, Acamania, and parts of Epirus, with Corinth, as a
colony, its chief town, though Athens and Sparta were also included.
As a senatorial province it was administered by a proconsul.
We have seen that the natural (though not the certain) sense of v.
11 is that Paul had already spent 18 months (or nearly so much) in
Corinth at the time when the Jews attacked him; only a relatively
short period (v. 18) remained. The wording of the present verse does
not show at what point in his proconsulship Paul was brought before
Gallio, but it is not unreasonable to suggest that Paul’s opponents
took an early opportunity of bringing their case before the new
proconsul in the hope that he would take their side. If this view is
correct we may suggest with some probability that Paul appeared
before Gallio in the Autumn—say, September—of 51; this would
mean, if Acts if trustworthy, that he arrived in the Spring—say,
March—of 50; see Romans 4f. These dates cannot be regarded as
certain (Tajra 55, for example, puts Paul’s appearance before Gallio
towards the close of the proconsul’s term, in Spring 52—Gallio
shows that he had had experience in dealing with Graeco-Jewish
rivalries), but they are nearer to a fixed point than any other event in
Paul’s career. They are moreover consistent with v. 2, if we may
accept (see the note) 49 as the date of Claudius’ edict against the
Jews. On the chronology see further D. Slingerland (JBL 110 (1991),
439-49); J. Murphy-O’Connor (JBL 112 (1993), 315-17).
κατεφιστάναι is hapax legomenon in the NT; ὁμοθυμαδόν is a
Lucan word; see on 1.14.
The Jews brought Paul ἐπί τὸ βήμα. It has often been supposed
that the βήμα is to be identified with the podium excavated in the
market place of ancient Corinth, which recalls the rostra in the
Roman forum and was probably a place for the making of public
speeches. E. Dinkier, however (‘Das Bema zu Korinth’, Signum
Crucis 118-33) has shown that this identification is quite uncertain.
The βήμα denotes the place where the judge holds his court, and is
determined by the presence of the judge, not topographically (cf.
25.10, 17). ‘Dass man ab und zu auch im Freien eine Rednerbühne
als βήμα benutzte, ist nicht ausgeschlossen, aber keineswegs als
üblicher Brauch nachweisbar’ (124). ‘Wir wissen nicht und können
872 COMMENTARY ON ACTS

schlechterdings nicht herausfinden, wo der Bericht von Apg lokali-

sierbar ist’ (129).
The verses that follow are of considerable historical importance,
On the legal issues raised see Sherwin-White (99-107). ‘The
narrative ... agrees very well with the workings of cognitio extra
ordinem. It is within the competence of the judge to decide whether
to accept a novel charge or not’ (99f.).
Schmithals writes on vv. 12-17: ‘Er [Lukas] stellt damit seiner
Zeit, einer Zeit akuter Christenverfolgungen, ein Ideal vor Augen,
von dem er hofft, dass es sich wie in der Frühzeit der Kirche auch in
seiner Gegenwart verwirklichen wird’ (157). Were there such acute
persecutions in Luke’s time? Weiser, in agreement with Haenchen
and Roloff, more cautiously: ‘Es hat eine Anklage des Paulus durch
Juden vor Gallio stattgefunden. Sie wurde mit einer Unzuständig-
keitserklärung abgewiesen. Daraufhin verprügelte die judenfeind-
liche Menge den jüdischen Sprecher Sosthenes’ (487).
For Rome’s attitude to Christianity cf. the attitude to Artemis,
illustrated in ND 4.77.
The Western text presents a more vivid picture. In place of τῷ
Παύλῳ καί, D h (syh** sa) have συλλαλήσαντες μεθ’ εαυτών ἐπί
τὸν Παύλον καί ἐπιθέντες τάς χειρας.

13. D h add, before λέγοντες, καταβοῶντες καί. These words add

to the narrative nothing of substance but serve merely to intensify
feeling; a characteristic Western addition.
The Jews accuse Paul of persuading men to worship God παρά
τόν νόμον (these words, placed at the beginning of the sentence,
carry a good deal of emphasis). ‘The question is whether Jewish
residents at Corinth, who presumably were not citizens of Corinth,
could expect the proconsul to enforce their domestic law within the
territory of a community that was a Roman colony’ (Sherwin-White
100). Sherwin-White has taken νόμος in this sentence to be the
Jewish law, noting that Gallio takes it in that way (v. 15, νόμου τού
καθ’ ὑμάς), even though that is not necessarily what the Jews
intended. Their best way of attacking Paul would have been to plead
that he was commending a religion that Romans could not legally
adopt (cf. the charge at Philippi, 16.21; also 17.7), and this may have
been the intention of the words in this verse, though, if it was, Gallio
either misunderstood or deliberately misinterpreted the charge. It
might have been a dangerous though not a legally sound charge, for
at this period proselytism and circumcision were not strictly contrary
to law and did not become so till the time of Domitian (see Tajra
21-3; NS 3.122-3). Circumcision was undoubtedly frowned upon,
and it may not have been known to the authorities that Paul did not
circumcise Gentile converts. It was known that many Corinthians
(v. 8) were being baptized and brought into something that looked
 46. PAUL AT CORINTH, WITH RETURN TO PALESTINE. 18.1-23. 873

like a variety of Judaism; the Jews may have hoped that this would
support a charge that Paul was contravening Roman law. If however
they were referring to their own law they would, it may be supposed,
be invoking the imperial protection of the Jewish religion. For this
see Josephus, Ant. 19.278-91, and especially 19.290 (καλώς οὖν
ἔχειν καί Ιουδαίους τούς ἐν παντί τῷ ύφ’ ήμᾶς κόσμω τά πάτρια
έθη ἀνεπικωλύτως φυλάσσειν—an edict of Claudius). Paul, though
himself a Jew, was a Jew of doubtful practice and was undoubtedly
responsible for a good deal of disturbance in the Jewish community.
It is true that the edicts issued by Claudius had been intended to
protect Jews against interference by non-Jews rather than from
Jewish heretics, but Paul (1 Cor. 9.20) could evidently regard himself
as something other than a Jew, and could from the standpoint of
orthodoxy be considered a menace. Perhaps the Jews wished to prove
that Christianity was so different from Judaism that it could not be
regarded as a religio licita (Stählin 246).
It may be that if Luke was writing as late as the time of Domitian,
when conversion to Judaism was punishable with death or at least
confiscation of property (Dio Cassius 67.14.2), he imported into his
narrative Circumstances that belonged to his day, not Paul’s. Perhaps
‘there was a deliberate and conscious ambiguity in the accusation
which demonstrates a certain cunning, but also the weakness of the
plaintiffs’ case against Paul’ (Tajra 56). According to Calvin (2.137)
the law in question was the Law of Moses.
As the next verses show, Gallio understood the proceedings as
being cognitio extra ordinem; see above, also Sherwin-White (14,
15).

14. ἀνοίγειν τό στόμα should be regarded as one of Luke’s OT

borrowings (e.g. Job 3.1), though the image is so obvious that it is
not surprising that it occurs elsewhere (e.g. Aristophanes, Birds 1719
(1716)). It proves to be unnecessary for Paul to offer any defence
(though he is prepared to do so). Gallio cuts him short and dismisses
the case. His conditional sentence (unfulfilled condition, correctly
formulated with ἄν in the apodosis) implies that no αδίκημα (an
intentional but not necessarily premeditated injury committed con-
trary to law; see especially Aristotle, Ethica Nicomachea 1135b
(8.8); Rhetoric 1.13.16) or <b>ᾳδιούργημα πονηρόν (there were many
kinds of ραδιούργημα—Lucian, Calumniae 20, μυρία
ραδιουργήματα—but see on 13.10; there may be stress on some
kind of deception, or, here, misrepresentation) had been committed.
If there had been such an offence, Gallio says, I should of course
have been forbearing with you. κατά λόγον is probably (Begs.
4.227) of course, but may be in accordance with what is right, or
what is natural; cf. Thucydides 3.39.4; Josephus, Ant. 13.195.
Dictionaries and commentaries, referring backwards and forwards to
874 COMMENTARY ON ACTS

one another, say that ἀνέχεσθαι is a legal and technical term and
means to accept a complaint or charge. Evidence is seldom cited,
and the meaning is not recorded in LS 136f., but see Wettstein
(2.205) on 2 Cor. 11.1.
On the use of ὦ with the vocative see Zerwick § 35, and on 1.1. Its
use in the present verse indicates no warmth or depth of feeling, and
accords with Luke’s use in general, which follows classical custom
(where ὦ is always used unless there is some special reason for
omission) rather than Hellenistic (where the custom is to omit, so that
ὦ becomes the mark of exceptional feeling).
ἄνδρες ’Ιουδαῖοι (D h vg) assimilates to other passages (e.g.
2.14) and is a secondary reading.
There is an interesting survey of the scene in Ehrhardt (Acts 77f.):
‘Here we have the whole case of Judaism in the Roman Empire in a
nutshell; their privileged status which made them approach imperial
dignitaries with very little reserve, and reluctance of even a man like
Gallio, the brother of Seneca ..., to deal with their internal quarrels,
and the open contempt for them on the part of the masses, who took
whatever opportunity offered itself to use violence against them.’

15. The new condition is open (not an unfulfilled condition as in v.

14): if in fact... then ... Gallio makes clear what he takes to be the
truth.
ζητήματα, plural, occurs in A B De E 33 323 614 945 1175 1739
2495 al lat sy co; it conveys a hint of disparagement not suggested by
the singular ζήτημα (in P74 D* Ψ 0120 <a> e). The word occurs only
in Acts in the NT (where also 3/7 of the occurrences of ζήτησις are
found), normally in a pejorative sense; see 15.2; 23.29; 25.19; 26.3.
Luke suggests that small and insignificant disputes, which sensible
people would not consider, are in mind. They are not important
enough to concern a Roman court.
έχετε (D gig) is an improvement, doubtless secondary, on έστιν,
which would be easier if ὑμίν were in the text. This would give the
same meaning as έχετε. ‘If you have disputes ...’.
λόγος is probably talk (contrasted with such acts as are referred to
in ν. 16), and ονόματα words (cf. Plato, Apology 17bc). It is
probably correct to say (Haenchen 515) that νόμου τοῦ καθ’ ὑμᾶς is
equivalent to τού ὑμετέρου νόμου, but it seems to lay some stress on
the fact that the law is Jewish—the law that is current among you but
not in this (Roman) court.
ὄψεσθε (future) αυτοί, You will have to see to it yourselves. This
may be a Latinismi; so, with a query, Μ. 3.86. ‘Magis ea latina
consuetudo: ipse viderit, tu videris: ad me nihil attinet’ (Blass 199). It
occurs on the lips of a Roman official at Mt. 27.24 (ὑμεῖς ὄψεσθε);
otherwise at Mt. 27.4. BDR § 362.2, n. 3 also treats this as a
Latinism, but notes parenthetically Epictetus and Marcus Aurelius,
 46. PAUL AT CORINTH, WITH RETURN TO PALESTINE. 18.1-23. 875

without giving references. Cf. also Vergil, Eclogue 3.108, Non

nostrum inter vos tantas componere lites. See J. L. North as cited on
15.6. The Jews of course did have their own synagogue machinery
for trying cases under Jewish law.
Theology and religious practice are matters for Jews, not Gallio, to
see to: κριτής (Ε Ψ <a> sy sa add γάρ, which gives the right sense, but
the speech is stronger, more abrupt, without it) ἐγώ τούτων (looking
back to the first part of the verse) ού βούλομαι (θέλω (D), I choose
rather than I wish is a classical improvement, but in Hellenistic Greek
the two words scarcely differed in meaning) είναι.

16. ἀπέλυσεν (D* h) is a much weaker variant of ἀπήλασεν.

Gallio had no patience with the Jews; if Luke’s account is correct he
judged that they were wasting the time of his court; that is, either the
charge was purely irrelevant (a matter of Jewish theology), or there
was not even prima facie evidence to support the belief that Paul
constituted a danger to the Empire. Either way, the precedent was a
useful one.
For βήμα see on v. 12.

17. Schille (366) regards this verse as a doublet of v. 16, but all
they have in common is Gallio’s indifference to the Jewish move
against Paul.
Who beat Sosthenes? According to the Western followed by the
Byzantine text (D Ε Ψ 0120 <a> gig h sy sa), πάντες οἱ Έλληνες.
This is probably correct interpretation. Jews were often unpopular;
they were for the moment out of favour and it would be safe to attack
one of them. Cf. the reference to Crispus in v. 8. Ehrhardt (Acts 77f.)
remarks on the inadequacy of Gallio’s police force. Alternatively
πάντες may be determined by the preceding αυτούς; all the Jews (so
explicitly 36 453 pc) beat Sosthenes, presumably because he had
mismanaged the case against Paul, failing even to secure a hearing.
We have not however been told that Sosthenes was in charge of the
proceedings. The reading that leaves πάντες undefined (P74 A B pc
vg bo) is to be preferred; and it may be that the two views mentioned
above should be combined: the Jews beat Sosthenes for his ineffi-
ciency, the Greeks because he was a Jew and out of favour with the
authorities. A Sosthenes is mentioned at 1 Cor. 1.1 as sharing with
Paul in the writing of the letter. He may be the former ruler of the
synagogue, a Corinthian Jew now become a Christian and travelling
with Paul. This identification is by no means certain; Sosthenes is not
an uncommon name (see BA 1596); if however it is accepted and
Sosthenes’ conversion dates from this period both Jews and Greeks
might have felt that they had an additional reason for beating him. It
is conceivable that he succeeded Crispus, or that the synagogue had
more than one ruler.
876 COMMENTARY ON ACTS

ἐπιλαμβάνεσθαι (often used of violent aggression; see ND 4.104

(A. L. Connolly)) is normally followed by a genitive. It would
however be incorrect to take the present sentence as an exception,
since Σωσθένην, accusative, takes its case as the object of the finite
verb ἔτυπτον; see BDR § 170.2, n. 2, and cf. 9.27; 16.19.
A similar point arises in the last clause. The general meaning is
clear: Gallio was not concerned either to persecute Paul or to protect
Sosthenes. The disturbance was evidently on a sufficiently small
scale to justify the view that public order was not threatened and
Gallio considered that it would do no harm if a few angry people
vented their wrath on a Jew. But is the verb ἔμελεν impersonal or
does it have ούδέν as its subject? According to BDR § 176.3, n. 4
ούδέν is the subject with τούτων as a partitive genitive: None of
these things troubled Gallio; so classical usage (many examples in
Wettstein 2.576), and cf. Hernias, Similitude 9.13.6. Radermacher
(26) however regards ούδέν as a strengthened negative, and Moule
(IB 28) compares 1 Cor. 9.9, ‘which makes it just possible that in the
Gallio passage also ἔμελεν is impersonal, τούτων is a Genitive of
Reference, and ούδέν is an Accusative of Respect’. The absence of
περί, on which Moule remarks, is in fact no problem; see LS, s.v.
μέλω A I 4, p. 1100.
There can be no doubt that Luke found Gallio’s attitude a matter of
interest and importance; it would serve as a precedent that Christians
on trial might cite with advantage. For the apologetic motive in Acts
see Introduction, pp. xlixf.
At the end of the verse d (D being erased) has: tunc Gallio fingebat
eum non videre. h (so Clark 369) has et Gallio simulabat [se non
vid]ere. Ephrem (Clark loc. cit.) comments: factus est (probably a
mistranslation of προσεποιεῖτο) quasi non videns.

18. ἔτι προσμείνας suggests the addition of a relatively short stay

(ἡμέρας ἱκανάς is a Lucan expression; cf. 9.23,43; 27.7; and the use
of ικανός with other nouns) to the 18 months of v. 11. Paul was not
driven out of Corinth. ἀποτάσσεσθαι is used only here and at v. 21
(twice in Lk., once in Mk, once in Paul; it is hardly the mark of a
source), τοῖς ἀδελφοῖς are the members of the church in Corinth. In
Acts as it stands this marks the end of the ‘Second Journey’ which
began at 15.35. Vv. 18-23 have however been taken as a doublet of
Paul’s last visit to Jerusalem (considered by Williams 213). Weiser
(502) thinks that 18.18b, 22a, 22c; 19.1b were taken from the
Itinerary, and that the following are Lucan: 18a (stay in Corinth), 18d
(haircutting), 19b-21a (work in Ephesus), 22b (to Jerusalem), 23a
(stay in Antioch), 23c (strengthening of disciples, possibly 23b
(journey). On this basis he concludes (503) that it is wrong to speak
of ‘Second’ and ‘Third’ Journeys; there is one continuous journey.
ἐξέπλει is ‘improved’ by D to ἔπλευσεν (Ec has ἐξέπλευσεν), but
 46. PAUL AT CORINTH, WITH RETURN TO PALESTINE. 18.1-23. 877

the imperfect can be understood in the sense of ‘began the voyage’.

This is very briefly narrated in the following verses, but it is possible
that Luke was working with an Itinerary that conveyed even less
information, the few details being supplied by Luke. He notes that
Priscilla and Aquila (cf. v. 2; Priscilla now takes precedence)
accompanied Paul. They travelled widely: from Rome to Corinth,
from Corinth to Ephesus, whence they send greetings to Corinth (1
Cor. 16.19). 2 Tim. 4.19 may also point to Ephesus. According to
Rom. 16.3 they are back in Rome (unless Romans 16 was directed
originally to Ephesus; see Romans 11, 257f.).
κειράμενος ... τήν κεφαλήν. It is not clear whether this statement
refers to Paul or to Aquila. Aquila is the nearest noun and is in the
appropriate case (nominative), but Paul is the effective subject of the
sentence and of the first singular verb in v. 19 (κατέλιπεν); he is
moreover the character in whom Luke is supremely interested. Page
(201) is confident that Paul is intended; Preuschen (113) and
Ehrhardt (100) that it was Aquila who shaved his head (a sign,
according to Ehrhardt, that he retained legalistic presuppositions).
The man in question shaved his head εἶχεν γάρ ευχήν. This is
usually taken to have been a Nazirite vow; cf. 21.23f. and see Num.
6.1-21, with StrB (2.747-51). The rule was that the hair was cut off
at the end of the period of the vow. Begs. 4.230 guesses that it may
have been customary to cut the hair at the beginning of the period
with a view to not cutting it again till the end, but we have no
evidence for such a practice. On Cenchreae cf. Rom. 16.1; it was one
of the two seaports of Corinth, situated on the Saronic Gulf; the
other, on the Gulf of Corinth, was Lechaeum. Assuming that it was
Paul who took the vow, Conzelmann (107) sees it as part of the
unhistorical picture of Paul as a good Jew, but if Luke thought of it in
this way he threw away his opportunity, for he makes nothing
whatever of the incident, using it rather to account (somewhat
obscurely) for the movements of the missionaries. Such vows were
sometimes taken before a difficult or dangerous undertaking; it is
possible that this vow should be connected with the vision of v. 9 and
the appearance before Gallio (vv. 12-17). But is this how the Paul of
the epistles would have confronted difficulty and danger? Paul
‘shaved his head for no other purpose except to accommodate
himself to the Jews’ (Calvin 2.140). But had he any occasion in these
circumstances to become a Jew to the Jews (1 Cor. 9.20)? ND 1.24
points out that ‘Paul’s vow (Acts 18.18) may appear to reflect his
Jewish background in view of the decision to cut his hair. But it may
be rather a standard Greek cultural reaction to some dream through
which came divine guidance.’ At ND 4.114f. there is a reference to
Juvenal, Satire 12.81f. where sailors after escape from shipwreck
rejoice with shaved head—gaudent ibi vertice raso garrula securi
narrare pericula nautae. Taylor (5.329) somewhat similarly thinks
878 COMMENTARY ON ACTS

that this incident marks the end of a vow taken previously; ‘pas de
voeu de naziréat évidemment’; Greek rather than Jewish.
Cenchreae was the natural port of embarkation for eastward
voyages. This is exemplified by the large number of eastern coins
found there (ND 3.60; 4.139f.).

19. κατήντησαν is in X A B E 33 pc d vgmss syp sa; κατήντησεν,

P74 Ψ 0120 <a> lat syh; καταντήσας, D h; there would be a natural
tendency to focus on Paul and his movements. The verb occurs 9
times in Acts (8 times with reference to journeys), 3 times in Paul,
once in Ephesians: an Acts word. They arrived at Ephesus, here
mentioned for the first time, missed on the outward journey because
the missionaries had been prevented from preaching in Asia (16.6),
of which province it was the chief city and the residence of the
governor. It lay at the mouth of the Cayster (the ruins are now found
inland because of a change in the coastline), and was a city of great
size and importance both commercially and culturally. It contained
the greatest of the seven wonders of the world, the temple of
Artemis. See further on 19.1.
The next words are expanded by the Western text; instead of
κάκείνους, D (614 h syh**) have καὶ τω ἐπίοντι σαββάτῳ εκείνους.
The Sabbath was a likely day for Paul to visit the synagogue. The
sentence might at first seem to suggest that whereas the whole party
arrived in Ephesus only Paul entered the synagogue, αυτός δέ,
however, is to be taken as introducing what is contained in vv. 20f.
That is, Paul left Priscilla and Aquila in Ephesus (there, αὐτοῦ in B
Ψ 0120 <a>, but ἐκει is well supported by P74vid A D E 33 104 326
1241 al and may be original), where they are still to be found in
18.26, but he himself continued his journey. It is clear from Acts and
from other sources that there were Jews in Ephesus, and it would be
amazing if they were to be found in many other parts of Asia Minor
but not here; yet among the many buildings of ancient Ephesus still
remaining no synagogue has been found and the earliest Jewish
inscriptions date from the second century. See NS 3.22,23, where the
literary and other evidence for Jews in Ephesus is given.
In the synagogue Paul διελέξατο (so P74 X AB 33 1739 1891 pc;
D pc bo have διελέγετο, E Ψ 0120 <a> διελέχθη, without substantial
difference in meaning); for the word see 17.2. Paul turned at once to
the Jews in the synagogue; 18.6 refers to Corinth, not to general
policy.
See Weiser’s analysis of this passage, given on v. 18. Many take
the view that Luke inserted the reference to Ephesus into the
Itinerary or some such source; so e.g. Haenchen (521); Schneider
(2.254). Pesch (2.155) thinks that Luke wished to make clear that
Paul was the first Christian to preach in Ephesus. This seems a rather
feeble reason for an insertion; unless a better can be given Paul’s
 46. PAUL AT CORINTH, WITH RETURN TO PALESTINE. 18.1-23. 879

visit must appear fruitless and pointless, and this is against its being a
Lucan insertion.
20. The Ephesian Jews were welcoming, at least to the extent of
wishing to hear more from Paul. They asked him to stay; D E <a> w sy
sams bo add παρ’ αὐτοῖς, unnecessarily. This however was contrary
to his plans; see below, vv. 21-23. He had, according to Acts,
established a foothold and received an invitation he could use in the
future (though Acts does not say that he did), and for the present this
sufficed.
ἐπί is unnecessary; the accusative of duration of time (πλείονα
χρόνον) would suffice, but the use of ἐπί in this sense is old and
widespread; see LS 623, s.v. ἐπί C II.
ἐπινεύειν occurs here only in the NT, but is as old as Homer.
21. ἀποταξάμενος, as at v. 18.
παλιν ἀνακάμψω, pleonastic; see BDR § 484. This reading, of P74
A B E 33 36 945 1739 1891 al vg, is very probably correct; the
pleonasm is avoided by D sa bopt, which omit πάλιν, not by Ψ <a> gig
sy, which have πάλιν δέ. Paul was soon back in Ephesus; 19.1.
τοῦ θεού θέλοντος is a pious formula with pagan as well as
Jewish parallels. The thought is Pauline: 1 Cor. 4.19, ἐάν ό κύριος
θεληση; 16.7; also Heb. 6.3; Jas 4.15. Josephus is virtually within the
biblical tradition: Ant. 2.333, τοῦ θεού θελήσοντος; 7.373. Plato
(with some emphasis) Alcibiades I 135d, ἐάν θεός ἐθέλη; Epictetus
1.1.17. DBS 252 quotes a papyrus letter (BGU 2. 423.18), τών θεών
θελόντων (2nd century AD).
ἀνάγεσθαι (see 13.13) is a regular word for put out to sea, set sail,
and is naturally answered by κατελθών in v. 22.
The verse contains a major Western addition. After εἰπών, the first
hand of D has δεῖ δέ πάντως τήν ἑορτήν τήν ἡμέραν ἐρχομένην
ποιῆσαι εις Ιεροσόλυμα. The corrector changes δέ to με (in
agreement with d: oportet me). This reading with με but without
ἡμέραν occurs also in Ψ <a> gig w sy. d continues in complete
agreement with D: solemnem diem advenientem facere. The words
recall 20.16 (see also 19.1), but unlike that verse do not specify the
feast in question. If we accept the Western text as original, and as an
accurate account of what Paul thought and said, and if further we are
right in the view (see v. 12) that Paul left Corinth in or about
September, he may have been thinking of the Day of Atonement
(though that was not a feast but a fast; cf. 27.9) or of Tabernacles. It
is however likely that the Western editor, who is followed here by the
greater part of the textual tradition, felt himself called upon to
provide a reason for Paul’s hurried journey and refusal to stay in
Ephesus. It may be possible to suggest a better reason; see on the
next verse. It will be necessary to raise at 20.16 the question whether
Paul found attendance at the festivals compulsive. Calvin (2.141)
880 COMMENTARY ON ACTS

thinks that this, like the vow in v. 18, was simply an accommodation
to Judaism. J. Μ. Ross (NovT 34 (1992), 247-9) defends the long
text of D.
A further textual variant runs over into v. 22; see on that verse.

22. Almost all MSS begin the verse with the simple statement that
Paul, having set sail from Ephesus, landed (κατελθών) at Caesarea.
Instead of this, 614 (syp syhmg) have τόν δὲ Άκύλαν εἲασεν ἐν
Έφέσω, αυτός δέ ἀνενεχθεἰς ἦλθεν εις Καισαρείαν. This repeats
what had been said in v. 19 (though now Priscilla is left out of
account) and in v. 21; the reading continues to the same destination
as in the other MSS. It is hard to detect any motivation here; the
Western text (or part of it) offers a rewording of the narrative.
Luke gives no account of the voyage which begins at Ephesus and
ends in Caesarea. It is unlikely that an ancient ship would be able to
make so long a voyage without calling at a number of ports to take on
supplies; if Luke had any information about what happened at any
port of call he suppresses it. It is probable that he had nothing to tell;
Paul may not have wished to linger on the way; see below.
For Caesarea see on 8.40. It lay in Judaea, not in Syria (v. 18).
Luke may possibly have been confused over this; or Paul may have
intended to land at Antioch, where he spent some time (v. 23) on the
return journey. Northerly winds may have made it impossible to put
in at Antioch. On points such as these nothing better than guesswork
is possible, though Conzelmann (107) thinks that had Paul been
compelled by unfavourable winds to travel to Caesarea, Luke would
have said so; he does not say so, so that Caesarea must have been the
intended destination, and a consequent visit to Jerusalem must have
been planned. Perhaps.
ἀναβάς is ambiguous. It may mean simply that Paul went up from
the docks where the ship had berthed and went up into the town and
there greeted (ἀσπασάμενος) the church of Caesarea. Cf. Herodotus
5.100, πλοία μἐν κατέλιπον ... αυτοί δέ ἀνέβαινον ... But
ἀναβαίνειν is also used for going up to a capital or holy place (see on
11.2), and it is therefore possible that, though its name is not
mentioned, Jerusalem may have been the goal of Paul’s journey, ή
εκκλησία was used for the church of Jerusalem ‘long after [this] had
ceased to correspond with the facts’ (C. H. Dodd, NT Studies 58). We
cannot be certain that a visit to Jerusalem took place at this point; if it
did we are left wondering why Luke should have referred to it
without mentioning it explicitly. That Jerusalem is not mentioned is
of course a reason for talcing the view that it was not visited (Clark
369: ‘It is incredible that so important an event [as a visit to
Jerusalem] could have been recorded so briefly’), but Luke’s reti-
cence may have an explanation. It is a reasonable suggestion that
Paul, profoundly disturbed by opposition that had already made itself
 46. PAUL AT CORINTH, WITH RETURN TO PALESTINE. 18.1-23. 881

felt in Galatia and may have begun to affect Corinth and other
centres also, made his way to Jerusalem to find out how far the
Jerusalem authorities were behind the trouble-makers, and if possible
to stop the trouble at its source. This might account for the speed of
his journey—he was not at this point concerned to visit churches en
route but to get to his destination as quickly as possible—and also
for Luke’s silence; he did not wish to dig up old troubles. Roloff
(277) thinks that Luke understood the story in this way, but was in
error; if such a visit had happened Paul would have mentioned it in
Galatians 1 and 2. This however depends on the date of Galatians.
Johnson (335) takes a different view. Paul went to Jerusalem ‘for one
reason only: to assert his continuing fidelity to the original apostolic
community’.
κατέβη reinforces the sense that has been suggested for ἀναβάς.
‘He went up from the harbour into the town of Caesarea and then
went down to Antioch’ does not make as good sense as ‘He went up
to Jerusalem from Caesarea and then went down from Jerusalem to
Antioch.’
For Antioch see on 11.19.

23. ποιήσας χρόνον τινα—at Antioch, as he had not done in

Jerusalem. It is easy to guess, impossible to prove, that Paul was
occupied in straightening out tangled relations with the church; cf.
Gal. 2.11-14. For the expression cf. 15.33.
The sequence of finite verb and participles, ἑξήλθεν, διερχόμενος,
στηρίζων, is discussed by BDR § 339.2a; § 421. διερχόμενος has
something of future significance and adds descriptive intention to
ἑξήλθεν, as if Luke had written ἑξήλθεν καὶ διήρχετο—that was
why he left, στηρίζων, in asyndetic relation with διερχόμενος, is of
less weight, as if Luke had written διήρχετο στηρίζων. Cf. Μ. 3.80,
similarly; but not quite convincingly. The sense requires that pur-
pose, intention, should be seen in στηρίζων rather than in
διερχόμενος. On the meaning Luke sometimes gives to διέρχεσθαι
see on 13.6.
καθεξής is a Lucan word (Lk. 1.3; 8.1; Acts 3.24; 8.4; nowhere
else in the NT). It is not easy to see why it should be used here unless
either (a) it is intended to emphasize that Paul omitted no group of
disciples on his way; but this is already covered by πάντας, or (b) it
is intended to emphasize the order in which the areas were covered,
first the Galatian territory, then Phrygia. This would conflict with
16.6, where the two districts are mentioned in opposite order, though
there too Paul is travelling from East to West. For the two terms see
on 16.6. It is hard not to agree with Haenchen (523): ‘καθεξής zeigt,
dass mit “galatisches Land’’ und “Phrygien’’ zwei verschiedene
Landschaften gemeint sind (gegen Ramsay).’ See however Hemer
(120). See also 19.1. If Paul’s journey to Ephesus had taken him
882 COMMENTARY ON ACTS

through the southern part of the Galatian province he would very

probably have passed through Colossae and Laodicaea, where
however he was not personally known (Col. 2.1—which is valid
evidence whether Paul wrote Colossians or not).
At this point what is conventionally known as the Third Mis-
sionary Journey begins. Cf. Schille (369): ‘Das ist die lukanische
Zäsur zwischen zwei Reisen.’ On this question, and on Paul’s routes,
see Introduction, pp. lviiif.
For ἐπιοτηρίζων (D Ε Ψ 0120 <a>), P74 A B 33 1891 pc have
στηρίζων, probably rightly, but without difference of meaning. The
aim is to make the disciples (μαθητάς here must be, as usual in Acts,
Christian disciples, notwithstanding 19.1) strong in their faith. To
infer from the absence of the word έκκλησία that in this area there
were disciples but no churches makes nonsense of Luke’s under-
standing of discipleship.
 XII
THE MISSION BASED ON EPHESUS
(18.24-20.38)

47. APOLLOS AND THE TWELVE DISCIPLES 18.24-19.7

(24) A certain Jew, Apollos by name, an Alexandrian in origin, an eloquent

man, arrived in Ephesus. He was powerful in the Scriptures. (25) He had
been instructed in the way of the Lord and1 was fervent in the Spirit as he
spoke and taught accurately the things concerning Jesus, though he knew
only John’s baptism. (26) He began to speak boldly in the synagogue. When
Priscilla and Aquila heard him they2 took him in, and expounded the way of
God to him more accurately. (27) When he wished to pass on to Achaea the
brothers encouraged him and wrote to the disciples that they should welcome
him. When he arrived he3 supported those who through grace had become
believers, (28) for he vigorously debated with the Jews, showing publicly
through the Scriptures that4 the Christ was Jesus.
(1) It was while Apollos was in Corinth that Paul5 passed through6 the hill
country of the hinterland and came down to Ephesus, where he found certain
disciples. (2) He said to them, ‘Did you receive the Holy Spirit when you
became believers?' They said to him, ‘We did not hear7 if there is a Holy
Spirit.’ (3) He said, ‘Into what then8 were you baptized?’ They said, ‘Into
John’s baptism.’ (4) Paul said, ‘John baptized with a baptism of repentance,
telling the people that they should believe in the one who was coming after
him, that is, in Jesus.’ (5) When they heard this they were baptized into the
name of the Lord Jesus, (6) and when Paul laid his hands upon them the Holy
Spirit came upon them, and they spoke with tongues and prophesied. (7) The
total number of the men was about twelve.

Bibliography
C. K. Barrett, in FS Reicke, 29-39.
Μ. Black, as in (46).
1NJB, preached with great spiritual fervour.
2NEB, took him in hand; NJB, attached themselves to him.
3NJB, was able by God’s grace to help the believers considerably.
4NJB, Jesus was the Christ.
5NJB, made his way overland.
6RSV, upper; NEB, inland.
7NJB, that there was such a thing as a Holy Spirit.
8NEB, Then what baptism were you given? NJB, Then how were you baptized?
883
884 COMMENTARY ON ACTS

C. Burchard, ZNW 52 (1961), 73-82.

E. C. Colwell, JBL 52 (1933), 12-21.
J. Coppens, Kremer, Actes 405-38.
K. Haacker, NovT 12 (1970), 70-77.
J. H. A. Hart, JTS 7 (1905), 16-28.
G. H. R. Horsley, as in (46).
J. H. Hughes, NovT 14 (1972), 214-18.
E. Käsemann, ZThK 49 (1952), 144-54 (= EVB 1.158-68).
G. D. Kilpatrick, JBL 89 (1970), 77.
J. Murphy-O’Connor, NTS 36 (1990), 359-74.
J. E. L. Oulton, ExpT66 (1955), 236-40.
J. K. Parratt, ΕχρΤ 79 (1968), 182f.; 80 (1969), 210-14.
H. Preisker, ZAW30 (1931), 301-4.
E. Schweizer, EvTh 15 (1955), 247-54 (= Beiträge 71-9).
B. T. D. Smith, JTS 16 (1915), 241-6.
W. A. Strange, NTS 38 (1992), 145-8.
Μ. Wolter, ZNW 78 (1987), 49-73.

Commentary
It would be easy to make out of what is here treated as a single
paragraph two distinct paragraphs, 18.24-28, the story of Apollos,
and 19.1-7, the story of (about) twelve disciples; easy, but mislead-
ing, for the most difficult problems and the most important observa-
tions would be missed. The Apollos story would depict the arrival in
Ephesus of a learned Alexandrian, who, richly endowed with the
Spirit, preached with great fervour and effect. His preaching showed
accurate acquaintance with Christian tradition (the way of the Lord,
the truths concerning Jesus). He had not received Christian baptism
but he had been baptized as a disciple of John the Baptist, and Paul’s
colleagues, Priscilla and Aquila, were glad to teach him the little he
did not yet know and to accept him as a fellow worker who could be
commended to the churches of Achaea. Some eyebrows might be
raised over such readiness to dispense with baptism into the name of
Jesus (or the Trinity), but the tracts of Acts (such as chs. 13 and 14)
that make no mention of baptism will be recalled. The ‘Disciples’
story also can be retold on its own. Paul himself is now back
(18.19-21) in Ephesus and there meets some twelve men who are
described as μαθηται. Unlike Apollos they show no sign of the
activity of the Spirit, and on inquiry profess that they have never
heard of the Holy Spirit; like Apollos they have been baptized with
John the Baptist’s baptism. Paul instructs them briefly in the relation
between John the Baptist and Jesus, baptizes them in the name of the
Lord Jesus, and lays his hands upon them. This done, the Holy Spirit
 47. APOLLOS AND THE TWELVE DISCIPLES. 18.24-19.7 885

comes upon them and his presence is manifested by the charismatic

phenomena of glossolalia and prophecy (cf. 2.4). Read on its own
this story assumes that baptism (with the imposition of hands) is the
indispensable mode of entry into the Christian life and that it and it
alone leads to the operation of the Holy Spirit.
It is however impossible to read either of these stories ‘on its own’,
and it may be assumed that Luke intended each to be read in the light
of the other. When they are so read a parallel and a difference
immediately stand out. In the first we meet a man who has received
John’s baptism and no other; he is given some instruction and then is
not merely received into the church but continues a preaching activity
he has already begun. In the second, there are (about) twelve who are
in the same position; they must receive a new baptism and the imposi-
tion of hands before they take part in Christian activities. Why, the
reader asks, should Apollos be treated differently from the twelve? Is
it because he is already ζέων τφ πνεύματι (18.25)? If so, it seems that
the purpose of baptism is simply to generate charismatic phenomena.
What was the more accurate instruction about the Way that Priscilla
and Aquila imparted to Apollos? Again, in the second story, why are
the twelve described as μαθηταί, a word which in Acts nearly if not
quite always refers to Christian disciples? Were they Christians,
though unbaptized? Or were they disciples of John? How could they
have been unaware of the very existence of the Holy Spirit?
The two paragraphs are united by two themes: the work of John
the Baptist, and the Holy Spirit, inadequate and adequate marks
respectively of the Christian faith. There is good though not over-
whelming evidence for the continuing existence of groups of dis-
ciples of John the Baptist after their master’s death, and the existence
of such groups, if they wished to become Christians, must have
presented the early church with a problem. What was to be done with
them? They were hardly in the ordinary sense members of the
church, but it was impossible to treat as enemies the disciples of the
one whom Christians accepted as the forerunner of the Christ, men
who had received the only baptism that Jesus himself, and no doubt
many of his earliest followers, had received. It is probable that
different answers to the problem were given, and that the two
fundamental ones are reflected in the double pericope with which we
are dealing. Some would hold: All that they need is to be instructed
more fully in what Christians believe about Christ. Others would take
the view that they were essentially like other unbaptized unbelievers
and must enter through the only door universally recognized. Many
discussions of this question (very notably that of Käsemann and to
some extent that of Schweizer—see the bibliography) are vitiated by
the assumption that Luke’s mind worked within a rigid framework
which assumed the absolute necessity of ecclesiastical regulations,
including the requirement of baptism. This is not so; the story of
886 COMMENTARY ON ACTS

Samaria (8.4-25) so far from manifesting a frühkatholisch view of

sacramental admission to the church views baptism and the gift of the
Spirit with very considerable freedom.
This is not to say that Luke has presented us with two historically
verifiable stories, told with objective interest in what really hap-
pened. Precise analysis of the stories (e.g. Weiser 509 thinks that in
vv. 24, 25a Luke uses traditions about Apollos; in 25b about his
Christian preaching in Ephesus; and in 26b the taking of Apollos into
their home by Priscilla and Aquila; whereas Luke added 24d
(Apollos’s arrival in Ephesus), 25c (Apollos baptized by John), 26c
(instruction by Priscilla and Aquila), perhaps 27abc, 28) is uncon-
vincing. It is much more probable that Luke wrote both straight out
on the basis of common talk in Ephesus. It is not to be thought that
Luke put them together in order to inform later historians of the
diverse attitudes to disciples of John in the first century. Haenchen
(534) is probably right: Luke ‘wollte in Kap. 19 ein Gesamtbild von
der erfolgreichen Arbeit des Paulus geben. Dafür trug diese
Geschichte einen wichtigen Zug bei: Paulus überwindet die Sekte.’
Perhaps not ‘überwindet’, but ‘absorbs’.
[I have drawn upon my contribution in FS Reicke, 29-39.]

24. A new paragraph is introduced without connection with the

preceding narrative. There is no reference to Paul whose story will
not be resumed till 19.1, when he is seen in—distant—relation to the
man who is the centre of the present story. The new figure is
connected not only with Ephesus but with the Christians there, and it
must be concluded that there was a pre-pauline church in Ephesus.
We hear nothing of any other founder, but Paul’s brief visit to the
synagogue (18.19f.) would hardly suffice for the foundation of a
church. The work of Priscilla and Aquila might have done so in the
time that the travels of 18.21-23 would take.
The new figure is a Jew, whose name is variously given in different
authorities. The majority of MSS have Άπολλῶς ὀνόματι; 36 453
1175 pc bo have ὀνόματι Άπελλῆς; D has ὀνόματι ’Απολλώνιος.
BDR § 29.4, n. 6 surprisingly write, 'Obschon Άπελλῆς und
Άπολλῶς etymologisch verwandt sind, scheint es sich um zwei
Personen zu handeln, indem Άπολλῶς aus 1 Kor 1.12 usw in Apg
eingedrungen ist; auch die Scholien (Cramer Catenen zu Apg p. 309f.)
scheinen die Verschiedenheit der Personen für möglich zu halten.’
Cramer has in fact one scholion, and it is wrongly taken by BDR. The
conclusion seems highly improbable; a baseless assimilation of the
name given in Acts to that of the person mentioned in 1 Corinthians is
unlikely to have affected very nearly all MSS. We must conclude (not
with certainty but with reasonable probability) that the same person is
mentioned in both books. They do not throw much light on each
other.
 47. APOLLOS AND THE TWELVE DISCIPLES. 18.24-19.7. 887

Apollos then was a Jew: also Άλεξανδρεύς τῷ γένει (D has γένει

Άλεξ., another of those variants that can have arisen only out of the
belief that the precise wording of the original did not matter so long
as the sense was given). As in 18.2 the word γένος cannot refer to
race; it must refer to place of origin and thus of political association.
Little can be made of the reference to Alexandria, of which Acts tells
us nothing except in a variant reading in v. 25. Philo was not a
representative Alexandrian Jew, and it should not be assumed that
Apollos must have been a philosopher and allegorist. If he was
instructed in Christianity in his native city (see v. 25) we can say
nothing about the kind of Christianity he must have learnt. Ehrhardt
(Acts 101, 102) thinks that Alexandria is represented here as
heterodox. ‘Alexandria had rejected the Jerusalem influence, which
the Church at Antioch had accepted.’ This view, like every other
about Christianity in Alexandria in the first century, is a guess, and
has no serious foundation. Luke does state that Apollos was ἀνἠρ
λόγιος. The adjective may mean eloquent or learned; it is fruitless to
inquire which is intended, since in the Hellenistic world education
was to a great extent training in rhetoric. Phrynichus disapproved of
the former rendering (Λόγιος' ώς οί πολλοί λέγουσιν ἑπί τοῦ δεινού
εἰπεῖν ... ού τιθέασιν οἱ ἀρχαῖοι... (176; Rutherford 284)). But the
early translations have eloquent (e.g. vg, eloquens; d gig, disertus).
See Foerster (Weltreich 102f.).
Apollos was also δυνατός ἑν ταῖς γραφαῖς, powerful, pre-
sumably, in his understanding of the Scriptures and in his use of them
in preaching and debate.
Surprisingly placed between ἀνὴρ λόγιος and δυνατός ὢν ἑν ταῖς
γραφαῖς is the statement that Apollos reached Ephesus—which
Paul had left in 18.21.
The name Apollos is ‘virtually unattested outside Egypt’ (ND
1.88; see also Hemer 233f.) Haenchen (528 n. 4) gives various
suggestions about Apollos and Weiser (509) gives a careful analysis
of the story. To Lüdemann (216) he was ‘urchristlicher Pneumatiker’.

25. The most natural way of understanding the first seven words of
this verse is that Apollos had been instructed in the Christian way
and was a Christian. In Acts, Christianity is described as the Way, ή
οδός, at 9.2; (16.17); 18.25, 26; 19.9, 23; 22.4; 24.14, 22; see
especially the note on 9.2. Here it is natural to take κύριος to refer, as
frequently, to the Lord Jesus. ἦν κατηχημένος is a periphrastic
pluperfect. For κατηχεῖv in the Lucan writings see Lk. 1.4, where it
may possibly refer to some more or less formal instruction on
Christian matters, as Acts 21.21 certainly does not—hearsay only,
and that inaccurate. In the present verse instruction must be intended.
D (gig) substitute the word of the Lord for the way of the Lord, and
place the instruction in Apollos’s home territory of Alexandria: ὃς ἦν
888 COMMENTARY ON ACTS

κατηχημένος ἀν τή πατρίδι τὸν λόγον τοῦ θεού. This means (if

accepted) that the Christian message had already spread as far as
Alexandria and had been established there. There is no evidence to
support this, but it is by no means impossible. There was a very
large Jewish community in Alexandria, and in the constant coming
and going between that city and Jerusalem there must have been
some Jews who had accepted and were concerned to spread the new
faith.
This interpretation is rejected by Kosmala (107,338), who here (as
elsewhere) takes the way to be the Essene way. See 1QS 8.13 for an
example of the use of the word at Qumran. This is most
improbable; see Fitzmyer (Essays 282), and above, on 9.2.
Apollos further was ζέων τῷ πνεύματι. This does not in itself
necessarily mean that he was a Christian, but the close verbal parallel
in Rom. 12.11 (τῷ πνεύματι ζέοντες) makes it seem probable, and it
is unlikely that one as interested as Luke in phenomena due to the
Spirit would use ζέων τῷ πνεύματι to mean no more than an
effervescent, lively human spirit (but cf. Blass (201), ‘animo ut
17.16, non spiritu s.’). Moreover, Apollos ἐλάλει (D, ἀπελάλει,
without significant difference in meaning—spoke, spoke forth) καί
ἐδίδασκεν ακριβῶς τὰ περί τού ’Ιησοῦ (so P41vid P74vid A Β Ε Ψ
0120 614 945 1175 1241 1739 2495 al; τοῦ κυρίου, <a>). This is how
Paul spoke in Rome (28.31), and it is impossible to take it as
anything other than a description of an inspired and accurate
Christian teacher.
Yet that inspired and accurate Christian teacher had not been
baptized, or rather, the baptism he had received had been that of John
the Baptist; he had not received Christian baptism. Nor (see v. 26)
did he now do so. This stands in sharp contrast with 19.1-7, in which
Paul baptizes a group of disciples of John the Baptist in Ephesus.
ἐπιστάμενος means being aware of having experience of Weiser
(507) finds the choice of this word significant. ‘Die behutsame Wahl
des Verbums “kennen” ermöglichte ihm, hernach den Mangel durch
Unterweisung beheben zu lassen.’ Whether he had been baptized by
John himself in Jordan, or had been a member of one of the groups of
disciples of John the Baptist which are believed by some (see J. H.
Hughes, NovT 14 (1972), 214-18) to have continued to exist after
the master’s death, is not clear. τό βάπτισμα Ίωάννου is the baptism
John preached and practised; it cannot mean the baptism of Jesus by
John.
The questions that arise out of the text are clear, and increase as we
proceed. Was Apollos a Christian? If he was, how had he escaped
baptism? Why was he not baptized now (v. 26)? Weiser’s explana-
tion is inadequate; no amount of instruction could confer baptism. If
he already knew so much, why was he further, ἀκριβέστερον,
instructed? Can an earlier form of the story be traced? Many answers
 47 APOLLOS AND THE TWELVE DISCIPLES. 18.24-19.7 889

have been given. Apollos was a Jewish Christian (Weiser 507). ‘Ap.
war also gewissermassen jüdischer “Jesus-anhänger” aber noch nicht
Christ’ (Schneider 2.226; cf. Schmithals 172). Unfavourable details
have been added to the picture of Apollos in order to depreciate him
(Käsemann). Baptism was introduced by Hellenistic Christians rather
than by the original Jerusalem disciples (Begs. 4.231). These sugges-
tions should be borne in mind, though none is entirely satisfactory.
26. Thus gifted, Apollos began to preach boldly in the synagogue.
For παρρησιάζεσθαι cf. 9.27, 28: Paul had already spoken boldly in
synagogues. Could this mean, could it in an earlier form of the story
have meant, that he was a Jewish, not a Christian preacher? On
ἤρξατο see Wilcox (125-7). It is unlikely that it here represents the
Aramaic Having reached Ephesus Apollos began to preach
there. It is not said, and is unlikely to be intended, that this was the
beginning of his preaching ministry; he had preached before he came
to Ephesus. παρρησιάζεσθαι corresponds to ζέων τῷ πνεύματι
(ν. 25).
Priscilla and Aquila, whom Paul had left in Ephesus (18.21),
apparently attended the synagogue, and there heard Apollos preach.
They were impressed, but, it seems, with his promise rather than his
achievement. He needed further instruction, so they took him aside
(προσελάβοντο) and set forth to him (ἐξέθεντο; cf. 28.23) the way
of God more accurately. If the way of God and the way of the Lord
(v. 25) were reversed in order we might guess that Apollos knew
already the (Jewish) way of God (for the expression cf. CD 20.18)
and now needed to be taught the (Christian) way of the Lord (Jesus).
But this is not what Luke says; moreover he has already said that
Apollos taught accurately the things concerning Jesus. ακριβώς and
ἀκριβέστερον have an artificial ring and sound contrived. Why did
not Priscilla and Aquila baptize Apollos? Was this feature of the
story for some reason omitted, and did Luke supply the
ἀκριβέστερον instruction, taking up ακριβώς from v. 25 and
improving on it? This is a more probable suggestion than Schweiz-
er’s (Beiträge 78), that ἀκριβέστερον was originally elative (cf.
24.22) but was taken by Luke to be a true comparative, which led
him to infer ἀκριβῶς in v. 25.
D Ψ 0120 <a> gig sy samss reverse the order of names and have Άκ.
καί Πρ. A mark of anti-feminism in the Western editor?
27. Apollos, at present in Ephesus (v. 24), wished to cross over
(διέρχεσθαι can hardly mean more than this here; contrast 13.6) to
Achaea. Whether he intended to travel by sea, sailing across the
Aegean, or to make the long land journey through Macedonia, is not
stated. For Achaea see on 18.12. The brothers (for this designation of
Christians see 1.15 and frequently) encouraged him (for this use of
the middle of προτρέπειν cf. e.g. Josephus, Ant. 12.166, τοῦ
890 COMMENTARY ON ACTS

πρεσβευτοῦ προτρεψαμένου) and wrote to the disciples (in Corinth,

we may guess, in view of 19.1). This was a commendatory letter,
συστατικὴ ἐπιστολή; such letters were known in the NT world; cf. 2
Cor. 3.1; Rom. 16.1; Col. 4.10; and see ND 1.64-66. The aorist
participle (προτρεψάμενοι may be antecedent to έγραψαν, or
contemporaneous—they encouraged by writing; see BDR § 339.1, n.
3. In place of the infinitive άποδέξασθαι the Western text (see
below) has ὅπως with the subjunctive; see BDR § 392.1, n. 5.
Marshall (304) writes ‘RSV says that they encouraged Apollos to go
to Achaia, but the text could also mean that they encouraged the
disciples at Corinth to welcome Apollos.’ This is undoubtedly
correct; but encouragement of the newly instructed preacher and the
furthering of his plans is perhaps more natural.
Having arrived (again we may guess) at Corinth, though other
Achaean towns cannot be excluded, Apollos gave support (for this
use of the middle of συμβάλλειν cf. Epictetus 3.22.78, Όμηρου
πλείονα τῇ κοινωνία συνεβάλετο Πρίαμος ...) to those who had
believed, οἱ πεπιστευκότες, the perfect participle implying that
having once come to believe they continued to do so—but Acts also
uses the aorist, οἱ πιστεύσαντες, without implying that those so
described were no longer believers.
διὰ τής χάριτος is used as at 15.11. That they became and
continued to be believers was due only to the grace of God. Luke
does not develop the theme of grace as Paul does, but makes it quite
clear that faith comes through divine not human initiative. It would
be possible to take διά τῇς χάριτος not with πεπιστευκότες but with
συνεβάλετο; Calvin (2.146) does so. But it seems better to take the
adverbial phrase with the nearer verb.
The verse appears in a different form in D (with some support
from P38 and syhmg). It runs: ἐν δέ τή Έφέσω ἐπιδημοῦντές τινες
Κορίνθιοι καὶ ἀκούσαντες αυτού παρεκάλουν διελθεῖν συν
αῦτοῖς εις τήν πατρίδα αὐτῶν. συνκατανεύσαντος δέ αυτού οἱ
Έφέσιοι ἔγραψαν τοῖς ἐν Κορίνθω μαθηταῖς ὅπως ἀποδέξωνται
τον ἄνδρα. δς ἐπιδημήσας εις τήν ’Αχαίαν πολύ συνεβάλλετο ἐν
ταῖς έκκλησίαις.
There is more here than a certain indifference to the exact
reproduction of words. The initiative is taken not by Apollos but by
Corinthians who happen to be resident in Ephesus and have heard
Apollos preaching. They were travelling home to Corinth—on
business?—and urged Apollos to travel with them. He agreed
(συνκατανεύσαντος; the word occurs nowhere else in the NT), and
the Ephesians backed up their Corinthian visitors with a commenda-
tory letter. Apollos took up residence in Achaea; this presumably
means that though he may have had a base in Corinth he travelled
over the province. This is supported by the plural ἐν ταῖς
έκκλησίαις—not only in the ἐκκλησία of Corinth but in other
 47. APOLLOS AND THE TWELVE DISCIPLES. 18.24-19.7. 891

Achaean Christian centres too. Or were these the house-churches

which seem to have been forming in Corinth (1 Cor. 1.1 1f.)?
Metzger (467f.) points out difficulties in the Western readings, but
does not discuss the question where or how they originated.
‘Nowhere else in Acts do we hear of members of one church acting
in another church, nor do we ever hear of an invitation to an apostle
or evangelist to come to a church (16.9 is not a parallel)' (468). It is
probably correct that these things are more likely in the more
developed situation of the second century than in the first. ‘If
Apollos’s visit is made on his own initiative, an introductory letter
recommending him to the Corinthians is appropriate; if, on the other
hand, he goes at the invitation of members of the Corinthian church,
why is it necessary that the Ephesians supply such a letter?' The
difficulty is not as great as Metzger suggests. Logically, the letter
seems unnecessary, but such actions are not always in strict accord
with logic; the Corinthians who had taken up temporary residence in
Ephesus would probably be only a small proportion of the whole
church, and might not be trusted by all—the church at Corinth was
notorious for its divisions. Here too, however, it may well be that we
see a glimpse of second-century rather than first-century church life.
There may also be a connection between the Western editor and
Corinth, where traditions about Apollos may still have been current
in the second century.

28. εὐτόνως (2 Macc. 12.23; 4 Macc. 7.10; Aristophanes, Plutus

1095), elsewhere in the NT only at Lk. 23.10: vigorously—often in a
physical sense, here determined by the verb διακατηλέγχετο, which
occurs here only in the NT. The double compound is quoted by LS
(397) only from this passage; ‘a very strong word’ (Page 203). Μ.
2.301f. says that it ‘might be taken as a sort of double perfective, a
combination of διελέγχομαι confute and κατελέγχομαι convict; but
Blass is probably right in classing it with διαλέγομαι’. Of this word
(with διαλαλέω, διαλογίζομαι, διερωτάω) Moulton says that it
recalls (by means of διά) the ‘two parties in a conversation’.
Apollos confuted the Jews energetically and δημοσία, publicly.
Cf. 16.37. The confutation naturally took the form of a scriptural
demonstration (ἐπιδεικνύς; cf. 9.39, demonstrating in a different
sense) that the Christ was Jesus. It is to be expected that the noun
with the article should be the subject (see Μ. 3.182-4; Moule, IB
115f., both referring to E. C. Colwell, JBL 52 (1933), 12-21; also
Zerwick § 172), and this makes good sense. Jews agreed with
Christians that there was, or was to be, a figure described by the term
Χριστός; what Apollos had to prove from Scripture was that this
figure had appeared in the person of Jesus. Luke finds it unnecessary
to specify the Scriptures used.
After δημοσία there are two different Western insertions, bearing
892 COMMENTARY ON ACTS

clear witness to a stage at which the wording of the text was handled
with some freedom while the meaning was retained. D 614 add
διαλεγόμενος καί (P38 has διαλ. only); this seems unnecessary after
διακατηλέγχετο. E adds καί κατ’ οἶκον, possibly recalling 20.20
rather than picturing private discussions. There is no indication
whose οἶκος might be used. At the end of the verse D has τόν
Ίησοῦν εἶναι χριστόν; P38vid has χριστόν είναι Ίησοῦν. The former
reading requires, the latter permits, the rather more obvious ‘ ... that
Jesus was (the) Christ’.
Knowledge of this verse may be reflected by Acts of Thomas
59(56) (L.-B. 2.2.176), ... ὑποδεικνύων ὅτι oὗτός έστιν Ιησούς ό
χριστός περί σὖ αἱ γραφαἰ ἐκήρυξαν; but the thought is too
commonplace to constitute proof.

1,2. For Luke’s use of ἐγένετο see Introduction, p. xlvi. Here it is

followed by an accusative and infinitives (ἐγένετο ... Παύλον ...
ἐλθεῖν ... καί εὑρεῖν ...), which is by no means an exact
reproduction of the Hebrew construction with It must be classed
as a Septuagintalism, part of Luke’s ‘biblical’ style. See Zerwick (§
389). Radermacher (149), however, takes the infinitives as subject:
‘was passierte, wird durch den Infinitiv ausgedrückt.’ We might
render crudely, ‘Paul’s coming to Ephesus happened’. It seems more
probable that Luke is telling his story in a way intended to recall the
OT.
According to 18.27 Apollos had gone to Achaea; it is not
surprising that he should find himself in Achaea’s largest city,
Corinth (the Western text of 18.27 says explicitly that he went there).
For his activity in Corinth see 1 Cor. 1.12; 3.4, 5, 6, 22; 16.12; and 1
Corinthians 8-11, 43-6, 81-7, 104-8, 391f. Of all this Acts has
nothing to tell except his general assistance to the believers and his
disputes with the Jews (18.27, 28). As the subject of είναι his name
appears in the accusative, which is given as Άπολλῶ by most
authorities, as ’Απόλλῶν by P74 Ac L 33 pc, and as Άπολλήν by
36 453 1175 pc bo (for this variant see 18.24). On the declension see
Μ. 2.121.
While (expressed by ἐν τω and the present infinitive; see BDR §
404.1, n. 2) Apollos was in Corinth Paul passed through (διελθόντα,
possibly preaching as he went; see on 13.6) the upper regions, τα
ἀνωτερικὰ μέρη (vg, superioribus partibus). The precise meaning of
this phrase is uncertain. The adjective is rare, and is not used
elsewhere as a geographical term. So far as we know it is attested
only for medical writers, but if this proves anything about Luke it
proves that he was not a doctor, for he was not thinking of medicines
delivered by mouth or of emetics. The adjective may be taken in the
most literal sense to refer to hill country or it may refer to the
hinterland (of Ephesus). άνω is used geographically (e.g. Herodotus
 47. APOLLOS AND THE TWELVE DISCIPLES. 18.24-19.7. 893

1.177, τά μὲν νῦν κάτω τής ’Ασίας . . . τὰ δέ άνω αυτής.). This

suggests with some probability the meaning of τά ἀνωτερικά μέρη
here. Paul was said at 18.23 to be passing through τήν Γαλατικήν
χώραν καὶ Φρυγίαν; the present verse takes up the same journey and
will refer either to the same territory or, more probably, to the
country between Phrygia and Ephesus. Paul was unknown to the
churches of Colossae and Laodicea (Col. 2.1) and therefore probably
did not use the route that follows (more or less) the line of the
Meander but a more northerly one. The route through the Cayster
valley was shorter and would also make possible the use of
ἀνωτερικά in both available senses: the hinterland was elevated. Cf.
Hemer (120): ‘τὰ ἀνωτερικά μέρη are plausibly understood to refer
to the traverse of the hill-road reaching Ephesus by the Cayster
valley north of Mt. Messogis, and not by the Lycus and Maeander
valleys, with which Paul may have been unacquainted.’
The nature of the hinterland makes κατελθεῖν a suitable verb (it is
read by P74vid A E Ψ 33 945 1739 1891 pc) but does not make
ἐλθεῖν (B <a> lat) unsuitable. The compound verb may have been
introduced in order to match άνωτερικά.
Paul came to Ephesus. When Rome took over the kingdom of
Attaius III in 133 BC Ephesus was already a large and notable city
and as part of the Empire it continued to expand in wealth and in
architectural splendour. It was the residence of the proconsular
governor of the province Asia, though not the titular capital. Further
details of its government will be mentioned later in this chapter. For
its fame as a centre of magic arts see on 19.18f. In Ephesus Paul
found τινὰς μαθητάς. The word itself (see on 9.36) strongly suggests
Christian disciples, but the content of the narrative has led some
(from Chrysostom onwards) to question this. Pesch (2.165) thinks
that the fact that the men appear to be separate from the synagogue
suggests that they are Christians, but in fact they are not. Kosmala
(106f.) and others take them to be disciples of John the Baptist (as the
context in part—v. 3—suggests). Schneider (2.263), Bruce (2.363),
Blass (203), Beyer (116) think that they are Christians; and there are
intermediate views. Marshall (306) (referring to K. Haacker, NovT
12 (1970), 70-77) thinks that they cannot be Christians and notes
that ‘Luke is not saying that the men are disciples but is describing
how they appeared to Paul.’ Stählin (253) prudently observes that
there were ‘mancherlei Zwischenstellungen zwischen Täufer- und
Jesusgemeinde’; but where is the evidence for these Zwischen-
stellungen? and how indeed can they be conceived? Weiser (515)
says that the men were not Christians but ‘auf der unmittelbaren
Vorstufe zum christlichen Glauben’. Again, there is some difficulty
in defining this Vorstufe. The problem will be considered below.
It is not clear why Paul should immediately ask whether these
disciples received the Holy Spirit when they became believers (the
894 COMMENTARY ON ACTS

aorist participle πιστεύσαντες is to be taken as coincident in time

with the finite verb ἐλάβετε; see Μ. 1.131). Paul does not ask if they
believe; this he accepts because they are Christians—so Preuschen
(115). It may be (in Luke’s understanding of the matter) that there
were no charismatic manifestations such as speaking with tongues and
prophesying; contrast v. 6. Luke (though not Paul) might think of this
as a sign that the Holy Spirit was not present. Schmithals (174) thinks
that the question is constructed so as to lead to the answer given.
The direct question is introduced by εἰ, which normally introduces
indirect questions. Cf. 1.6. ‘Der Gebrauch ist unklassisch, findet sich
dagegen auch in LXX (Gen 17.17 us), ist also wohl Hebraismus als
Übersetzung (neben μή) von hebr. das in der indirekten Frage εἰ
entspricht, und das auch direkte Fragen einleitet’ (BDR § 440.3,
n. 5). See however 1.76.
The disciples’ reply is surprising: We have not even heard if there
is a Holy Spirit. Is it conceivable that Christian disciples should have
said this? Is it conceivable that disciples of John the Baptist should
have said this? According to Acts itself, Jesus had promised that the
disciples would be baptized with the Holy Spirit (1.5; 11.16) and
John had foretold this (Lk. 3.16). Readers of the OT must have been
aware of the existence of the Holy Spirit. Bengel (463): ‘Nam neque
Mosen neque lohannem Baptistam sequi potuissent, quin de Spiritu
sancto ipso audissent.’ Haenchen (530) comments with an exclama-
tion mark: ‘Der heilige Geist der Christen ist ihnen völlig unbe-
kannt!’ Marshall (306) quotes Rom. 8.9 and concludes that the
disciples cannot be Christians. It must be asked however whether
Luke in this passage and Paul in Romans 8 mean precisely the same
thing by the Holy Spirit. The variant (instead of έστιν), λαμβάνουσίν
τινες (P38 P41 D* syhmg sa) was no doubt introduced in order to obviate
the difficulty. On the construction (οὐδ’ εἰ ... ἠκούσαμεν) see C.
Burchard in ZNW 52 (1961), 73-82—εἰ is used for that. Page (203)
explains the substance of the disciples’ reply with the suggestion that
their words mean, ‘We did not at our baptism hear whether there is a
Holy Spirit’; that is, 'Our baptism (John’s baptism) was simply a
baptism of repentance and conveyed no promise beyond that of the
forgiveness of sins.’
There are further variants to be observed in these verses (see W. A.
Strange, NTS 38 (1992), 145-8). The opening sentence, ἐγένετο ...
Έφεσον, is given in P38vid D syhmg as follows: θέλοντος Παύλου
κατά τήν ιδίαν βουλήν πορεύεσθαι εἰς Ιεροσόλυμα εἶπεν αὐτῷ τό
πνεύμα ὑποστρέφειν εις τήν ’Ασίαν, διελθὼν δέ τά ἀνωτερικά
μέρη έρχεται εις ’Έφεσον. In addition to a good deal of verbal
resemblance there are two major differences: Apollos and his
residence in Corinth disappear, and we hear of a change of plan on
Paul’s part—he wishes to return to Jerusalem (we should probably
infer that the writer did not suspect, as we have done, a visit to
 47. APOLLOS AND THE TWELVE DISCIPLES. 18.24-19.7. 895

Jerusalem at 18.22) but is directed to Asia and comes to Ephesus.

The dropping of Apollos is relatively easy to understand. His work in
Corinth contributes nothing to the new paragraph apart from the fact
that it explains why Paul did not meet him in Ephesus. The account
of Paul’s overruled plan underlines the fact that Paul’s plans are not
made κατά σάρκα (cf. 2 Cor. 1.17) but under divine direction.
Metzger (469) quotes B. Weiss (Der Codex D, p. 94, n. 1): ‘the
whole antithesis between ἰδία βουλή and an order of the Spirit is
neither in the character of Paul nor of Luke, who brings expressly
into prominence how Paul allows all his decisions to be made by the
will of God made known to him through the Spirit.’ But this seems to
be exactly what happens in the text before us, as it does also e.g. at
16.6, 7: They passed through Phrygia and Galatic territory having
been forbidden to speak the word in Asia [as they evidently would
otherwise have done] ... they tried to enter Bithynia [this was
evidently their intention] but the Spirit of Jesus did not permit them
to do so. The Western text seems here to be quite in the spirit of the
Lucan narrative, but this of course does not establish its originality;
the old Lincial text also is Lucan. The very fact that Lucan thought
and language are used in the Western text is in a sense an argument
against its authenticity; had the reading stood in the text from the
beginning it would have seemed—it would have been—a passage
that no one would have wished to remove. G. D. Kilpatrick (FS
Greeven, 193) thinks that ‘we may avoid some of our difficulties by
assuming that part of the original text survives in WT [= Western
text] and part in GT [= General text]’. He sees further difficulties
however and takes the suggestion no further.
A further variant spans vv. 1, 2:
καὶ εὑρεῖν τινας μαθητὰς εἶπέν τε πρός αυτούς: P74vid A B
33 36 453 945 1175 1739 1891 al vg co
καί εὑρών τινας μαθητάς εἶπεν πρ. αὐ. D Ε Ψ <a> gig syh boms
καί εἶπεν τοῖς μαθηταῖς P38vid
The second is the neater sentence; probably to be regarded as an
‘improvement’ on the first.

3. The surprising reply of the disciples leads to a further question.

εις τί οὖν ἐβαπτίσθητε; οὖν ‘presupposes that if they had been
baptized into the name of Jesus, they would have received the Spirit
at Baptism’ (Knowling 403). What sort of baptism can you have had
if you profess ignorance of the Holy Spirit? The answer is, (We were
baptized) εἰς τό Ίωάννου βάπτισμα. εις τί and εις βάπτισμα are
unexpected. When εις is used with βαπτίζειν it is usually followed
by the name of a person, normally in the NT Christ (or some variant
or fuller form of the name). It is probable that the expressions used
here presuppose the usual baptismal formula and were used because
896 COMMENTARY ON ACTS

it was impossible (or was thought by Luke to be impossible) to say,

We were baptized into (or into the name of) John (Conzelmann
110f.). Zerwick (§ 101) thinks that εις is used as if it were
instrumental ἐν, but in fact the preposition requires no further
definition than that of εἰς and ἐν with ὄνομα and ονόματι.
But if these men were Christian disciples, why had they received
only John’s baptism? According to Wettstein (2.580) they were
Christians ‘quos Apollos docuerat’—had taught, presumably, before
Priscilla and Aquila took him in hand (see vv. 24-8).
Variants in this verse again suggest freedom in copying a text that
was respected because of those of whom it told but was not regarded
as canonical and sacrosanct. Thus at the beginning of the verse εἶπέν
τε is read by B 36 453 614 1175 pc d; εἶπέν τε πρὸς αυτούς by <a>
vgmss (syp sa); ὁ δέ εἶπεν by P41vid P74 A E 33 pc bo; εἶπεν δέ by D
Ψ 945 1739 1891 (2495) pc; ὁ δὲ Παύλος πρὸς αυτούς by Ρ38. The
disciples’ answer is introduced by εἶπαν in the majority of MSS, by
ἔλεγον in P45 D.

4. One might have expected Paul to reply to the disciples, John’s

baptism, which you have received, was a baptism with water; the
new baptism, which was promised by John and is now given in the
name of Jesus Christ, is a baptism with the Holy Spirit (Lk. 3.16;
Acts 1.5). This will make good the deficiency implied by your words.
In fact nothing more is said about the connection between baptism
and the Holy Spirit (though the Spirit is in due course given; v. 6);
instead Paul takes up another aspect of John’s prediction. John
ἐβάπτισεν (one would have expected the imperfect; the aorist is
constative and sums up John’s ministry) βάπτισμα μετανοίας. This
is the only place where the noun βάπτισμα is used as cognate
accusative with the active verb (for the passive cf. Mk 10.38, 39; Lk.
7.29; 12.50). Elsewhere in Acts John preaches, proclaims a baptism
(10.37; 13.24). John’s ministry was a summons to repentance (Lk.
3.3, 8), and this was expressed in baptism. Christian baptism also
included repentance and led to the forgiveness of sins (Acts 2.38).
John pointed forward not only to a new baptism with the Holy
Spirit but to a Coming One, who would be the agent of the new
baptism. So Lk. 3.16: έρχεται δέ ό ισχυρότερος μου, οὗ οὐκ εἰμὶ
Ικανός λῦσαι τὸν Ιμάντα τών υποδημάτων αυτού. John continues
with the promise that this Stronger One will baptize with the Holy
Spirit and fire, but does not expressly require faith in the Stronger
One. It may however reasonably be said that this is implied; only
those who have faith in him will accept his baptism. The clause in
which the requirement of faith (‘ἵνα πιστευσ. pro πιστεύσαι’, Blass
203) is made is oddly expressed. The order of words ‘is strange,
considering how natural and easy it would have been to write λέγων
τῷ λαῷ ἵνα πιστεύσωσιν εις τόν ...’ (Moule, IB 169). It is not a
 47. APOLLOS AND THE TWELVE DISCIPLES. 18.24-19.7. 897

simple matter of the prolepsis of the subject of the subordinate clause

(Μ. 3.325), which is not uncommon in Greek. It may be that we must
be content to note another example of the lack of revision that
appears in a number of passages in Acts.
The last five words of the verse may be read in different ways,
which illustrate the problem of the passage as a whole. They may be
either : the Coming One—that is, of course, as you, being Christians
though unbaptized know, Jesus; or: the Coming One, who, I now
inform you disciples of John, is to be identified with Jesus, to whom
you should now, in accordance with your teacher’s word, transfer
your loyalty.
The name Jesus was too simple for many copyists, though it is
undoubtedly the best reading (P38 P74 A B E 614 2495 pc vg
syh samss bo); D r have Χριστόν; has τὸν Χριστόν Ίησοῦν; ψ 945
1175 1739 1891 pc gig syp samss have τὸν Ίησοῦν Χριστόν.
5. Paul had not said that the ‘disciples’ should be rebaptized but
that John had bidden them believe in the one who was to come;
believing in him however could be expressed by being baptized in
his name, and the rite was duly carried out.
After τοῦ κυρίου ’Ιησοῦ the Western text (D 614 syh**; also P38
but without Χριστού) makes the pious but secondary addition
Χριστού εις ἄφεσιν αμαρτιῶν. Schille (377) regards this as ‘litur-
gisch gerundeter’, but it makes little sense since the baptism the men
had already received was εις ἄφεσιν αμαρτιών.
A curious interpretation of the verse is given by Barth (CD 4.4.62,
75), who takes the subject of ἀκούσαντες ἐβαπτίσθησαν to be the
crowds who listened to John and were baptized by him. His baptism
was, by anticipation, baptism into the name of Christ. This being so
there was no need for further baptism of the ‘disciples’, and, their
baptism being now removed from the verse, they no more than
Apollos receive a second baptism. Apart from its general improbabil-
ity this is an impossible way of understanding the sentence. Those
who heard and were baptized were the αὐτοῖς on whom Paul laid his
hands, the αυτούς on whom the Spirit came (v. 6), the group of men
who numbered about twelve (v. 7). It is unfortunate that Barth’s
important discussion of baptism should be marred by this piece of
exegesis.
’Manus imposuit Paulus: actum baptismi aliis reliquit’ (Bengel
463). This is a possible but by no means necessary way of
understanding the passive έβαπτίσθησαν.
6. Whether Paul or some others deputed by him carried out the
baptism of the disciples (v. 5) is not stated; Paul himself laid hands
on them. For the relation in Acts between baptism, the laying on of
hands, and the gift of the Spirit, see on 8.17. That Luke intended to
affirm and commend as necessary some supplement to the baptismal
898 COMMENTARY ON ACTS

rite, or for that matter the rite itself, is excluded by the preceding
story. Apollos, who equally knew only the baptism of John (18.25),
was not baptized again, nor were hands laid upon him; moreover he
was already ζέων τῷ πνεύματι, though without Christian baptism. It
is probable (see FS Reicke 29-39) that the two stories reflect
different ways of receiving disciples of John the Baptist into the
church. Some would accept them on the ground of some further
instruction; others thought that the new Christian baptism should be
given. The laying on of hands was ‘oratio super hominem’ (see
1.606), prayer directed to an indicated intention. Marshall (308)
thinks that it incorporated the disciples into the fellowship of the
church. ‘... so ist damit aber keineswegs eine Abhängigkeit oder gar
Kontrollierung und regulierung des Geistes von Seiten der Menschen
oder kirchlicher Amtsträger gemeint; denn Lukas hebt an anderen
Stellen das freie, von menschlicher Beeinflussung unabhängige
Wirken des Geistes ... und seine Unverfügbarkeit ... deutlich
hervor’ (Weiser 518). The question whether ἐπιτιθέναι represents
or is unanswerable, and for Luke is probably meaningless.
See on 6.6.
The ignorance of v. 2 is immediately ended by the coming of the
Spirit. For ήλθε, P38vid D (vgmss) have ευθέως ἐπέπεσεν, a variant that
adds at most a little vividness to the narrative. According to Turner
(Insights 20), the article with πνεύμα is anaphoric and refers back to
ν. 2, 'that very Holy Spirit came upon them’. This is no doubt true,
but should not be pressed lest we come to the conclusion that Luke
believed that there was a multiplicity of Holy Spirits.
For speaking with tongues and prophesying see 2.4; for Luke this
was the clearest indication that the Spirit was at work. Cf. 2.4; 10.46.
Whether the Paul who wrote 1 Corinthians would have been satisfied
with this outcome of his work is another matter.
The imperfects ἐλάλουν, ἐπροφήτευον, are inceptive: they began
and continued to do these things.
Instead of ἐλάλουν δέ γλώσσαις καί syhmg has et loquebantur
linguis aliis et cognoscebant ipsi eas, quas et interpretabantur ipsi
sibi; quidam autem etiam prophetabant (Ropes, Begs. 3.181).

7. Luke is interested in numbers, and likes to give them. He is also

aware that he can as a rule give only an approximation, and indicates
this by the use of ὡσεί (as here) or ώς.
οἱ πάντες ἄνδρες means the whole company of the men, the sum
total. So Zerwick (§ 188): ‘Attributive positum dicit [πᾶς] rei vel
rerum complexum, summam (in oppositione ad partem) e.g. ...
“summa virorum erat duodecim” A 19.7 (cf. 27.37).’ Similarly
Moule, IB 94; Μ. 3.201; BDR § 275.3.
There is no reason to think that Luke saw any special significance
in the number twelve; most think that the ὡσεί proves this, but
 47. APOLLOS AND THE TWELVE DISCIPLES. 18.24-19.7. 899

Williams (220) thinks that the ‘about’ can be disregarded. ‘Does this
section point to the existence of a primitive “college” of Twelve at
Ephesus, recognized perhaps by Paul, who governed the Church
there?’
Kosmala, whose view of the ‘disciples’ is mentioned above, adds
in a footnote (116, n. 27), ‘Nach jüdischer (und essenischer) Weise
zählten nur die Männer; aber vielleicht waren die “Jünger”, die
striktere Regel befolgend, überhaupt nicht verheiratet.’ Vielleicht:
but there is no reason to think so; in first-century Ephesus Paul was
more likely to encounter a group of men than a mixed group, or a
group of women, even though the men might be married.
 48. PAUL’S SUCCESSFUL MINISTRY AT EPHESUS 19.8-20

(8) Paul1 went into the synagogue and2 spoke boldly, for three months
arguing and persuading about the kingdom of God. (9) But when some grew
hard and disbelieved, and spoke evil of the Way3, in the hearing of the4
populace, he separated from them and withdrew the disciples, arguing daily
in the school of Tyrannus. (10) This lasted for two years, with the result that
all who lived in the province of Asia5 heard the word of the Lord, both Jews
and Greeks. (11) God performed no common works of power by Paul’s
hands, (12) so that6 sweatbands and sweatcloths were carried from contact
with his skin to the sick, and their diseases left them and the evil spirits went
out.
(13) Some of the7 itinerant Jewish exorcists8 set about naming the name of
the Lord Jesus over those who had evil spirits, saying, T adjure you by Jesus
whom Paul preaches.’ (14) There were seven sons of a certain Sceva, a Jew,
a chief priest, doing this. (15) But the evil spirit answered them, 'I9 know
Jesus and I10 am acquainted with Paul, but who are you?’ (16) and the man in
whom the evil spirit was leapt upon them, overpowered them all, and
mastered them, so that they fled from the11 house naked and wounded. (17)
This became known to all, both Jews and Greeks, who lived in Ephesus; fear
fell upon them all, and the name of the Lord Jesus was magnified. (18) Many
of those who had believed would come, confessing [their sins]12 and
disclosing their magical practices. (19) A good many of those who practised
magic gathered their books together and burned them in the presence of all.
They counted up the prices [of the books]13 and found that they came to
50,000 silver [drachmae]14 (20) Thus the word of the Lord grew mightily and
prevailed.

Bibliography
C. K. Barrett, FS Lohse, 96-110.
K. Berger, NTS 20 (1973), 25, n. 95.
1Greek; RSV, NJB, he; NEB, During the next three months he attended the
synagogue.
2RSV, for three months spoke boldly.
3NJB, in public
4RSV, congregation; NEB, whole congregation.
5NJB, were able to hear.
6RSV, NJB, handkerchiefs or aprons; NEB, handkerchiefs and scarves.
7NEB, strolling.
8NEB, tried their hand at using; NJB, tried pronouncing.
9NEB, I acknowledge; NJB, I recogmze.
10RSV, NJB, know; NEB, know about.
11Greek, RSV, that.
12Greek, RSV, omit.
13Greek, omit; RSV, of them; NJB, of these.
14Greek, omit; RSV, NEB, NJB, pieces.
900
 48. PAUL’S SUCCESSFUL MINISTRY AT EPHESUS. 19.8-20. 901

E. Delebecque, RScPhTh 66 (1982), 225-32.

J. A. Fitzmyer, FS Schneider, 299-305.
G. H. R. Horsley, as in (46).
P. W. van der Horst, ZNW 69 (1978), 187-202.
P. Lampe, BZ 36 (1992), 59-76.
T. J. Leary, JTS 41 (1990), 527-9.
G. Μ. Lee, Bib 51 (1970), 237.
B. A. Mastin, JTS 27 (1976), 405-12.
B. A. Mastin, Bib 59 (1978), 97-9.
F. Miltner, Ephesos. Stadt der Artemis und des Johannes, Vienna 1958.
W. A. Strange, JTS 38 (1987), 97-106.
B. E. Taylor ExpTSJ (1946), 222.
P. Wexler, REJ 140 (1981), 123, 124, 133.

Commentary
This section falls into four parts. The first describes in summary form
Paul’s two-year ministry in Ephesus. It follows the usual pattern.
Paul begins in the synagogue and is able to continue there for three
months before he is obliged to move to a non-Jewish site, discoursing
daily in the school of Tyrannus. The result of his unusually long
period of teaching there is the dissemination of his message through
the whole of Asia. This plain account could well be the entry under
‘Ephesus’ in a Pauline itinerary. Verse 8 reproduces words
(εἰσελθών εις τήν συναγωγήν) used in 18.19; it is quite possible that
all the intervening material was introduced by Luke into the itinerary
from other sources.
The rest of this section, together with 19.21-40, is probably best
understood as consisting of local traditions picked up by Luke in
Ephesus. Their historical value is uneven. Ephesus was a great centre
of magical practices (see below, especially on v. 19), and it is not
surprising that Paul should leave the impression of an opponent of
magic; nor is it surprising that his opposition should be expressed in
various ways—by his appearance as a more striking wonder-worker
than any of his rivals (vv. 11, 12), as one who discomfited the
professional exorcists (vv. 13-17), and as one who banished magic
from the city (vv. 18, 19). The paragraph is wound up with a Lucan
summary. The next incident (19.23-40) is related to pagan religion
rather than magic.
Of the opening sub-paragraph (vv. 8-10) Haenchen (536) writes
that it ‘lässt naturgemäss viele Fragen unbeantwortet, die der Histor-
iker stellen muss’. The questions are in fact not many, and they are
not of great weight. How is it that the Jews did not turn against the
902 COMMENTARY ON ACTS

new faith while Aquila and Priscilla taught in Ephesus (18.26)? Why
did they wait until Paul had been teaching for three months? The
question assumes that the Apollos episode is rightly placed in the
chronology, which is by no means certain, and that after 18.19
Aquila and Priscilla immediately began to teach. We do not know
how these two presented the Christian message and what sort of
provocation their teaching may have caused. It is true that Paul seems
to have left the synagogue at Corinth more quickly (18.6), but he
probably spent most of his eighteen months in Corinth before the
Jews took serious action against him (18.11,12). And how quickly do
people take notice of and react to a new message? At the same rate on
every occasion? There are too many unknowns here for rational
comment. There are questions that a historian may feel bound to ask,
but it is well to recognize that there are some that are not important
and many that cannot be answered because the necessary evidence
does not exist. The three months of 19.8 is probably a guess, or at
best an approximation; and it probably never occurred to Luke to
wonder how things went before Paul arrived. 'Oder eine andere
Frage. Paulus sondert die Jünger ab—hat die christliche Gemeinde
dort ... nicht schon vorher ihre Sonderversammlungen gehabt, bei
denen sie das Herrenmahl feierte?’ (Haenchen 537). Perhaps they
had; but we may say with confidence that such meals as they had
were church suppers which would naturally be held not in the
synagogue but in private houses (cf. Rom. 16.5; 1 Cor. 16.19; Col.
4.15; Phm. 2; and see 1 Corinthians 263, 325; Ch., Μ., S. 65f.). The
school of Tyrannus no more than the synagogue would be suitable
and used for such purposes; and Luke probably thought church
suppers less important than Paul’s preaching and teaching. The
question is pointless. 'Oder noch eine Frage: hat Paulus nach einem
Wirken von drei Monaten die Gemeinde (die er nicht gegründet
hatte!) so souverän beherrscht, dass er ihre Trennung von der
Synagogue verfügen konnte?’ (Haenchen 537). Another ill-thought-
out question. The separation probably received a powerful impulse
from the Jewish side; and it would be foolish to underestimate Paul’s
power of leadership.
Haenchen’s questions do nothing to destroy the historical worth of
vv. 8-10, but it is equally true that this cannot be demonstrated. Luke
probably had before him a reference to a long stay in Ephesus; on
this ground he probably felt it safe to extend the period of work in the
synagogue (beyond for example the three weeks in Thessalonica,
17.2). And he may have had good reason to know that the word of
the Lord had spread pretty widely in Asia, and among both Jews and
Greeks.
It was natural to represent Paul in Ephesus as the foe of magic,
because of the reputation enjoyed by Ephesus and because Luke was
always ready to use an opportunity of expressing one of his favourite
 48. PAUL’S SUCCESSFUL MINISTRY AT EPHESUS. 19.8-20. 903

antagonisms (see 8.9-25; 13.6-12, with the notes). The stories about
Paul’s successful healings (w. 11, 12) and of the marked lack of
success of the sons of Sceva were no doubt told with delight—and
probably with some exaggeration—by Ephesian Christians, and
listened to by Luke with equal pleasure. 19.23-40 is matter of a
different kind, though acquired in the same way. ‘The present
passage [this includes 19.1-7]... compresses into three vivid scenes
the essence of Paul’s prophetic ministry as an apostle and serves to
“legitimate” him firmly in the reader’s eyes as having fulfilled
precisely what was predicted of him’ (Johnson 344).

8. Εἰσελθών δὲ εἰς τήν συναγωγήν. Cf. 18.19, where the same

words are used. The repetition suggests that Luke may here be taking
up again a source that he was using at the earlier point and into which
he inserted material that took Paul to Palestine and back to Ephesus.
There is here no reference back to a previous visit to Ephesus and the
synagogue, to Apollos, or to the disciples of 19.1-7. This verse could
be describing the beginning of Paul’s work in Ephesus. For Jews in
Ephesus see NS 3.22, 23, 88, 122, 123; also Trebilco, 17, 18, 24f.,
167. Cf. Josephus, Apion 2.39.
D syhmg, adding ἐν δυνάμει μεγάλη, underline the power and
effectiveness of Paul’s preaching. No copyist would have omitted
this had it been original. The phrase is constructed adverbially with
ἐπαρρησιάζετο; cf. 9.27. This is an Acts word (7 times; once in 1
Thessalonians; once in Ephesians), used of Paul and at 18.26 of
Apollos.
ἐπί μήνας τρεις. For the use of the preposition cf. 18.20. After
these three months Paul remained in Ephesus a further two years (v.
10); the total is summed up at 20.31 as three years (counted
inclusively). Three months was a long time for Paul to be tolerated in
a synagogue; Pesch (2.167) considers that this synagogue was one
that tolerated Sondergruppen; they had been prepared to accom-
modate the disciples of John the Baptist (19.1-7). For the surprising
lack of evidence of a synagogue community in Ephesus see on
18.19.
ἐπαρρησιάζετο may suggest (it hardly proves) inspired speech;
διαλέγεσθαι (see 17.17) suggests reason: Paul argues and debates.
πείθων belongs to the same realm of discourse. This word also is
characteristic of Acts, though it is often used in the passive of those
who are convinced by Christian argument and believe (e.g. 17.4). For
the active cf. 13.43; 18.4. It usually has a personal object, and here
one must be understood—those who frequented the synagogue,
whom Paul persuaded concerning the kingdom of God, i.e., that it
had been manifested in Jesus; but kingdom of God is a term Luke
uses in Acts as a summary of the Gospel preached by the apostles
and others; see e.g. 20.25. The great majority of MSS have τά περί
904 COMMENTARY ON ACTS

τῆς β. τ. Θ., but B D Ψ 1175 189Γ pc are right to omit the article,
which other MSS have added in order to conform with 1.3.

9. ὡς δέ τινες ἐσκληρύνοντο (D has τινὲς μὲν οὖν αυτών,

introducing a characteristically Lucan μὲν οὖν, but this is probably
due to assimilation to other passages); as usual, Paul’s preaching in
the synagogue brings only limited success and stirs up trouble.
ἐσκληρύνοντο is passive in form but probably intransitive in
meaning; Luke is not thinking of an extraneous hardening agent.
They grew hard; possibly (taking the verb as middle), they hardened
themselves, hardened their hearts against Paul and his message; so
Schneider (2.268). The word is active and transitive at e.g. Exod. 7.3;
Rom. 9.18, and we should note 13.48, but Luke will be more inclined
to blame the recalcitrant synagogue members than a divine decree.
ἀπειθεῖν, for unbelief, is used at 14.2; cf. πείθων in v. 8.
Not content with disbelief they also slandered (κακολογεῖν here
only in Acts) the Christianity preached by Paul so as to give it a bad
name with the public. According to MM 316 the verb ‘in the NT
seems always to be used in the weaker sense of “speak evil of’”. ND
1.28f., but especially 2.88, disagrees, though the papyrus evidence
quoted there is not stronger than that given by MM where the editors’
translations are ‘abused ... in the most unmeasured terms’, ‘insulted
me immoderately’). ND shows that the word is often used in
litigation. The fact is that κακολογεῖν means to speak evil, and only
a context can show how evil evil is. Here there is no illuminating
context apart from the fact that Paul felt it necessary to take decisive
action. For ή οδός as a term meaning Christianity see 9.2. It is not
clear how πλήθος is to be taken. It may refer to (a) the Christians in
the synagogue; for πλήθος as a local community of Christians cf. e.g.
15.30; the effect on them might be to cause them to give up the faith
they had accepted; (b) the synagogue community as a whole, who
might in consequence expel or punish the Christians; (c) the general
public of the city, who would decide not to become Christians and
perhaps to persecute those who were. See Fitzmyer, Essays 290. The
third possibility is perhaps the best. It was adopted by D (E) syp
Syh** which add τῶν ἐθνών. Τότε. See however Ropes (Begs.
3.182).
Paul’s response was to separate (ἀποστάς) from the Jews and to
withdraw the disciples (as usual μαθηταί are Christian disciples)
from the synagogue. Cf. 18.7. Instead of teaching there he continued
his work ἐν τῇ σχολῇ Τυράννου (D, Τυραννίου). For the name, not
uncommon in inscriptions in Ephesus, see Hemer (120f.), who gives
a reference also to a building described by the Latin-Greek word
αὐδειτώριον. σχολή here can hardly mean anything other than a
building, though ND 1.129f. think that it may mean not a place but a
‘group of people to whom addresses were given during their leisure
 48. PAUL’S SUCCESSFUL MINISTRY AT EPHESUS. 19.8-20. 905

hours’. The meaning building is unusual and late (see FS Lohse,

96-110, especially 96f.). It is not clear from their contexts that
Josephus, Apion 1.53 and Epictetus 1.29.34 (cited BA 1591) refer to
places or buildings. Building seems to be correct in Plutarch, De
Recta Ratione Audiendi 8 (42A), where σχολή is used with
διδάσκαλεῖov; not correct in De Curiosiate 9 (579F), where the
parallels are θεάματα, ακούσματα, διατριβαί; not correct in De
Exilio 14 (605A), with the parallel διατριβαί. Cf. Cicero, De Oratore
1.22 (102), ... qui cum in schola assedissent ... For the name
Tyrannus see above; he may have been a philosopher, otherwise
unknown, who lectured in the class-room, or the owner of the
building. The name occurs in Josephus, Ant. 16.314; Ephesian
inscriptions are given in ND 4.186; 5.97. ND 5.92, referring to P.
Wexler, REJ 140 (1981), 123, 124,133, considers the possibility that
σχολή may mean synagogue. This can hardly be the meaning here.
Paul taught here καθ’ ἡμέραν. Roloff (283) assumes that in the
synagogue the Christians could meet only on the Sabbath, so that the
move will have made not only a change of venue but a multiplication
of gatherings and of publicity. Roloff also thinks it must be assumed
that in addition to the lectures in the lecture-room the Christians also
had ‘Mahlversammlungen in den Häusern’ (as in 2.42-46). It could
be so, but these are not in the text. διαλέγεσθαι, as in v. 8.
At the end of the verse D (614 gig syh) add, ἀπό ὥρας έ ἕως
δέκατης, that is, from 11 a.m. to 4 p.m. (or with the greater precision
provided by J. Carcopino, La Vie Quotidienne à Rome à l'Apogée de
l'Empire (1939), 178f. from 11.15 to 2.58 in the winter and from
10.44 to 5.2 in the summer—the figures will of course have varied
daily so that the calculations are worth little). It is probably a correct
observation that these were the siesta hours when the lecture-room
would not normally be in use. In quintain [horam] varios extendit
labores; Sexta quies lassis (Martial 4.8.3f.).

10. τοῦτο δὲ ἐγένετο ἐπί έτη δύο. Cf. Plato, Phaedo 84bc, σιγή
σὖν ἐγένετο ... ἐπί πολὐν χρόνον. The imperfect rather than the
aorist might have been expected (this went on for two years); the
effect of the aorist is to treat Paul’s activity as a single unit; his
ministry lasted for two years. Plato’s ἐγένετο is similar: Silence fell
and lasted ... For ἐπί see v. 8. The two years must be added to the
three months already mentioned (v. 8); taken together they justify the
τριετία of 20.31. The dates are probably Autumn 52 to Spring 55; see
Introduction, pp. lvi-lxi and Romans 4f.
The result (ώστε) was that all who lived in Asia heard the word of
the Lord, spreading outward from the chief city of the province. For
Asia see 16.6; there is no reason to think that anything less than the
whole province is intended. The verse does not assert that Paul
personally proclaimed the word even in all the major cities of the
906 COMMENTARY ON ACTS

province (for his own inclusive and representative way of referring to

the extent of his missionary activity see Rom. 15.19). According to
Col. 1.7; 2.1 Paul had not himself evangelized Colossae; Epaphras
had represented him there. Revelation 2 and 3 are sufficient to show
that leading cities in Asia had been evangelized at a fairly early date;
at the time of writing some at least of the seven churches addressed
were in a state of decline. Luke simply affirms widespread evange-
listic activity, and the affirmation could be based simply on the fact
that Asia as he knew it was one of the most developed Christian
mission fields. He takes the opportunity of adding that the Christian
groups were mixed, comprising both Jews and Greeks. The seven
letters of Revelation probably reflect a later stage when there was
trouble between Jews and Gentiles; see Rev. 2.9; 3.9. Of Paul’s stay
in Ephesus Stählin (255) rightly notes, ‘Es war für Paulus nicht nur
eine Zeit des Wirkens, sondern auch des Leidens (vgl 20.19; 2 Kor
1.8-11; 1 Kor 15.32).’
The latter part of the verse is given by D* (e syp) as follows: έως
πάντες οἱ κατοικοῦντες τήν ’Ασίαν ἤκουσαν τούς λόγους τού
κυρίου, ’Ιουδαίοι καί 'Έλληνες. Most of this is simply a moderately
free paraphrase which does not change the meaning, but the plural
τούς λόγους is unusual. It is probably a slip (corrected in the MS
itself). Zahn thought the reading original (and indeed it is unusual
enough to have been assimilated, if it was original, to the more usual
‘word of the Lord’), and inferred from it—correctly from a theologi-
cal if not from a textual point of view—that the Lord himself spoke
in the words of the preacher.

11. The preaching of the word of the Lord (v. 10) was accom-
panied, as often in Acts (cf. e.g. 4.29f.), by miracles (here δυνάμεις;
the words σημεῖον and τέρας are not used in this context, or
anywhere after 15.12). God himself performed the mighty works,
though διά τῶν χειρῶν Π.; this is a Semitism, due not to translation
but to Luke’s imitation of biblical (i.e. LXX) language. Schille (379)
thinks that there may be a magical reference here; in view of what is
about to be said about Ephesian magic (vv. 13-19) this seems highly
improbable. The mighty works were οὐ τὰς τυχούσας, not any
works that might happen to anyone at any time, but special ones,
uncommon miracles; cf. a second-century AD inscription from
Ephesus, οὐχ ὡς ἔτυχεν, in no ordinary manner (BMI 4.481*.340).
Wettstein 2.580f. gives many examples of the use of the participle;
e.g. Artemidorus 2.13, κινδύνους οὐ τούς τυχόντας; Porphyry, De
Abstinentia 1.7.1; Herodian 2.3.7. It may have been a late develop-
ment. Bruce (1.357) notes that it is common in Vettius Valens.
‘V.11f. zeichnen das Bild des Paulus so, wie es nach den
Forderungen des urchristlichen Pneumatikertums hätte aussehen
müssen’ (Haenchen 540); over against this, ‘Lk setzt natürlich
 48. PAUL’S SUCCESSFUL MINISTRY AT EPHESUS. 19.8-20. 907

voraus, dass es wirklicher Glaube war, der so handelte, nicht Glaube

an den Wundertäter, sondern an seinen Gott’ (Bauernfeind 230).
Weiser (529) rightly emphasizes that it is God who works, but
through man. See further on the next verse.

12. Physical contact between the healer and the sick person is a
common feature of miracle stories; the wonder is heightened here in
that the contact is indirect.
σουδάριον translates the Latin sudarium, σιμικίνθιον, semicinc-
tium. The precise meaning is not known with certainty; no better
suggestion is available than the definition of Ammonius (Fragmenta
in Acta Apostolorum, ad 19.12; MPG 85, 1576 = Cramer 3.316f.),
ἀμφότερα νομίζω λινοειδῆ είναι πλήν τὰ μέν σουδάρια ἐπὶ τής
κεφαλής ἐπιβάλλεται, τὰ δέ σιμικίνθια ἐν ταῖς χερσὶν κατέχουσιν,
οἱ μή δυνάμενοι οράρια [= Latin oraria, (pocket) handkerchief]
φορέσαι. Thus probably both were sweat-rags, σουδάρια worn on
the head to prevent the sweat from running into the eyes, σιμικίνθια
carried in the hand for general mopping up. An alternative possibility
for σιμικίνθιον is apron. In each case the important point would be
contact (this is against belt—T. J. Leary, JTS 41 (1990), 527-9)
with Paul’s skin. χρώς is not common in prose, but is used in the
LXX.
The effect of these objects, presumably applied to the sufferers,
was that their diseases left them (for ἀπαλλάσσεσθαι cf. Pseudo-
Plato, Eryxias 401c, εἰ aἱ νόσοι ἀπαλλαγείησαν ἐκ τών σωμάτων),
and evil spirits went out from those whom they had possessed. It is
customary to point out the parallel with the effect of Peter’s shadow
(5.15f.). The parallel is valid, but it is doubtful whether Luke went
out of his way to draw up a precise but varied balance between Peter
and Paul. He was probably more concerned to claim that Paul could
beat the Ephesian magicians at their own game. It was ‘ein typisches
Beispiel von Mana-Glauben’ (Stählin 255).
Cf. Acta Johannis 62 (L.-B. 2.1.181): Μετά δέ ταῦτα γινόμεθα
εις τήν ’Έφεσον ... Τῶν ποδῶν αὐτού ἁπτόμενοι, καί τὰς χεῖρας
αυτού εἰς τά ἴδια πρόσωπα τιθέντες ἐφίλουν αὐτάς, ὡς ὅτι καν
ἥψαντο τών ἐκείνου ἐνδυμάτων.

13. Paul’s success as healer and exorcist prompted imitation, and

the use by others of the means he had so successfully employed. It is
not stated in vv. 11, 12 but implied by v. 13 that the name of Jesus
had been invoked when the miracles were performed.
ἐπιχειρειν is used by Luke only in the NT (Lk. 1.1; Acts 9.29;
19.13). περιέρχεσθαι (cf. 28.13, si v.1.; Heb. 11.37) means to go
around, but, as 1 Tim. 5.13 may confirm, could (but does not
necessarily) suggest the migration of wandering charlatans. Xeno-
phon, Cyropaedia 8.2.16 suggests the behaviour of a wandering
908 COMMENTARY ON ACTS

beggar; Luke, who had no high opinion of wandering magicians, no

doubt thought of itinerant exorcists in this, or a less favourable way.
See also Betz (142), ‘die wandernden Bettelpriester der Syrischen
Göttin’. Cf. Lucian, Asinus 37, τήν χώραν περιήειμεν. It is here only
that Luke uses the word ἐξορκιστής, but he probably thinks of the
persons in question as belonging to the same class as Simon (8.9-11)
and Elymas (13.6-8). ND 2.11 has interesting evidence of the non-
inclusion of exorcists in a recognized state health service.
For Jewish exorcists see StrB (4.534f.); NS 3.342-79; Hemer
(121). Their existence is presupposed in Mt. 12.27 = Lk. 11.19.
Josephus, Ant, 8.45-49 is good proof of the fact that Solomon’s
legendary power as an exorcist was believed to have been trans-
mitted, through incantations and formulas, to first-century exorcists.
For the use outside orthodox Christian circles of the name of Jesus
see Mk 9.38-41. Rabbinic disapproval of the practice (StrB 1.468)
shows that it existed. For the use in Acts of όνομα see on 3.6; here
only (apart from Eph. 1.21; 2 Tim. 2.19, where the usage is different)
do we have ὀνομάζειν τό ομομα—a pointer to the fact that when
Luke speaks of the proper use of the name by authentic Christians he
is thinking of something different from the incantation of a formula.
The story that follows Luke would regard as a clear proof of his own
understanding of the name: the name itself has anything but the
desired effect (so rightly Conzelmann 111). Again, Christians do not,
in the NT, use the verb (έξ)ορκίζειν. The verb is found in the sense
that it has here in magical papyri (LS 1251; BA 1178; MM 457),
sometimes, as here (ὁρκίζω ὐμᾶς τόν Ίησοῦν), with a double
accusative, sometimes with other constructions. The relation of the
NT narratives to magic is difficult to assess (according to Lüdemann
the story of vv. 13-16 is ‘schon aus formgeschichtlichen Gründen
unhistorisch’, but he does not explain the judgement), but a narrative
such as the present suggests that Luke was aware of a resemblance
between Christian miracle-working and contemporary magic but at
the same time wished to make a fundamental distinction. For the
borrowing of names cf. PGM 4.3019f. (in Deissmann, LAE 252):
ορκίζω σε κατὰ τού θεοῦ τῶν Εβραίων 'Ιησοῦ; Origen, C. Celsum
1.6; 6.40.
ὁρκίζομεν (<a>) and ἐξορκίζομεν (36 453 614 945 1739 1891 pc)
are assimilations to the preceding plurals and to the compound noun
έξορκιστής.
’No rule will account for’ the use of the article with names here
and in v. 15 (Μ. 3.166). ον Παύλος κηρύσσει is simply a means of
identification; for the content of Paul’s preaching cf. 17.18.

14. The general statement of v. 13 is supported by a particular

example, ἦσαν ... ποιοῦντες should not be taken as a periphrastic
tense; ποιοῦντες is predicative (there were seven sons ... doing this;
 48. PAUL’S SUCCESSFUL MINISTRY AT EPHESUS. 19.8-20. 909

cf. 21.23; Lk. 2.8; and see BDR § 353.2b, though this example is not
quoted there). That is, they were saying to demons, ορκίζω υμάς,
κτλ. The construction is changed by Ε Ψ <a>, which add oἱ before
τούτο.
These were the sons of one Sceva, a Jew, a high priest. On the
name see Hemer (234); add Horace, Epistle 1.17.1, Quamvis, Scaeva
... There is no Sceva in the list of Jewish High Priests otherwise
known (see Jeremias, Jerusalem 377f.). What else could ἀρχιερεύς
mean? The word is used in the plural in the gospels and Acts (e.g. 4.1
(v.1. ἱερεῖς), 23) to denote, it seems, members of the Jewish priestly
aristocracy, or of the court that determined issues relating to the
priests and the Temple (Jeremias, Jerusalem 178). On this question
see B. A. Mastin, JTS 27 (1976), 405-12. Rejecting the views of
Haenchen (Luke thought that Sceva really had been high priest—this
had the effect of magnifying Paul), Burkitt (Begs. 4.241, ‘Scaeva, “a
rascally Levantine (real race very uncertain)”, claimed the title “as
an advertisement” ’), and B. E. Taylor (Sceva was a renegade Jew
who held the office of ἀρχιερεύς in the imperial cult) Mastin refers
to Apuleius, Metamorphoses 2.28-30, in which Zatchlas (Zachlas),
an Egyptian, raises a dead man to life. Zatchlas is described twice as
propheta, once as propheta primarius, once addressed as sacerdos.
In a similar way, the description of the father of the seven exorcists
as a priest, whose sons will have been priests too, is consistent with
their work as exorcists, and helps ‘to show how true religion
triumphed’ (409). Cf. K. Berger (NTS 20 (1973), 25, n. 95). For
pagan αρχιερείς in Asia see Mommsen (Provinces, 1.347f.); cf. the
Asiarchs of 19.31. A better parallel however than Mastin’s is to be
found in Juvenal, Satire 6.544, where the fortune-telling Jewess is
described as magna sacerdos. Luke should have known, and indeed
did know, a great deal better than Juvenal what a priest, or chief
priest, was in Judaism; that he uses the word here probably means
that he found it in the tradition, written or oral, that he was using. The
story may not have been connected with Paul (Weiser 524, and
others).
For τινός, τινές is read by P74 A Ψ <a> lat syh, rightly accordingly
to Ropes (Begs. 3.182); it is the sons, not their father, who are being
introduced. According to this view, seven exorcists were involved; in
ν. 16 they are collectively referred to in the word ἀμφοτέρων. See on
that verse. The suggestion that the number came into the text as a
gloss, Σκευᾶ = = ἑπτά, seems very improbable (Μ. 1.246). For
seven gig alone has duo.
The text appears in a different form in (P38) D w syhmg, as follows:
ἐν οἷς καί υιοί (4- ἑπτά syhmg) Σκευᾶ (+ Ιουδαίου) τίνος ἱερέως
ἠθέλησαν τὀ αυτό ποιῇσαι. ἔθος είχαν τούς τοιούτους ἐξορκίζειν
καί εἰσελθόντες πρὸς τόν δαιμονιζόμενον ἤρξαντο ἐπικαλεῖσθαι
τό όνομα λέγοντες παραγγέλλομέν σοι ἐν ’Ιησοῦ ον Παύλος
910 COMMENTARY ON ACTS

κηρύσσει ἐξελθεῖν. The basic D form of this reading omits the two
problem words, seven and in all forms there is priest, not high
priest. See further Clark (370-3); he thinks priest original and
suggests that the number seven may have arisen through a mis-
understood marginal ζ (= ζήτει), a critical sign, or a mark of
interrogation.
It is interesting to compare Testament of Levi 18.12: the sons of
the true high priest will have authority to trample on τά πονηρά
πνεύματα.

15. The variant in D (τότε ἀπεκρίθη ... καί εἶπεν) seems to be an

arbitrary piece of rewriting; it resembles the Syriac but need not be
ascribed either to underlying Aramaic or to a reaction from a Syriac
version.
τό πνεύμα τό πονηρόν is introduced abruptly. Evil spirits (plural)
were mentioned in v. 13; we must suppose that Luke has now moved
on to a particular example of the work of the sons of Scaeva
(themselves already invoked as a special example of the Jewish
exorcists of v. 13), who entered a house (v. 16; cf. v. 14 D) and
attempted to exorcise a single spirit from a possessed man. The spirit
is not to be so easily overpowered and driven out.
The evidence for the inclusion of μέν (P41 B E Ψ 614 2495 pc)
and the evidence for its omission (P38 P74 A D <a> latt sa) are
evenly balanced. On the whole there is a stronger case for omission.
τὀν μὲν Ίησοῦν ... υμείς δέ ... makes a pleasing Greek sentence,
which copyists would be more likely to create than to destroy. This is
confirmed by the fact that the first hand of did not include the
word, whereas a corrector did.
With regard to τόν Ίησοῦν ... τόν Παύλον there seems to be no
explanation of the use of the article; see the reference on v. 13 to Μ.
3.166. It is probably simply a Hellenistic development; Turner points
out that the article is always used with proper names in Modern
Greek. If further meaning is sought in the articles the point may be
‘The great Jesus ... the great Paul ... but who in comparison are
you?’ Cf. Mk 1.34, the demons ᾔδεισαν αυτόν (cf. also Acts 16.17).
In the present verse the verb is γινώσκειν, but no difference in
meaning is to be sought; this is probably true also with regard to
ἐπίσταμαι. According to Weiser (524), for Luke ἐπιστᾶσθαι always
has as object ‘historische Faktoren’, and it is possible that Luke
intends, I recognize Jesus as the one with authority, and I know who
Paul is, namely that he, unlike the sons of Sceva, is entitled to use
Jesus’ name; but it is more probable that Luke is introducing verbal
variation.
υμείς δέ τίνες ἐστέ; That is, What right have you to invoke the
name of Jesus? The point is important for Luke’s understanding of
the name (see on v. 13; cf. Schneider 2.266). υμείς is emphatic,
 48. PAUL’S SUCCESSFUL MINISTRY AT EPHESUS. 19.8-20. 911

‘thrown forward contemptuously’ (Page 206). Cf. Isaeus, De Heredi-

tate Cironis 24, σὺ δὲ τίς εἶ; σοι δὲ τί προσήκει θάπτειν; ού
γιγνώσκω σε.

16. In ν. 15 the subject of the verb was the evil spirit; here it is the
man possessed by the spirit, which is apparently able to speak but can
act only through its host. ἐφαλόμενος (aorist participle, P74 AB
1175 pc) is more suitable to the narrative than the present participle
(ἐφαλλόμενος, P41 E Ψ <a>; ἐναλλόμενος, D); this could mean
however that the present was original and was improved by copyists.
The changed order (ἐπ’ (or εις) αυτούς ό άνθρωπος), which brings
the antecedent nearer to the relative pronoun, is secondary. The
spirit, taking action through the man it controls, suits the deed to the
word and the would-be exorcists are discomfited, κατακυριεύσας
(against the simple verb, κυριεύσας, D pc) may be accepted, and
the καί that precedes it in (Ψ) 104 323 453 1241 al rejected as
introduced by copyists in a wooden attempt to avoid asyndeton
between the participles ἐφαλ. and κατακ. ‘Die Partizipia folgen sich
asyndetisch (ohne καί bzw τε) wenn sie nicht den gleichen Wert im
Satzgefüge haben’ (BDR § 421). κατακ. is indeed virtually unneces-
sary with ϊσχυσεν.
ἀμφοτέρων would normally mean both, which is inconsistent with
επτά in v. 14. For a variety of conjectures and variant readings see
Metzger (471f.). There is an interesting but inconclusive discussion
in Μ. 1.80. There are a few somewhat later papyrus examples of the
use of ἀμφότεροι in the sense of all (MM 28; BA 93) and it is
probably best to suppose that Luke here gives us the earliest known
occurrence. See Begs. 4.241f., and cf. 23.8. The omission by E and
the substitution of αυτών by (Ψ) <a> are certainly secondary.
The possessed man, provoked by the demon, tore the clothes
off the would-be exorcists (γυμνούς) and beat them
(τετραυματισμένους) so that they fled ἐκ τοῦ οἵκου εκείνου. This
should refer back to a previously mentioned house; there is none.
Ephraim (Ropes, Begs. 3.184) gives a fuller picture: et stridit
dentibus daemonium ad rectam et sinistram et expulit eos a domo.
Roloff (285) concludes from the unexplained reference to a house
that Luke is abbreviating tradition. But the whole of vv. 14—16
creates a somewhat unsatisfactory impression, which probably
accounts for the rewriting in D. One is inclined to suppose that a
fragmentary and unsatisfactory traditional narrative has been incor-
porated here; its unsatisfactory state might be held to speak well for
Luke’s faithfulness to tradition. But he could have tidied it up
without unfaithfulness, and it is hard to know why he did not. The
alternative possibility, that he himself caused the untidiness by
incorporating into his story of Paul an incident that did not belong to
it, has much to be said for it. It is surprising that Paul does not in the
912 COMMENTARY ON ACTS

end drive the spirit out. But the spirit had shown that it was on Paul’s
side.
We may note once more that ‘Die Nennung des Namens Jesu
wirkt nicht automatisch’ (Weiser 532; cf. Conzelmann 111).
17. τοῦτο δὲ ἐγένετο γνωστόν: cf. 9.42, a Lucan phrase, but not
an unnatural one. Ίουδαίοις τε καἰ 'Έλλησιν also is by way of
becoming a formula; cf. 19.10. For Jews in Ephesus see on v. 8, and
18.19. ἐπέπεσεν φόβος ἐπί: cf. 2.42; 5.5, 11.
ἐμεγαλύνετο (cf. 5.13; 10.46; Wilcox 66 for the Septuagintalism)
τό όνομα τού κυρίου Ιησού. The negative effect of the name as it
reacts on those who use it improperly is as impressive as its positive
effect. For the significance of the name in Acts see on 3.6. The mere
pronouncing of the name was not effective; it is a means of access to
the person whose name it is; hence in the present verse it is Jesus
who is magnified (glorified). The reference to his name is a way of
saying this.
Schneider (2.267) sees in vv. 17-20 Paul confronted by syncre-
tism; he refers to G. Klein, Das Synkretismus (1967), 59.
18. πεπιστευκότων: D Ψ 614 2495 pc have πιστευόντων, E has
πιστευσάντων; the perfect, as here, occurs at 15.5; 18.27; 21.20, 25.
In a sentence determined by a verb in a past tense (ἤρχοντο,
imperfect, describing a sequence of events, many would come, one
after another) it will have a pluperfect sense: those who had believed
(and of course continued to believe; that is, those who had become
believers).
The general sense of the next words is clear, but not their precise
construction and meaning. (1) Is πράξεις governed only by
ἀναγγέλλοντες or by ἐξομολογούμενοι also? Is ἐξομ. used abso-
lutely, without object? (2) Does πράξεις mean actions in general, or
does it have its specialized meaning of (magical) spells? For this
meaning see MM 533; BA 1399 (e.g. PGM 4. 1227, πρᾶξις γενναία
ἐκβάλλουσα δαίμονας), ἐξομολογείσθαι usually has a direct object
(e.g. Mk 1.5) but it seems best here to separate it from ἀναγγέλλειν,
which otherwise would come near to duplicating it, and take the
meaning to be that they confessed (their sins, in general) and
disclosed their magical spells (which otherwise would have been
kept secret in the books burned in v. 19). For this use of
ἐξομολογείσθαι cf. 2 Clement 8.3 (οὐκέτι δυνάμεθα ἐκει
ἐξομολογήσασθαι ή μετανοεῖν ἔτι). The special sense of πράξεις is
supported by the use of πράσσειν in v. 19.
19. ικανοί, one of Luke’s favourite words of quantity, is about as
imprecise as ‘a good many’. Like πράξεις (v. 18), περίεργα is a
semi-technical term for magical practices. The meaning originates
with ‘things better left alone, not meddled with’; cf. Plato, Apology
 48. PAUL’S SUCCESSFUL MINISTRY AT EPHESUS. 19.8-20. 913

19b, Σωκράτης αδικεί καί περιεργάζεται ζητών τά τε ὑπό γης καί

τά ἐπουράνια. So too Ecclesiasticus 3.23 (Page 206). Cf. Xenophon,
Memorabilia 1.3.1. For περίεργα itself see MM 505; BA 1303.
πράσσειν derives meaning from its object, but the use (v. 18) of
πρᾶξις gives it added force and direction.
τάς βίβλους, the books in which the spells were written down and
thus kept ready for use. They would no doubt resemble the papyri
edited and published by K. Preisendanz in Papyri Graecae Magicae
(1928, 1931). There is an example in Background 34-7. See also
Betz (154, n. 4, quoting Lucian, Philopseudes 12, ... ἐπειπών
ἱερατικά τινα ἐκ βίβλου παλαιᾶς ονόματα επτά ...). Ephesus was
noted for such products, and the term Έφέσια γράμματα was
current. Thus Plutarch, Symposium 7.5.4 (706D), ώσπερ γάρ οἱ
Μάγοι τούς δαιμονιζομένους κελεύουσι τά Έφέσια γράμματα
πρὀς ἀυτούς καταλέγειν καί ὀνομάζειν... ; Clement of Alexandria,
Stromata 5.8.45.2, τά Έφέσια καλούμενα γράμματα ἐν πολλοῖς δὴ
πολυθρύλητα όντα. Magic was officially discouraged (see probably
Tabula VIII, Qui malum carmen incantassit... ) but almost univer-
sally believed in. Only sceptics such as Lucian (not Christians, who
disapproved but did not disbelieve) denied its power. For the burning
of the books cf. Suetonius, Octavian 31, ... quicquid fatidicorum
librorum Graeci Latinique generis, nullis vel parum idoneis auctor-
ibus, vulgo ferebatur, supra duo millia contratta undique cremavit.
But here the owners were not consulted and Augustus’s motive was
different. Cf. Livy 40.29, Libri in comitio, igne a victimariis facto, in
conspectu populi cremati sunt; Diogenes Laertius 9.52; Lucian,
Alexander 47.
συνεψήφισαν: no subject is expressed, unless we are to think of
the owners of the books, who brought them. It may be better to take a
subject out of πάντων. The public who witnessed the conflagration
were so impressed that they computed the value of the books they
saw on the bonfire—no doubt in a rough and ready way. Wilcox
(127), comparing the use of the third person plural active instead of a
passive and the frequency of this construction in Hebrew and
Aramaic, thinks of a Semitism, but in the present context this is most
unlikely. An impersonal plural—‘they did this or that’—occurs from
time to time in many languages.
In the computation of the value of the books that were destroyed
no unit of currency is expressed. For the ellipse see Μ. 3.17, where
Turner rightly notes that δραχμῶν must be supplied, and adds that
with this αργυρίων (rather than αργυρίου) would be expected—
50,000 silver (pieces). For the omission of the monetary unit cf.
Plutarch, Galba 17 (πέντε καί είκοσι μυριάδας αργυρίου); Jose-
phus, Ant. 17.189 (αργυρίου ἐπισήμου μυριάδας πεντήκοντα). Was
it customary to use the singular ἀργυρίου when no unit was
specified? It might have seemed more natural to give an estimate of
914 COMMENTARY ON ACTS

the number of books burned (as Suetonius does). That Luke puts a
price on them may reflect his strong dislike of the money-making
side of magic and his clear rejection of it from the Christian side; cf.
his treatment of Simon Magus (8.4-25). The whole episode, with the
kind of judgement and attitude it reflects, is important. Luke has a
strong dislike of magic and of anything related to it. A miracle-
working magus is the last sort of model he would use for an apostle.
Cf. Pseudo-Phocylides 149: Make no potions, keep away from
magical books.

20. A characteristic concluding summary. For ὁ λόγος ηὔξανεν cf.

6.7 (ὁ λ. τ. θεού) and 12.24. ἰσχύειν is not used elsewhere in this
sense, but it is a Lucan word (Lk., 8 times; Acts, 6 times, including
19.16, above; rest of the NT, 14 times), κατά κράτος is not used
elsewhere in the NT. It may mean with all one's might (e.g. Plato,
Laws 692d), or by force, with violence, usually in military contexts
(e.g. Thucydides, 8.100.5). The latter, taken metaphorically, is the
sense here, though the metaphor is living in view of the discomfiture
of the sons of Scaeva and the burning of the magical books. An
alternative possibility is to take κράτος with the following words,
τού κυρίου, ‘by the power of the Lord the word grew ...’. This
cannot be ruled out but it does not seem probable; one would expect
κράτος to have the article.
If this possibility is rejected we have the unusual word order τού
κυρίου ό λόγος (this is reversed in P74 (Ε Ψ ) lat syh) Cf. 4.33,
and see Moule (IB 169). The text of D* is also strange: ἐνίσχυσεν
καί ή πίστις τού θεού ηὔξανε καί ἐπλήθυνε (ἐπληθύνετο, D2). The
first verb lacks a subject. It would be easy to conjecture an original
from which both forms of the text might have diverged, but
impossible to establish it with any certainty. The general sense of the
verse is not in doubt.
Luke here brings one incident to a close and is about to embark on
a fresh piece of tradition unconnected with the last, except so far as
19.26 refers to the great success of the Christian mission described in
19.11-20.
 49. RIOT AT EPHESUS 19.21-40

(21) When these events were done, Paul1 formed the intention to pass
through Macedonia and Achaea and travel to Jerusalem. He said, ‘After I
have been there I must see Rome too.’ (22) He sent2 into Macedonia two of
his assistants, Timothy and Erastus, and himself extended his stay in Asia.
(23) At that time there arose no small disturbance concerning the Way.
(24) One Demetrius by name, a silver-smith,3 by making silver shrines of
Artemis4 provided the craftsmen with5 no small amount of business. (25) He
gathered them and6 the workmen7 engaged in this business together, and
said, ‘Men, you know that our prosperity arises out of this business; (26) and
you see and hear that not only in Ephesus but in almost all Asia this fellow
Paul8 has persuaded and led astray a large crowd of people, saying that these
that are made with hands are not gods. (27) Not only does this mean for us a
risk that this line of business may come into disrepute but also that the
temple of the great goddess Artemis9 may be reckoned as nothing, and she
whom all Asia and the inhabited world worship will be cast down from her
greatness.’
(28) They listened to this and were filled with rage, and shouted, ‘Great is
Artemis9 of the Ephesians.’ (29) The city was filled with confusion; they
seized Gaius and Aristarchus, Macedonians and travelling companions of
Paul’s, and rushed with one accord into the theatre. (30) Paul wished to10 go
[into the theatre] to the people, but the disciples would not permit him to do
so. (31) And some of the11 Asiarchs, who were well disposed to him, sent
and begged him not to12 go into the theatre. (32) Some shouted one thing,
others another; for the assembly was confused, and the majority did not
know why they had come together. (33) Some of the crowd13 instructed
Alexander, and the Jews put him forward. Alexander made a gesture with his
hand14 and wished to make a defence to the people, (34) but when they
recognized that he was a Jew there arose one cry from all, who for about two
hours shouted, ‘Great is Artemis9 of the Ephesians.’
1RSV, resolved in the Spirit.
2NJB, sent ahead of him.
3RSV, who made; NEB, who made shrines of Diana.
4NJB, Diana.
5NEB, a great deal of employment.
6NJB, others.
7RSV, of like occupation; NEB, in allied trades.
8NEB, with his propaganda has perverted.
9NEB; NJB, Diana.
10RSV, to go among the crowd; NEB, to appear before the assembly; NJB, to make
an appeal to the people.
11NEB, dignitaries of the province.
12RSV, to venture; NEB, to take the risk of going.
13RSV, prompted; NEB, explained the trouble to; NJB, prevailed upon.
14NJB, with the intention of explaining things.
915
916 COMMENTARY ON ACTS

(35) The town clerk stilled the crowd and said, ‘Men of Ephesus, what
man is there who does not know that the city of the Ephesians is temple
warden of the great Artemis9 and of15 the stone that fell from heaven? (36)
Since these matters are not open to contradiction you must remain quiet and
do nothing rash. (37) For you have brought these men [here]16 who are
neither guilty of temple profanation nor blasphemers of our goddess. (38) So
if Demetrius and the craftsmen associated with him have a suit against
anyone, courts are held and there are proconsuls: let them accuse one
another. (39) And if you17 seek anything more than that it will be dealt with
in the lawful assembly. (40) For as for this day, we run the risk of being
accused of riot, there being no cause for it. We shall not be able to give a
reason for this meeting.’18 With these words he dismissed the assembly.

Bibliography
A. Bammer, Die Architektur des jüngeren Artemisiums von Ephesos,
Wiesbaden 1972.
S. Μ. Baugh, NTS 36 (1990), 290-4.
W. H. Buckler, FS W. Μ. Ramsay, 27-50.
C. Burchard, ZNW 61 (1970), 167f.
H. J. Cadbury, JBL 50 (1931), 42-58.
E. Delebecque, RScPhTh 66 (1982), 225-32.
G. S. Duncan, NTS 3 (1957), 211-18.
E. Fascher, FS Bultmann (1954), 247f.
J. A. Fitzmyer, as in (48).
E. L. Hicks, Expositor 1 (1890), 401ff.
G. H. R. Horsley, as in (46).
L. J. Kreitzer, JSNT 30 (1987), 59-70.
P. Lampe, as in (48).
G. Μ. Lee, Bib 51 (1970), 237.
B. A. Mastin, JTS 27 (1976), 405-12.
F. Miltner, as in (48).
C. F. D. Moule, ExpT 65 (1945), 221.
R. Oster, JAC 19 (1976), 24-44.
R. Oster, HThR 77 (1984), 233-7.
F. Sokolowski, HThR 58 (1965), 427-31.
R. F. Stoops, JBL 108 (1989), 73-91.
W. A. Strange, JTS 38 (1987), 97-106.

15NJB, her statue.

16Greek; RSV, omit.
17NJB, want to ask any more questions.
18RSV, there being no cause that we can give to justify this commotion; NEB, if the
issue is raised, we shall be unable to give any explanation of this uproar; NJB, we can
give no justification for this gathering.
 49. RIOT AT EPHESUS. 19.21-40. 917

Commentary

This paragraph falls into two parts. The second, vv. 23-40, must be
based upon information derived by Luke from Ephesus, either on a
visit to the city or by inquiry, oral or possibly written, from Ephesian
residents. It is full of material that presupposes accurate, or at least
reasonably accurate, knowledge of Ephesus. Not only is the great
temple of Artemis known—this would prove nothing, since most
people in the ancient world had heard of it; there are details about the
work and trade of the silversmiths (of whom one is named),
appropriate references (the adjective μεγάλη) to Artemis, mention of
the Asiarchs, and a part for the town’s γραμματεύς. For details see
the notes below. Paul’s fellow travellers, Gaius and Aristarchus, are
involved (and could possibly have provided information). It is most
improbable that Luke’s account (a page and a half of NA26) is
complete or that it is correct in all its details, but the objections to its
historicity brought by Haenchen (553f.) are superficial and uncon-
vincing. He speaks first of Demetrius, who, in Luke’s narrative,
appears at first to be in command of the situation. ‘Aber dann kommt
das Unbegreifliche: er lässt nicht mehr von sich hören. Wie kann ein
Mann, der als ein so gut Organisator dargestellt wird, sich gerade
dann in Schweigen hüllen, wenn er—vor einem günstigen Pub-
likum—mit einem konkreten Vorschlag herauskommen müsste?’
We do not in fact hear a great deal about Demetrius’ organizing
ability, and why should we hear more? He has raised a mob that
threatens to tear Paul in pieces, and secures his immediate (20.1)
departure from the city. Nothing unbegreiflich here. Again: the
Asiarchs. It is hard to see why Paul should not have had friends
(Luke does not say converts) among them. ‘Diese Asiarchen wohnen
doch nicht alle beisammen. Reagieren sie also alle in gleicher Weise?
Oder halten sie gerade eine Sitzung ab?’ Luke does not say that all
the Asiarchs reacted in the same way; he speaks of τινὲς τών
Άσιαρχῶν. And some (many? a few?) of them would hardly need to
call a meeting. Finally (Haenchen says endlich though he has in fact
another point to make) there is the town clerk, who is said to dismiss
the complaints too easily, and incredibly frightens the crowd with a
threat of the charge of στάσις. But a charge of στάσις could have
been a very serious matter for the city; see on v. 40. We cannot tell
how far the town clerk’s tongue may have been in his cheek when he
declared, ‘Die Christen lästern die Göttin nicht,’ and there was no
need for him to add ‘sie leugnen nur ihre Gottheit’ (554). If ‘das hat
der Ratschreiber ebenso wie seine Hörer vergessen’, one must add
that the town clerk would in these circumstances regard it as his
business to forget and to cause the crowd to forget. It is of course true
that Paul could hardly have been satisfied with what is said in v. 37,
but the town clerk was not speaking in order to satisfy Paul but to
918 COMMENTARY ON ACTS

quell a riot; he may or may not have been sincere in what he (in
Luke’s narrative) says, but a certain economy of truth has often been
practised in a good cause.
In Haenchen’s opinion an even greater difficulty arises out of a
comparison with 2 Cor. 1.8ff. According to these verses Paul in Asia
experienced so severe a θλιψις, ‘dass er den Tod für unausweichlich
hielt ... Von diesem Ereignis schweigt Lukas—warum?’ Because,
Haenchen answers, Paul was obliged to flee for his life from
Ephesus. Luke is silent? To have a theatre full of people thirsting for
one’s blood might well have seemed life-threatening. Moreover,
though 2 Cor. 1.8-11 is an obscure passage at least a strong case can
be made for seeing in it a reference to illness (see 2 Corinthians
63f.). It is of course true that the town clerk’s speech makes a
conclusion that will have pleased Luke more than a description of a
hurried flight (and Paul wastes no time in 20.1). The speech forms
part of the picture of favourable treatment by unbiased authorities
that Luke intends to present. His motivation is undoubtedly visible in
the story of the riot—Tn dem allen tritt deutlich die apologetische
Tendenz als die beherrschende Intention dieser lukanischen Erzäh-
lung ans Licht’ (Schmithals 181). But it is not a story riddled with
improbabilities, and 1 Cor. 15.32 and 2 Cor. 1.8 between them,
though they are anything but clear, show that Paul’s ministry in
Ephesus was a disturbed one.
The first, and much shorter, part of the paragraph (vv. 21,22) is of
a different kind. Paul determines to make a roundabout journey to
Jerusalem, travelling by way of Macedonia and Achaea; after visiting
Jerusalem he will go to Rome. Two of his assistants, Timothy and
Erastus, are sent ahead into Macedonia, whether by land or sea is not
stated. Paul is to follow them into Macedonia, and continue into
Greece (20.1, 2). For the rest of this journey see 20.3—21.15. In the
whole of it a number of place names occur but there are only two
incidents, the church meeting in Troas (20.7-12) and the address to
the Ephesian elders in Miletus (20.17-38). If any part of Acts may be
described as an Itinerary it is this; it is easy to see how the two
events, which are best explained as independent Pauline stories
collected by Luke, together with the account of the riot, have been
inserted into it. That two assistants should have been sent ahead of
the main party is quite understandable, especially since (though Luke
makes no mention of the matter) this journey to Jerusalem was
primarily if not exclusively for the purpose of picking up the sums
collected in the various churches for the relief of poverty in
Jerusalem and conveying them to their destination. The two assis-
tants would tell the members of the churches to complete their
preparations: Paul would soon be there. The reader will recall that
assistants were sent to Corinth for precisely this purpose; see
2 Corinthians 8. The difficulty arises that Paul mentions three
 49. RIOT AT EPHESUS. 19.21-40. 919

messengers, of whom the only one named is Titus. Timothy was sent
to Corinth (1 Cor. 4.17; 16.10f.); this however appears to have been
at an earlier date, and no mention is made of a companion. There is
much to be said for the view that v. 22 refers, in mistaken terms, to
the mission of 2 Corinthians 8. Titus is not mentioned in Acts (see FS
Black (1969), 2). It may be that Luke in omitting almost all reference
to the collection found it expedient to omit Titus too.

21. Here only in Acts is πληροῦσθαι used of events; cf. Lk. 1.1,
πεπληροφορημένων πραγμάτων. ταῦτα almost certainly means
‘these events, just described’; it could refer to ἔτη in 19.10 (Begs.
4.244), but this noun is too remote to form a likely antecedent. For
ὡς δὲ ἐπληρώθη ταῦτα, D has τότε. For temporal ὡς cf. e.g. 18.5.
ἔθετο ἐν τῷ πνεύματι, Paul formed the intention, τιθέναι,
τίθεσθαι is used in something like this sense from an early time in
Greek literature (LS 1790, s.v. A II 6); so in Homer, Odyssey 4.729
as here with an infinitive (ἐνὶ φρεσὶ θέσθε ... μ’ ἀνεγείραι). ἔθετο
alone could hardly mean ‘purposed’, so that τῷ πνεύματι cannot be
taken to refer to the Holy Spirit (as e.g. in RSV, Paul resolved in the
Spirit), though this would accord with other passages in which Paul’s
plans are said to be controlled by the Holy Spirit (e.g. 16.6). For the
infinitive πορεύεσθαι, see BDR § 392.3, n. 12, though the reference
there to the influence of Hebrew is doubtful.
Paul’s intention is to pass through Macedonia (see 16.9) and
Achaea (see 18.12) and travel to Jerusalem, διελθὠν... πορεύεσθαι.
D has διελθεῖν ... καί πορεύεσθαι, which Black (AA 67) regards as
an example of significant parataxis in the Western text, suggesting its
affinity with Aramaic tradition. It is perhaps more likely than he
allows that, though D is joined here by A, the parataxis is due to the
influence of the Latin (d has transire ... et sic ire ... ). For διέρχεσθαι
as meaning not merely a journey but a preaching tour see on 13.6; this
is a probable meaning here, since Paul could hardly have avoided
work in the familiar mission field even if he had wished to do so.
It appears to be Luke’s understanding of events that the journey
contemplated here was undertaken at 20.1. The riot (v. 23) happened
‘at that time’ and after the uproar had been quelled (v. 40) Paul set
out for Macedonia, went on to Greece (20.2), spent three months
there (20.3), and then travelled by a circuitous route to Jerusalem.
This account however seems to represent one of Luke’s major
omissions. Whether the sending of Timothy (v. 22) is to be identified
with that forecast in 1 Cor. 4.17; 16.10 is not certain; Paul’s final visit
to Greece (20.2, 3) seems to follow upon the troubled history to
which 2 Corinthians alludes (see 2 Corinthians 5-21). This troubled
history is passed over in complete silence. It is difficult to think that
Luke was totally unaware of it; it was not part of the story he wished
to tell. Paul’s intention at this time was to travel to Achaea by way of
920 COMMENTARY ON ACTS

Macedonia (1 Cor. 16.5); later he changed this plan, and intended to

travel to Macedonia by way of Corinth, and then to retrace his steps,
giving Corinth the advantage of a double visit (2 Cor. 1.15f.). This
plan also was changed (2 Cor. 1.23), a fact which led to unpleasant
comments at Corinth. Whether Paul would on the occasion of his
projected visit continue to Jerusalem was unsettled (1 Cor. 16.4).
τὴν Μακ. καί Άχαῖαν, Β Ψ <a> unites the provinces into one
goal. τὴν Μακ. καί τήν ’Αχ., P74 A D Ε 33 945 1739 pc distinguishes
them.
εἰπών must be regarded as contemporaneous with ἔθετο; it is not
conversation but the formulation of the purpose as it arose in his
mind; the immediate intention to go to Macedonia and Achaea was
part of the longer purpose to travel to Jerusalem and Rome.
γενέσθαι με εκεί looks on the visit to Jerusalem as a unit but it
could do so either with reference to arrival (when I have got there) or
to the stay (when I have been there). The difference is real but does
not affect the sense. Jerusalem must be visited first, but the more
remote objective, Rome, was beginning to fill Paul’s mind, according
to Acts, and according to Paul himself (Rom. 15.22-29). For Paul,
Rome was to be a staging post on the way to Spain. This Luke does
not mention (possibly because he knows that Paul did not get so far);
Rome is the goal of his story, and if he can show the faith planted,
and its great teacher at work, in the capital he will have accomplished
his task. If the mission can reach Rome, and within a generation,
there is nowhere it cannot go. Rome was probably alluded to at 1.8;
Aquila and Priscilla had come from Rome, which probably had
already been evangelized (18.2). Acts shows nothing of Paul’s
diffidence in writing to a church that he himself had not founded. The
journey is one that Paul must (δεῖ) make; cf. 23.11; 27.24; and see E.
Fascher in FS Bultmann (1954), 247f. Stählin (258) points out that
the journey was intended (a) to strengthen the churches, (b) to deal
with difficulties, and (c) to put together the collection, but Marshall
(313) is probably right in commenting that Luke had recognized that
Paul’s arrest and subsequent story, including the journey to Rome,
were in the end more important than the collection.
’Vielleicht haben wir hier [21f.] ein ziemlich unverändertes Stück
Itinerar vor uns’ (Bauemfeind 232). The Ephesian incidents were
inserted into it.

22. Timothy (see 16.1-3) was sent to Corinth (and might well
have been sent through Macedonia) at 1 Cor. 16.10 (cf. 4.17). A visit
by Timothy to Macedonia (Philippi) is contemplated in Phil.
2.19-23; this may have been from Ephesus, but a more likely place
of writing for Philippians is Rome. According to Pesch (2.176) the
Macedonian visit of 1 Corinthians 16 belongs to an earlier stage of
the Ephesian ministry; it differs in that nothing is said of Erastus. An
 49. RIOT AT EPHESUS. 19.21-40. 921

Erastus is mentioned at Rom. 16.23 as ὁ οἰκονόμος τής πόλεως. The

city is not named, but Romans was probably written in or near
Corinth, and it is a fair guess that Erastus was city treasurer of
Corinth. The name occurs again at 2 Tim. 4.20: Erastus ἔμεινεν ἐν
Κορίνθῳ. This would agree with but does not require his being a
Corinthian official. We cannot be certain who wrote these words (in
2 Timothy), though even if Paul did not write them they may have
been based on sound tradition; we cannot however place in time the
residence of Erastus referred to in 2 Timothy. Begs. 4.244 refers to a
Corinthian inscription that mentions an aedile Erastus; see H. J.
Cadbury in JBL 50 (1931), 42-58, but also Hemer (235), who (like
Cadbury) doubts whether any identification is possible.
Timothy and Erastus are described as two τῶν διακονούντων
αὐτῷ. Timothy’s role is not defined at 16.3; cf. 13.5, Paul and
Barnabas had John as υπηρέτης. It is clear from the epistles that
Timothy was a trusted and valued colleague (Phil. 2.20), though in 1
Cor. 16.10 (written probably somewhat earlier than Philippians) it
seems that Paul has to tell the Corinthians to treat him with proper
respect. διακονεῖν suggests a somewhat menial position; cf. Phil-
emon 13 (Onesimus will attend to Paul’s needs in prison). The
expression ‘personal assistant’ may cover what is meant, but it is
clear that this in fact (whether Luke knew it or not) included pastoral
responsibility in the churches.
τήν Μακεδονίαν: the article is given in P74 A B D Ψ <a>, omitted
by E 36 323 614 1175 1891 al.
When his assistants left for Macedonia Paul stayed in Ephesus; cf.
1 Cor. 16.8 (though it is not clear that the two verses refer to the same
occasion), ἐπέχειν is sometimes used with χρόνος (cf. Herodotus
1.132.3, ἐπισχών δέ ολίγον χρόνον; D in this verse includes ολίγον)
for spending an undefined or unspecified time in a place. Luke is
making room for one of his best stories.
For the use of εἰς and the accusative when ἐν and the dative would
have been expected see 2.5. Here D corrects to ἐν τῆ ’Ασίᾳ. For a
number of minor Western variants in vv. 22-37 see Metzger
(472f.).

23. A new narrative, complete in itself and free (except in v. 26)

from any cross reference to Paul’s work in general begins here. This
verse is Luke’s own introduction to the following narrative (so
Lüdemann 224); its basis is not in itself Christian (Lüdemann 226).
Though not dependent on it is reasonably consistent with such other
knowledge as we possess, and Luke may have heard it told in
Ephesus, or by Ephesian Christians elsewhere. Reicke (DFZ 313f.)
refers to evidence for similar riots in W. H. Buckler (FS Ramsay,
27-50) and T. R. S. Broughton (Roman Asia Minor: An Economic
Survey of Ancient Rome, ed. T. Frank, 4.846ff.)
922 COMMENTARY ON ACTS

Έγένετο: this is not the Semitic use of this word. A disturbance

(τάραχος), and no small one (the litotes is characteristic of Luke; see
12.18), arose—happened. It happened κατά τόν καιρόν ἐκεῖνον;
this means no more than ‘during Paul’s Ephesian ministry’, though
Luke (20.1) takes it to have provided the cue for Paul’s departure,
which is likely enough. The disturbance concerns ‘the Way’; on οδός
as a term for Christianity see on 9.2. In the disturbance, ‘Dabei wird
deutlich, wie sich Religiosität, Patriotismus und wirtschaftliches
Interesse zu einem untrennbaren Ganzen miteinander verbunden
haben’ (Roloff 291). On the motivation of the story see Schmithals
quoted on p. 918.

24. Demetrius: the name occurs at 3 Jn 12, but (though 3 John may
well have originated in or in the neighbourhood of Ephesus) it is
unlikely that the two should be connected; Der Kleine Pauly
1.1462-71 mentions 33 persons bearing the name. This Demetrius
was an ἀργυροκόπος, a silversmith. The participle that follows,
ποιῶν, may be either attributive (D, reading ἦν instead of ὀνόματι,
makes this certain),... who made ..., or adverbial,... by making ...
There is no way of deciding with certainty which construction is
correct. Μ. 3.152 notes the ambiguity, and Moule, (IB 105) noting
both possibilities, adds ‘the latter seems to be the more probable’.
Certainly it helps to bind the narrative together. It was by making
silver shrines that the silver trade flourished, and it was this activity
that was attacked, implicitly or explicitly, by monotheists such as
Paul. Hence the τάραχος.
Papyrus examples of the use of ἀργυροκόπος (MM 74; BA 211)
have been supplemented by an Ephesian inscription (IEph VI
(1980) 2212.4-7, 9; in ND 4.7-10) which mentions ‘Μ. Antonius
Hermeias, ἀργυροκόπος, νεοποιός’, and also ‘τὸ συνέδριον τών
ἀργυροποίων’, the gild of silversmiths. See further Hemer (235f.).
Demetrius made ναούς ἀργυροῦς. The adjective is omitted by B
gig: possibly by homoeoteleuton, though it is possible that the
adjective was not originally in the text and was added on the strength
of Demetrius’s occupation, ναός usually means a temple, or the most
sacred part of a temple, the inner shrine where the image or other
sacred object was placed, but it has been conjectured that the word
was also used for small portable shrines which were carried in
religious processions. Those made by Demetrius were shrines of
Artemis (not representations of a deity, so that Lucian, Alexander 18
is not a parallel). On Artemis, the great Ephesian—and Asian—
goddess, see especially L. R. Taylor in Begs. 5.251-6, supplemented
by ND 4.74-82. As a Greek goddess Artemis was the daughter of
Zeus and Leto, and sister of Apollo, worshipped already in Myce-
naean times. She was a virgin who helped women in childbirth, a
huntress armed with a bow, the goddess of death. The establishment
 49. RIOT AT EPHESUS. 19.21-40. 923

of an Ionian colony at Ephesus, and similar acts of colonization

elsewhere in Asia Minor, led to assimilation of the Greek Artemis to
deities of oriental origin. Worship of a goddess (perhaps of fertility)
seems to have been practised in Ephesus before the arrival of the
Greeks, and images (which may once have included the great golden
image in the temple at Ephesus) have often been interpreted as many-
breasted, suggesting that she was a fertility goddess. An alternative
interpretation of the supposed breasts as bull’s testicles would
suggest fertility even more strongly (see D. W. J. Gill and B. W.
Winter in The Book of Acts 2.88). The view that Artemis was a
fertility goddess is however effectively criticized in the same volume
(319f.) by P. A. Trebilco. The Ephesian goddess was probably
related to Cybele and to Ma (who had her own temple in Ephesus).
The fame and sanctity of the temple of Artemis are widely attested;
see below, v. 27, and the note.
There is evidence for the addition, between ποιων and ναούς, of
the words ίσως κιβώρια μικρά, perhaps small cups (chalices might
be better, since there is evidence for the liturgical use of κιβώριον).
the support for this reading is very slight, but it includes Chrysostom.
In view however of the word ἴσως, and of the fact that Chrysostom
professes ignorance of the silver shrines, it is probably wise to regard
this not as a genuine variant but as a guess at what the ναούς
αργυρούς might be.
Another suggestion was made as long ago as 1890 by E. L. Hicks
(Expositor 1 (1890), 401ff.). Hicks suggested that the text ναούς
ποιούν was a corruption of the word νεοποιός (there are various
spellings, ναοποιός, νεωποιός), the term used in the Roman period
for the officials elected by the city tribes to supervise the fabric of the
temple of Artemis. For details see most conveniently Sherwin-White
(90f.). It is an interesting speculation, and if it is correct it could be
used to argue for Luke’s knowledge of Ephesus in the first century
(though Hicks himself attributed the present text to Luke's mis-
understanding). But our only ground for it is our ignorance (already
in Chrysostom?) of the custom of making silver shrines. Hicks
further suggested that the Demetrius of the riot might be identified
with a Demetrius referred to in an Ephesian inscription, dating
probably (see Sherwin-White 91f.) from the first century, as a
νεωποιός (see AGIBM III 578, Begs, 5.255). ND 4.127-9 quotes a
thanksgiving to Lady Artemis, from a νεοποιὸς αὐθαίρετος (volun-
tary) (IEph 3.961.3).
For descriptions of the great temple of Artemis see Strabo
14.640ff.; Pausanias 2.2.5; 4.31; also A. Bammer, Die Architektur
des jüngeren Artemisiums von Ephesos (Wiesbaden 1972). The
temple measured 120 x 70 m.; it had 128 pillars, 19 m. high.
Demetrius provided (παρείχετο; παρείχε A* D E pc, without
substantial difference in meaning) the craftsmen (τεχνίταις, the
924 COMMENTARY ON ACTS

skilled workmen who actually produced the shrines, whether in his

regular employment or under contract) with a good deal of business,
trade (εργασίαν), literally, no little trade—litotes, as in v. 23.
25. Demetrius gathered together the craftsmen (οὕς, pointing back
to ν. 24) and the ἐργάτας; the latter seem to be the relatively unskilled
workmen. The word is used of farm labourers (Herodotus 5.6.2,
ἀργόν εἶ ναι κάλλιστον, γης δέ ἐργάτην ἀτιμότατον—admittedly, of
the Thracians). The use of περί is classical, ‘den Gegenstand des Tuns
oder der Bemühung bezeichnend’ (BDR § 228.2). ‘Alii erant
τεχνῖται, artifices nobiliores; alii ἐργάται operarii' (Bengel 466).
ἄνδρες: in most passages (see 1.16) this address is given a further
qualification. Demetrius goes on to appeal to common knowledge,
and to natural human instinct. Work, and thereby prosperity, is
afforded by the local cult. Anything that threatens the cult therefore
threatens also the silver-workers of all grades at a very sensitive
point—in their pockets.
For οὔς συναθροίσας καί, D 614 (2495) pc gig sy sa have οὗτος
συναθροίσας.
For ἐργάτας εἶπεν ἄνδρες, D (syp syh**) sa have τεχνίτας ἔφη
πρὸς αυτούς ἄνδρες συντεχνίται. In this reading the ἐργάται
disappear.
26. θεωρείτε και ακούετε. The order is surprising; one would
have expected hearing first, then the appeal to first-hand knowledge
(sight).
’Εφέσου: ‘Apg 19.26 οὐ μόνον ... όχλον können die Gen. als
Gen. des Ortes erklärt werden (Latinismus? = Ephesi ...), möglich
aber auch als Gen. von όχλον abhängig zu denken (sehr weite
Trennung ... ); D hat ἔως Έφεσίου [sic]...’ (BDR § 186.1, n. 2). Μ.
1.73 compares Sophocles, Oedipus Tyrannus 236f. (γῆς τῆσδε) but
thinks that dependence on όχλον may be right. ‘The gloss έως (D),
“within”, may possibly express the meaning.’ It is probably best to
take the genitive as a sort of possessive: a crowd of people belonging
to Ephesus ...; and to suppose that Ephesus and Asia are placed first
for emphasis.
σχεδὸν πάσης τής ’Ασίας: cf. 19.10. It must be remembered that
Paul had assistants (see 19.22); Col. 1.7 appears to mean that
Epaphras acted on Paul’s behalf in evangelizing Colossae (and Col.
2.1 suggests that there were Christian groups in Laodicea and
elsewhere which Paul had not founded personally though he took
responsibility for them). These statements may well reflect local
tradition even if Paul did not write the epistle. Demetrius would no
doubt be prepared to exaggerate in a good cause, but Luke would
equally be prepared to take over the exaggeration in a somewhat
different one.
πείσας μετέστησεν: by persuasion he has moved people from one
 49. RIOT AT EPHESUS. 19.21-40. 925

side (idolatry) to another (Christianity). Cf. Xenophon, Hellenica

2.2.5, τὰ ἐκεῖ πάντα πρὀς Λακεδαιμονίους μετέστησεν.
The rest of the verse contradicts a somewhat cruder paganism than
that contemplated in 17.29. It is no longer a question of deity
resembling material objects, but whether the material objects were in
fact gods. Christians were by no means the only critics to deny this.
Jews of course did so; this it is unnecessary to document. There was
also a considerable amount of pagan criticism of crude idolatry (see
above, pp. 849f., and cf. Betz 40,43). What the average inhabitant of
Ephesus thought about the images he saw on every side of him is not
easy to determine; it would certainly be mistaken to credit him with
the views of Epicurus or of Lucian. Many if pressed would probably
have agreed that the products of human manufacture (oἱ διά χειρῶν
γινόμενοι) were not themselves θεοί, but would have seen in the
symbolic representation of the gods more than bare symbols. T think
that a man who is altogether burdened in soul and has endured many
misfortunes and griefs in his life and does not enjoy sweet sleep,
would, if he stood before this image [of Zeus, by Pheidias, at
Olympia], forget all the grievous and dreadful things it may befall
one to suffer in human life’ (Dio of Prusa, quoted by A. D. Nock in
Early Gentile Christianity (1964), 5). On the whole the philosophical
attitude to popular religion grew more tolerant as the Hellenistic age
progressed. Thus Zeno (SVF 1.264), 'It will not be necessary to build
temples, for a temple ought not to be held to be worth much or holy
(πολλοῦ ἄξιον καὶ ἅγιον); nothing is worth much or holy which is
the work of builders and artisans (οικοδόμων ἔργον καί βαναύσων)’
(in Clement of Alexandria, Stromata 5.12.76); also Plutarch, quoted
on 17.24. But later see Maximus of Tyre 2.1, 2: 'It is not that the
divine Being stands in any need of images or statues. It is poor
humanity, because of its weakness and the distance dividing it from
God ... which has contrived these things as symbols. People who
have an exceptionally strong power of mental realization, who can
lift the soul straight away to heaven and come into contact with
God—such people, it may be, do not stand in any need of images.
But such people are few amongst men.’
The position of Paul, the Jew and the Christian, was more radical,
and is not unfairly represented by Demetrius: the objects were not to
be identified with gods—indeed, whatever men might think (1 Cor.
8.6), there existed no θεοί with whom they could be identified; there
was but one (1 Thess. 1.9f.; Rom. 1.21-3). The question is whether
Paul (if rightly represented by Demetrius) was fair in thinking that
anyone supposed that the idols actually were gods. It is however
clear that Christianity was mounting a major attack on such religions
as that of Artemis. Both could not prevail. The town clerk (v. 37) was
either more tactful or less intelligent than Demetrius. Cf. Schille
(386): ‘Demetrius wirkt weitsichtiger als der Stadtschreiber, der die
926 COMMENTARY ON ACTS

Dinge v.35 herunterzuspielen versteht.’ Demetrius can see the

‘weltweite Gefahr’. ‘Was Lukas Demetrius in den Mund legt, grenzt
an Prophetie.'
Παῦλος σὗτος is a disparaging expression. After it D adds τις
τoτε, which must be a mistake for τἱς ποτε, nescio quem (gig). See
Ropes (Begs. 3.186f.), and for papyrus parallels (without reference to
this passage) BDR § 303, n. 2.

27. ού μόνον, repeated from v. 26; a common expression in the

later part of Acts (21.13; 26.29; 27.10); characteristic of Jn, and
classical. Demetrius puts business considerations first, but adds a
religious point.
κινδυνεύει is often used in a weakened sense, ‘there is a possi-
bility’, but here the context suggests that the basic notion of danger is
intended—κινδυνεύει ἡμῖν, there is a risk for us. τούτο τό μέρος
now becomes an accusative subject with the infinitive ἐλθεῖν
(otherwise τούτο τό μέρος would be the subject of κινδυνεύει, it
may come ... ). μέρος is the part played, or enjoyed, by Demetrius
and his colleagues; ‘line of business’ (MM 399) is probably right,
though PFlor 1.89.2 (3rd century AD), which MM quote, does not
quite prove their point. τά μέρη τής διοικήσεως are branches of the
administration (der Verwaltung) rather than branches of a business,
Geschäftszweige (BA 1025). But was Demetrius νεοποιός after all
(v. 24), and care for the administration of the temple of Artemis his
part in the administration of the city? It is safest to say, Not only is
there a risk for us that this line of business may come into disrepute
(ἀπελεγμόν, from ἀπελέγχειν)... Schille (386) takes εις ἀπελεγμὸν
ἀλθεῖν to be a Latinism, which d and the Vulgate put back as in
redargutionem venire. See Kilpatrick, JTS 10 (1959), 327. The Latin
suggests that the word was ‘taken to mean something like “reproof,
censure, adverse comment, public criticism” ’.
Two more infinitives dependent on κινδυνεύει follow, λογισθήναι
and μέλλειν. καθαιρεῖσθαι is dependent on μέλλειν. D* have
μέλλει (in parallel with κινδυνεύει) instead of μέλλειν. This simplifies
the sentence, but is not for that reason necessarily original. There is a
danger, a risk ... that the temple of the great goddess Artemis may
come to be reckoned as nothing. Artemis is described as θεά. This
feminine form is not uncommon, but the masculine form θεός is often
used for both genders (e.g. τοῖς θεοῖς εὔχομαι πᾶσι και πάσαις,
Demosthenes 18.1 (225), cf. 141 (274); 21.52 (531); Μ. 3.22 adds
inscriptions and papyri). But see especially Μ. 1.60, 244; and note
ν. 37, where the town clerk speaks of τήν θεόν ημών. At 1.60 Moulton
quotes (abbreviating a little) Blass (209f.): ‘Usitate dicitur ή θεός (ut
37); verum etiam inscriptio Ephesia Brit. Mus. [= AGIBM] nr. 481,
ν. 324 (cf. 220.278) τή μεγίστη θεά Έφεσία Άρτέμιδι, cum alibi
(ibid. 373.375 etc. nr. 482) ή θεός eadem dicitur (in illa formula,
 49. RIOT AT EPHESUS. 19.21-40. 927

v. 13). Itaque formulam sollemnem ή μεγάλη θεά Ά. mira diligentia

L. conservavi.’ At 1.244 (Additional note to the second edition),
however, Moulton prefers the suggestion of G. Thieme. The town
clerk used the technical term; ή θεός is used in inscriptions for the
great goddess of the city; other people’s goddesses were θεαί, and this
use determined that of Demetrius. This convinced Moulton, but one
would have thought that the silversmith leader would have used the
official designation of the goddess whose rights he was defending.
There is still much to be said for Blass; also for the view that Luke was
not interested in consistency in such a matter. It is worthwhile to note
that Xenophon of Ephesus (1.11.5) writes,... τήν πάτριον ἡμῖν θεόν,
τήν μεγάλην Έφεσίων Άρτεμιν.
Notwithstanding Isa. 40.17; Wisdom 3.17; 9.6; Dan. 4.35 (The-
odotion) it would be unwise in such a context as this to take
λογισθῆναι εἰς as a Semitism (predicative represented by εις). See
Μ. 2.462, and cf. Xenophon, Cyropaedia 3.1.33, χρήματα ... εἰς
ἀργύριον λογισθέντα τάλαντα πλείω τῶν τρισχιλίων (not identical
in sense, but a pointer to usage).
For οὐθείς (instead of ούδείς) cf. 15.9 and the note.
And she will be cast down from her greatness; on τής
μεγαλειότητoς see BDR § 180.1, n. 2; a genitive of separation? Page
(208) quote Diodorus Siculus 4.8.2, καθαιρεῖν τι τής τοῦ θεοῦ
δόξης. Some MSS have the accusative, meaning be deprived of.
Ephesian Artemis was universally recognized and revered; e.g.
Pausanias 4.31.8, Ephesian Artemis is the title which all cities
recognize, and by which men privately worship her as the greatest of
the gods; Xenophon of Ephesus, 1.11.5, quoted above; Livy 1.45.2,
iam tum erat inclitum Dianae Ephesiae fanum; id communiter a
civitatibus Asiae factum fama ferebat.
The effect of Christianity on pagan religion is shown in Pliny’s
account of the restoration of paganism when Christianity was
repressed: Certo satis constat prope iam desolata templa coepisse
celebrari et sacra sollemnia diu intermissa repeti inveniebatur (Epis-
tles 10.96.10).

28. Demetrius’ speech had its intended effect in rousing his

colleagues in opposition to the Christian movement.
The Western variant at the beginning of the verse (ταῦτα δέ
ἀκούοαντες, D (lat) syp) is another example of an insignificant
variation which attached no importance to the precise reproduction
of wording, πλήρεις is read here by the majority of MSS, and is to be
accepted. There is evidence (see MM 519; Μ. 2.162; BDR § 137.1)
for πλήρης as indeclinable; this was a Hellenistic practice which was
spreading in the first century AD. It occurs in the NT, but there is no
reason to suspect it here.
θυμός has a wide range of meanings, here certainly, anger, rage.
928 COMMENTARY ON ACTS

The Western text (D (614) syhmg) adds that they ran into the street,
δραμόντες εἰς τὀ ἄμφοδον. This adds nothing to the sense. It could
have been removed by an editor motivated by verbal economy but
more probably added to brighten the narrative. Ropes (Begs. 3.186)
however thinks it ‘one of the few intrinsically interesting “Western”
additions’.
μεγάλη ή Άρτεμις Έφεσίων. See the passages quoted on v. 27;
many more could be added. For the acclamation cf. Bel and the
Dragon 18, 41. ή is omitted by D* pc, rightly according to Ramsay
(Church 139-41) on the ground that the article was not used in
acclamations. It seems however that there was no uniform practice.
The reading with the article violates Apollonius’s Canon (that nouns
in regimen either both have or both lack the article); but this does
happen with proper names and national appellations (Moule, IB 115).
Cf. vv. 34, 35.

29. Again the Western text begins the verse with what seems to be
a pointless paraphrase, for it is hard to see any real difference
between ἐπλήσθη (a Lucan word) ή πόλις τής συγχύσεως (in almost
all MSS) and συνεχύθη ὄλη ή πόλις (D* (gig syp)). D*, it is true,
surprisingly adds αισχύνης. This however has been plausibly
explained as due to the influence of d which (for the Greek
συγχύσεως) has confusionem. This word was often an equivalent for
αισχύνη.
ὁρμᾶν is used here and at 7.57, each time of a violent onrush of
people; elsewhere in the NT only of the Gadarene swine.
ὁμοθυμαδόν occurs 10 times in Acts, elsewhere in the NT only at
Rom. 15.6. As usual in Acts it suggests physical association as well
as unanimity. Luke appears to be freely writing up the information he
had received.
The θέατρον at Ephesus is the only one mentioned in the NT (also
at ν. 31). The Hellenistic theatre was an imposing building, now fully
excavated (see F. Miltner, Ephesos (1951(8)), 30-32). Estimates of
its capacity vary, but the lowest seems to be 24,000. An inscription
cited in Deissmann (LAE 113f.) from Jahreshefte der Österrei-
chischen Archäologischen Instituts 2 (1899), Supplement 43f., seems
to presuppose that meetings of the town ἐκκλησία (vv. 32, 39, 40)
were held in the theatre. In AD 103-4, C. Vibius Salutaris presented
a silver image of Artemis, together with other statues, ἵνα τέθηνται
κατ’ ἐκκλησίαν ἐν τῶ (sic) θεάτρω (sic) ἐπὶ τών βάσεων. The
corresponding Latin of the bilingual inscription runs, ... ita ut
[om]n[ie]cclesia supra bases ponerentur. Cf. AGIBM 3.481.395. Less
formal gatherings also took place in theatres. Thus AGIBM
4.792.4ff.: ὁ μὲν δᾶμος ἐν ού μετρίᾳ συγχύσει γενόμενος ... μετά
πάσας προθυμίας συνελθὼν εις το θέατρον; Cicero, Pro Flacco 7
(16): Cum in theatre imperiti homines, rerum omnium rudes
 49. RIOT AT EPHESUS. 19.21-40. 929

ignarique consederant: turn bella inutilia suscipiebant; turn seditiosos

homines reipublicae praeficiebant; turn optime meritos cives e
civitate ejiciebant.
The rioters seized and dragged along with them (συναρπάσαντες)
the two Macedonians Gaius and Aristarchus. The name Gaius recurs
at 20.4; for the textual question, on which the question whether the
Gaius there can be the Macedonian of this verse depends, see the
note. The singular Μακεδόνα (36 453 pc), which in the present verse
makes only Aristarchus a Macedonian, probably arose through
identification of Gaius here with Gaius the Derbaean of 20.4.
Aristarchus is mentioned at 20.4 as a Thessalonian and at 27.2 as a
Macedonian from Thessalonica; we need not doubt that the same
man is in mind in all these passages. An Aristarchus is mentioned at
Col. 4.10; Philemon 24. For the movements of Aristarchus and his
relation with Paul see on 27.2.
Gaius and Aristarchus were συνέκδημοι of Paul’s. The word is
used in a similar way at 2 Cor. 8.19, where the person concerned,
whose praise in the Gospel circulated in all the churches, had been
appointed by the churches (χειροτονηθείς ὑπό τῶν εκκλησιών) to be
Paul’s συνέκδημος in the matter of the collection that Paul was
organizing. The word, which in itself means simply fellow-traveller,
may thus have acquired a semi-official meaning, ‘travelling col-
league’, or the like. The word has a somewhat similar sense at
Josephus, Life 79 and Plutarch, Otho 5 (1068). In IG 12(8). 186 line 9
(Samothrace, first century BC) the Doric form συνέγδαμοι is used
‘of private persons accompanying a public mission’ (LS 1706). The
word seems very suitable for men who were not simple members of
the church but trusted and authorized assistants of Paul. Presumably
they were publicly known and thus natural targets for the mob’s
violence.

30. It was not Paul’s intention to avoid dangers to which his

subordinates were exposed; he wished εἰσελθεῖν εις τόν δήμον, that
is, to go into the theatre where the people (of the city) were
assembled. This however the disciples would not permit; Paul’s life
was too valuable to be risked in this way. There is no difference in
meaning between the Old Uncial and Western texts here, since in the
former the negative (οὐκ) and the verb (εἴων) combine to form what
is virtually a negative verb (BDR § 433.1), so that oἱ μαθηταὶ
ἐκώλυον (D syp) is equivalent to the text of the great majority of
MSS.
Begs. 4.248 has ‘δήμον is unlikely to mean assembly’; this was a
riot. Yet the same company of people becomes in v. 32 ή εκκλησία.
For the question whether these events are reflected in 1 Cor. 15.32;
16.9; 2 Cor. 1.8 see 1 Corinthians 365f.; 2 Corinthians 63f. The
answer seems to be, ‘Possibly in 1 Cor. 15.32.’
930 COMMENTARY ON ACTS

In this sentence Paul appears first in a genitive absolute, then

(αυτόν) as the accusative object of the main verb; Luke is not writing
in his best style.

31. It was not only Christian disciples who were concerned for
Paul’s safety. There were also officials who were ready if not to take
his part at least to advise caution. These were some of the Asiarchs.
The meaning of this term is disputed, and the question is complicated
by the fact that it seems to have changed in the course of time.
Literary evidence is meagre; inscriptional and numismatic more
plentiful, and there are analogous terms—Galatarchs, Bithyniarchs,
Lyciarchs. The main problem lies in the relation (if any) between
the office of Asiarch and that of High Priest of the cult of Rome in
the league (κοινόν) of Asia. ‘My explanation ... is that from the
Asiarchs designated in each year as the foremost men of Asia one
was chosen to act as high priest of the emperor, and then, as the
temples of the league were built, one was selected to serve at the
league temple in each city. Thus all Archiereis would have been
Asiarchs, but all the Asiarchs would not have acquired the distinction
of the highpriesthood. As the number of league temples grew, in time
there would have been a priesthood for every Asiarch, and the two
terms would thus come to be identical in meaning. It is possible that
this was already the case in the time of Paul, or at least in that of
Luke’ (L. R. Taylor, in Begs. 5.261; see the whole admirable
discussion, 256-62; also Sherwin-White 88-90). Basic information
is given by V. Chapot, La Province Romaine proconsulaire d'Asie
(1904), 482ff. It is interesting to note Martyrdom of Polycarp 12.2
(ἠρώτων τον Άσιάρχην Φίλιππον, ἵνα ἐπαφῇ τω Πολυκάρπω
λέοντα), but the passage gives no fresh information and may in any
case reflect conditions of a later date. Taylor’s discussion is too good
to be out of date, but some recent work should be noted. There is
more recent bibliography in Hemer (121f. and ND 1.82), and there is
a particularly important discussion (by R. A. Kearsley) in ND
4.46-55. Her last sentence is ‘All this indicates that the Asiarchy was
quite separate from the provincial high-priesthood, at least during the
period covered by the evidence considered above’ [fifty years after
Acts 19].
These Asiarchs were kindly disposed to Paul, ὄντες (D,
ὑπάρχοντες; it is hard to see any motive for a change in either
direction) αὐτῷ φίλοι (adjective rather than substantive because of
the dative: BDR § 190.1, n. 1).
For the construction (πέμψαντες ... παρεκάλουν) cf. Mt. 11.2
(where it is helped out by διὰ τῶν μαθητών αυτοῦ).
δούναι ἑαυτὀν εις τό θέατρον: Begs. 4.248 translates ‘to venture
into the theatre’. This gives a sense appropriate to the context, but the
words do not mean more than ‘go into ...’. Some have seen in
 49. RIOT AT EPHESUS. 19.21-40. 931

δούναι εαυτόν an allusion to Mk 10.45 (δούναι τήν ψυχήν αυτού).

This is most improbable.
Stählin (261) describes the Asiarchs as ‘Wächter über die Loy-
alität’. On the basis of this sort of observation Lüdemann (225)
writes, ‘Aus diesem Grunde muss die apologetische Erklärung derart
ergänzt werden, dass Lukas mit der Freundschaft der Asiarchen die
Weltläufigkeit des Christentums betont.’ From a different angle,
Luke may be said to be pointing out that Christianity in Ephesus
already had sympathy in the highest circles (Roloff 293). Differently
again, that the Asiarchs were Paul’s friends shows that he ‘spumed
the Pharisaic ideas of ritual purity which would have excluded such a
friendship’ (Ehrhardt, Acts 103).

32. μὲν οὖν: for this frequent construction, often, as here, resump-
tive, see 1.6; 15.3. After two verses dealing with Paul Luke returns to
his description of the crowd.
άλλοι... άλλο τι ἔκραζον: some were shouting one thing, others
another. For the construction cf. 2.12 and more precisely 21.34;
Xenophon, Anabasis 2.1.15, οὗτοι... άλλος άλλα λέγει.
ή εκκλησία. For the Christian use of this word see 5.11; Introduc-
tion, pp. lxxxviif. In its non-biblical sense it is correctly used at v. 39
(see the note) of the duly constituted assembly of citizens (at
Ephesus, held in the theatre; see v. 29); the use in vv. 32, 40 is
doubtful, since the assembly seems to be informal, unofficial, and
riotous, δήμος possibly (vv. 30, 33), ὄχλος (vv. 33, 36), and
συστροφή (v. 40) are more suitable terms. The use of εκκλησία is
however understandable; the persons concerned were those who
would have been summoned to a lawful assembly even though they
were not at the time engaged on lawful business.
συγκεχυμένη; cf. σύγχυσις (v. 29).
οἱ πλείους. It is hard to see more than a free rewriting, in which
the editor sometimes wrote from memory without checking his text,
in the variant οἱ πλεῖστοι (D). It is true that this superlative is
sometimes used for the best, noblest (people), and it is conceivable
that the Western editor meant that even the best people in Ephesus
did not understand what was going on; but it is doubtful whether so
much can be read out of the text.
τίνος ένεκα, on account of what, i.e., why. ένεκα most often
follows its case, but in the NT does so only here and at Lk. 4.18 (in a
quotation of Isa. 61.1, LXX).
After ἤδεισαν the perfect would have had the force of the
pluperfect. Thus συνεληλύθεισαν is a case of ‘Attraktion des
Tempus’ (BDR § 345).
For the verse as a whole cf. Charito of Aphrodisias 1.5.3: αλλά καὶ
ό δήμος άπας εἰς τὴν αγοράν συνέτρεχεν άλλων άλλα
κεκραγότων.
932 COMMENTARY ON ACTS

33. ἐκ δέ τοῦ ὄχλου. The preposition ἐκ is used in place of a

partitive genitive and the phrase serves as subject of the verb: some
of the crowd. For this (‘rather barbarous’—Begs. 4.249, but see BDR
§ 164.2, n. 6) construction cf. 21.16 (without ἐκ); Lk. 21.16; Jn 7.40;
16.17. A not impossible alternative is to take the subject of
συνεβίβασαν to be a simple ‘they’ and ἐκ του ὄχλου as a qualifying
adverbial phrase: They put forward out of the crowd ...
συνεβίβασαν P74 A B E 33 323 945 1739 1891 pc
κατεβίβασαν D* lat (detraxerunt; there are some strange Latin
equivalents)
προεβίβασαν Dc Ψ <a> vhmss syh
Of these words συμβιβάζειν occurs at 9.22 (affirm, allege; teach);
16.10 (conclude); also at 1 Cor. 2.16; Eph. 4.16; Col. 2.2, 19;
καταβιβάζειν at Mt. 11.23 (v. 1.) = Lk. 10.15 (v. 1.) (bring down—
to Hades); προβιβάζειν at Mt. 14.8 (put forward). The last of these
seems at first sight to give the sense required by the passage: some of
the crowd put forward Alexander, presumably to act as their
spokesman. This however would anticipate προβαλόντων (see
below), and it is hard to see why, if προεβίβασαν was the original
reading, nearly all the earliest witnesses, Old Uncial and Western,
should change it (into συνεβ. and κατεβ.). This argument will carry
with it the corollary that συνεβ. does not mean (as many translators
have taken it to mean) put forward. It is probably best to take the
word in the sense (which is that of the LXX—see on 9.22) of teach,
though here with the quasi-legal connotation of instruct; MM (s.v.
συνβιβ.; 603) give examples of the legal use of ἐκβιβ. Those
authorities that read κατεβίβασαν (detraxerunt) may have taken it
with the following clause: When the Jews put forward Alexander
some of the crowd brought him down. The general sense, however,
seems best served by: Some of the crowd instructed Alexander, and
the Jews put him forward. But it remains quite unclear who
Alexander was, why he was chosen, and what he was expected to do
or to say. See further below.
προβαλόντων αυτόν τών ’Ιουδαίων undoubtedly means that the
Jews (for Jews in Ephesus see on 18.19) put forward Alexander. In
what sense and for what purpose? For the role of the Jews in the
whole episode see on v. 40. The next words show that the crowd in
general was unfavourably disposed to the Jews, of whom Alexander
was one, and we may probably accept the meaning ‘to cite in one’s
defence’; see Isaeus 7.3 (μάρτυρας προὐβαλόμην); Plato, Laches
201b; Demosthenes 18.149 (277), though in none of these passages
is the verb active. But why else should the Jews have put forward one
of their own number? The Jews would be under attack (a) because
they were not always differentiated from the Christians, who had
provoked the mob, and (b) because in any case they were known to
be opposed to idolatry, and were unpopular. See on v. 40.
 49. RIOT AT EPHESUS. 19.21-40. 933

κατασείσας. The word is characteristic of Acts (12.17; 13.16;

21.40). In all the other passages it is used with the dative, τῇ χειρί,
here with the accusative direct object, τὴν χεῖρα. The dative is ‘more
frequent’ (LS 910).
ἀπολογείσθαι is in the NT a Lucan word (Acts 24.10; 25.8; 26.1,
2, 24; Lk. 12.11; 21.14; elsewhere in the NT, twice; cf. ἀπολογία,
Acts 22.1; 25.16), to make a defence. It is implied that the Jews have
been attacked (Weiser 547—they felt themselves threatened by the
attack on Paul). τω δήυῳ. Cf. v. 30.
The name Alexander occurs at 4.6; this cannot be the same man.
No identification is offered. The name is found at 1 Tim. 1.20
(Hymenaeus and Alexander have made shipwreck concerning the
faith and have been delivered to Satan that they may learn not to
blaspheme) and at 2 Tim. 4.14 (Alexander the coppersmith did me
much harm; he will get his deserts; Timothy must beware of him).
The Pastoral Epistles are probably pseudonymous, but this does not
mean that all the names contained in them are purely fictitious. It
would however be rash to identify any of these Alexanders; see
Pastorals 47f., 121. A more plausible suggestion (Klausner, Jesus
351) connects the present passage with Mk 15.21 and Rom. 16.13.
There was little point in mentioning the names of Simon’s sons
unless they were known in Christian circles. The suggestion would
gain in force if the view were accepted that Romans 16 was
addressed not to Rome but to Ephesus; this however is unlikely
(Romans 257f.). And we do not know that Alexander of Acts 19 was,
or became, a Christian. See further on v. 34.

34. ἐπιγνόντες: a sense construction—those who recognized

uttered the φωνή (below). This verse still leaves us guessing who
Alexander was, why the Jews put him forward, and to what end his
defence (ἀπολογείσθαι) was directed, but we do learn that he was
himself a Jew. The name was borne by Jews (Jastrow 70), notably by
an Amora of the third century. The crowd presumably recognized
that he was a Jew by the line taken in his defence.
φωνή ἐγένετο μία ἐκ (omitted by D lat) πάντων, one cry arose
from them all, that is, from the mass of the mob, who were not Jews.
It is not at this point clear why the Jews were brought into the
episode; see below. The separation of μία from φωνή stresses the
word and thus the unanimity of the crowd—who behave with the
irrationality still characteristic of large assemblies.
ὡς, about, approximately; see LS, s.v. E, 2039. In Acts this is
more often expressed by ὡσεί (see 1.15), which is read here by P74 B
33 pc.
κράζοντες, read only by A, should probably be accepted, as
lectio difficilior; it is easy in view of πάντων to understand the
omission of the word by Ψ and the change by all other authorities to
934 COMMENTARY ON ACTS

κραζόντων (though Ropes (Begs. 3.188) thinks that the reverse is

what happened). See BDR § 466.4, n. 5. The nominative takes up the
case of ἐπιγνόντες, but this does nothing to remove the anacolouthon
since the only finite verb in the sentence is ἐγένετο and its subject is
φωνή. There is no reason why Luke should not have written (BDR,
loc. cit.) ἐπιγνόντες ... ἐβόησαν ὁμοῦ πάντες ... κράζοντες.
μεγάλη ή Άρτεμις Έφεσίων. See ν. 28. Β repeats the clause. ‘Die
Epanadiplosis, d.h. die nachdrückliche Verdopplung eines gewichti-
gen Wortes, ist dem NT nicht unbekannt, aber nirgends als rhetorisch
zu betrachten, sondern überall als Wiedergabe der wirklichen Rede’
(BDR §493.1).
C. Burchard (ZNW 61 (1970), 167f.) quotes T. Benjamin 3.7 (α
and β), περιλαβών αυτὸν ἐπί δύο ώρας κατεφίλει, and Τ. Judah 3.4
(β), πολεμήσας τον Άχώρ ἐπί ώρας δύο ἀπέκτεινα αυτόν, and
asks, ‘Eine redensart?’
In vv. 33, 34 we have noted some unusual, in some cases also
difficult, expressions.
ἐκ τού όχλου as subject
Unusual sense of συμβιβάζειν
An otherwise unknown (and unexplained) Alexander
Unexplained action of the Jews
Unexplained defence (against what?) by Alexander
Anacolouthon: ἐπιγνόντες ... φωνή ἐγένετο ... κράζοντες
(κραζόντων)
We may add the observation that v. 32 could be followed imme-
diately by ν. 35, i.e., vv. 33, 34 form a self-contained unit which
could be removed without leaving a noticeable gap. It seems worth
while at least to raise the question whether these verses contain
supplementary information added to a source that gave an outline of
the events at Ephesus; possibly Luke supplemented by local inquiry a
narrative given him by a travel document (or similar source); or
perhaps he had his own reason for making the addition. A probable
consequence might be that Alexander either was at the time of the
riot, or subsequently became, a Christian.

35. For the connection see the note on v. 34. This verse could be
linked directly with v. 32.
καταστέλλειν is used in the NT only here and in v. 36. Cf.
Josephus, War 2.611, καταστείλας τόν θόρυβον αυτών; 4.271; Ant.
14.156; 20.174; 2 Macc. 4.31. It has the same meaning (to restrain)
in the papyri, e.g. BGU 4.1192.5, τών ’Αράβων κατεσταλμένων καί
πάντων ἐν τήι μεγίστη[ι] εἰρήναι γεγονότων. The rare occurrence of
the word accounts for its replacement by κατασείσας (D E Ψ 614
pc), used above in v. 33.
ὁ γραμματεύς. This title of an official at Ephesus (as well as at
 49. RIOT AT EPHESUS. 19.21 –40. 935

other Greek cities, including Athens) is confirmed by inscriptions.

The noun occurs in Dittenberger, OGIS 2.493.11, Λούκιος ...
άποδε[δει]γμένος γραμματεύ[ς τού] δήμου, the corresponding verb
at 510.11, γραμματεύοντος Ποπλίου Ούηδίου Άντ[ω]νείνου
Άσιάρχου (cf. ν. 31). These inscriptions come from the second
century AD, but there is no reason why the office should not have
been in existence in the middle of the first. See Sherwin-White (86f.).
The word itself was much older; see e.g. Thucydides 7.10. Hanson
(197) quotes A. Η. Μ. Jones (The Greek City (1940), 181). It seems
that some of the wealth of Artemis got into the city treasury and that
the town clerk might not have welcomed an inquiry.
ἄνδρες Έφέσιοι. For this mode of address see 1.11.
τίς γάρ ... ; γάρ is ‘inde ab Homero saepe in interrogationibus,
fere ut german. wer denn' (Blass 212). See BDR § 452.1, n. 2. Either
it simply emphasizes the question (cf. Lk. 22.27, and see below on
ανθρώπων), Why, what man is there ... ?, or it assumes an
unexpressed accusation or admonition, You really must restrain
yourselves, for who is there ... ? A variant on this might be, I have
tried to quieten you, for who ... ? ανθρώπων is emphatic: Where can
you find anyone in the whole human race who does not know ... ?
The verb of knowing is followed by the accusative and participle
(οὖσαν) except in D where the infinitive (είναι) is used in place of the
participle, possibly through the influence of the Latin of d (esse).
As subject of the dependent clause τήν Έφεσίων πόλιν makes
good sense, but D has τ. ἡμετέραν πο., which comes perhaps more
naturally on the lips of a speaker who is using local patriotism as a
means to his end of quietening the crowd. Έφεσίων could have been
introduced from the end of the preceding verse. The reading of d,
vestram, plausible in itself, for the town clerk could so address the
townspeople, is probably an internal Greek variant, since ὑμετέραν
could come from ἡμετέραν even more easily than vestram from
nostram.
The city is νεωκόρον τής μεγάλης Άρτέμιδος. For Artemis and
the adjective great see vv. 28, 34. νεωκόρος meant originally
warden of a temple; the cognate verb is used by Josephus (War
5.389, πάλιν τόν αυτών σύμμαχον ἐνεωκόρουν, they (the Jews after
the Exile) re-established the temple worship of their Ally (God)). The
noun became a ‘title assumed by Asiatic cities in Imperial times,
when they had built a temple in honour of their patron-god or the
Emperor’ (LS 1172), in Ephesus, of Artemis. See Sherwin-White
(88f.); Hemer (122). The present seems to be the oldest use that we
have. See Dittenberger, OGIS 2.481.3; 496.7; BMuslnscr
481*.4—all of the second century AD.
Ephesus was thus temple warden of Artemis και τού διοπετοῦς.
διοπετής is an adjective but it seems scarcely necessary to speak of
the ellipse of a noun (άγαλμα), with Μ. 3.17; LS 433. It is Greek
936 COMMENTARY ON ACTS

idiom to turn the adjective into a substantive: that which fell down
from Zeus. To supply άγαλμα is over-precise, though it is used in
Euripides, Iphigenia in Tauris 977f. (at 87f., 1384f. διοπετής is not
used). The object was presumably some kind of meteorite, having
perhaps human form. Begs. 4.250 makes the point that such objects
(also e.g. the Palladium at Troy, the Minerva Polias at Athens, the
Cybele at Pessinus) could be used as a counter to Jewish and
Christian attacks on paganism as the worship of objects of human
manufacture, and quotes Cicero, In Verrem 2.5. 72 (187), Sim-
ulacrum Cereris ... quod erat tale, ut homines, cum viderent, aut
ipsam videre se Cererem, aut effigiem Cereris non humana manu
factam sed de caelo lapsam [caelo delapsam] arbitrarentur. See also
Herodian, 1.11.1. But the religious evaluation and use of meteorites
must have long preceded this polemical interest. For other parallels
see Betz (168, n. 2).
D has διοσπετοῦς, which makes clearer that the object not merely
fell from the sky (cf. άγαλμα διοπ. in Iph. in T. 1384f., referred to
above) but from Zeus himself. It also probably explains the surpris-
ing Vulgate reading, ‘iovisque prolis’, which WW suggest could be
derived from τοῦ Διὸς παιδός (written as πεδός). d has huius iovis,
hcl mg has her διοπετής The words διοπετής, διοσπετής, were
probably not well understood.
36. ἀναντιρρήτων, passive: since these matters are not open to
contradiction. Contrast 10.29, ἀναντιρρήτως, active: without gain-
saying. Presumably Paul and his colleagues would have contradicted
if they had been given the opportunity to do so.
κατεοταλμένους. For the word see v. 35. ὐπάρχειν rather than
είναι, perhaps in order to suggest remain quiet.
προπετές. Cf. 2 Tim. 3.4, the only other occurrence in the NT. But
the word is not uncommon with the meaning rash, precipitate.
37. ἱεροσύλους: classical usage would have had a participle with
the predicate; Μ. 3.159; BDR § 418.6, n. 9. For the cognate verb
ἱεροσυλεῖν cf. Rom. 2.22. According to Betz (186 n. 3) the word ‘ist
oft ein Schimpfwort’; here however it will have its proper sense in
view of the charge brought against Paul and his companions, which
is virtually that of profaning the temple of Artemis by alleging that
she herself was not a deity, and that if she were the worship offered
to her was improper. In this sense Paul was a ιερόσυλος, and he was
engaged in speaking evil of (βλασφημοῦντες) the goddess. Thus
Chrysostom (Homily 42.2) of the town clerk’s words, Άρα τό πᾶν
ψεύδος, ταῦτα δέ πρὸς τὸν δήμον ώστε κἀκείνους ἐπιεικεστέρους
γενέσθαι, φησίν; he is putting the best possible appearance on the
matter so as to quieten the crowd. Paul and his colleagues were not
profaning or robbing the temple or blaspheming the goddess in a
vulgar way. This was the ‘correct’ Jewish attitude; see Begs. 4.251,
 49. RIOT AT EPHESUS. 19.21-40. 937

with reference to Josephus, Apion 2.237; Ant. 4.207, βλασφημείτω

δὲ μηδεὶς θεούς οὓς πόλεις ἄλλαι νομίζουσι. μηδὲ συλᾶν ιερά
ξενικά. See however Apion 1.249, 311. Cf. also Bengel (466):
‘Apostoli non collegere multa absurda ex mythologia, sed propo-
suere veritatem DEI.’
After τούτους, D 614 syhmg co add ἐνθάδε, unnecessarily.
For θεόν, D* Ec P 614 1241 2495 al have θεάν. On the use of θεός
and θεά for female deities see on v. 27.
For ἡμῶν, E* <a> vg syh bo have ὑμῶν. Editors and copyists
thought it might be well to suggest that the town clerk, who evidently
has to be counted as one of the (at least, relatively) good should be
distinguished from the wicked and idolatrous Ephesians.
38. Δημήτριος ... τεχνῖται. See on v. 24.
ἔχουσι πρός τινα λόγον, have a case (law-suit) against anyone,
that is, a private suit against individuals, as Demetrius and his
colleagues might seem to have against Paul and his. This takes due
note of πρός τινα, and of the contrast expressed in the next verse.
See Sherwin-White (83). The force of the sentence is not sig-
nificantly changed if, with D (gig) samss, we read πρὸς αυτούς τινα
λόγον, have any case against them. See v. 39.
For αγοραίοι, αγοραίος, see on 17.5. In the present verse it is
clear that we must take ἀγοραῖοι as the adjective with ἡμέραι or
σύνοδοι understood; it is equivalent to the Latin fora, or conventus,
aguntur. The town clerk points out to Demetrius that there are courts
in which private suits may be presented in due form of law and that
there is therefore no occasion for riotous assembly. See BDR § 5.4,
n. 21; Moule (IB 192). ‘Ephesus was capital of one of the conventus,
or assize-districts’ (Hemer 123).
Courts are held, and ανθύπατοί εἰσιν. ‘There are such persons as
proconsuls’—before whom such matters may be brought. There was
of course only one proconsul of the province of Asia; the plural is
pluralis categoriae (Μ. 3.26)—Luke is not to be accused of
ignorance of an elementary and universally known fact, nor is it
likely that there is a reference to the period after the murder of Μ.
Junius Silanus (Tacitus, Annals 13.1; Dio Cassius 61.6) when Helius
and Celer (whom Agrippina had employed in the murder) were
temporarily in charge in Asia.
39. εἰ δέ τι περαιτέρω επιζητείτε. The town clerk now addresses
Demetrius and those associated with him directly, in the second
person. The sense is correctly given by Sherwin-White: ‘If they are
after something more than a private lawsuit’ (83). τι περαιτέρω is the
reading of P74 B 33 36 453 945 1739 1891 pc d gig (alterius); AD
(Ε) Ψ <a> (vg) syh co have περί ετέρων (the Vulgate’s ‘alterius rei is
an attempt to represent περί ἑτέρων without departing too far from
the Old Latin rendering’ (Ropes, Begs. 3.189)). This widely attested
938 COMMENTARY ON ACTS

reading may have arisen by accident (at being written, as often in the
MSS, ε); it goes too far to say that it does not suit the context
(Metzger 473).
Anything beyond a private civil action will be dealt with ἐν τῇ
ἐννόμω ἐκκλησία. See ν. 32. A similar expression occurs in an
inscription (AGIBM 3.481.339-40: κατά πάσαν νόμιμον
ἐκκλησίαν. ‘According to the latest restoration of the main document
(1. 54) there was one monthly meeting called ἱερὰ καὶ νόμιμος
ἐκκλησία. Chrysostom in his commentary [Homily 42.2] said that
there were three monthly meetings. Presumably there was one
regular and two extra meetings a month’ (Sherwin-White 87). This
otherwise reasonable inference does not take account of the fact that
customs may have changed between Luke’s time, in the first century,
and Chrysostom’s, in the fourth. It appears that in any case Deme-
trius would not have had to wait long for a regular meeting. The
present meeting was clearly not ἔννομος. This word can be further
documented: Lucian, Deorum Concilium 14, ἐκκλησίας ἐννόμου;
Dittenberger, SIG 2. 672.37, ἐν τᾶι ἐννόμωι ἐκκλησίαι; 2.852.20
(Hemer 123), ἀγομένης ἐκκλησίας ἐννόμου.
For ἐπιζητειτε, P74 Ε 36 453 945 1739 1891 pc have ζητείτε,
without difference of meaning.
For τῇ ἐννόμω, D* has τῷ νόμῳ, which is meaningless and must
be a slip.

40. The syntax of this verse is very confused, and it is not

surprising that there are several variants. Hort supposed that there
had been a primitive corruption of the text; so also, tentatively,
Ropes (Begs. 3.189). See Weiser (549) for a good survey of various
views of the conclusion of the episode.
The demonstrating artisans, and the whole city with them, run the
risk, κινδυνεύομεν, of being accused (ἐγκαλεῖν, as in v. 38). Cf.
Begs. 4.252, ‘... as though the secretary meant to say the real risk is
not in loss of business but trouble with the police for disturbing the
peace’. στάσις is a strong word (in the στάσις of Mk 15.7 murder had
been committed); riot. In classical usage that of which a party was
accused was expressed in the accusative, but the genitive case tended
in the Hellenistic age to oust the accusative (Radermacher 99).
Moule (IB 39), however, notes that this would be the only occurrence
of this construction in the NT; in 26.6, 29 περί with the genitive is
used. He doubts whether the clause can be construed in this way, but
offers no alternative. In the present verse περί introduces not the
charge but the ground of the charge, in (the events of) this day—
περί τής σήμερον.
μηδενὸς αιτίου ὑπάρχοντος (D, ὄντος, without substantial differ-
ence in meaning). Taken as a unit (genitive absolute) this is
intelligible enough: there being no cause (for it, for that which has
 49. RIOT AT EPHESUS. 19.21-40. 939

happened today). Elsewhere αίτιον may mean crime (perhaps as the

ground for punishment), but this meaning (negatived by μηδενός) is
impossible here. See Μ. 2.341. Overbeck, referred to by Knowling
(419), takes αιτίου as masculine, ‘there being no man guilty’. Cf. the
Vulgate, cum nullus obnoxius sit... This may to some extent avoid
the difficulty of the negative.
The next clause contains several difficulties. (1) What is the
antecedent of the (masculine or neuter) relative ού? It cannot be
στάσεως or τής σήμερον, both of which are feminine; presumably it
is αιτίου. (2) Taking this to be so, what is meant by saying
‘concerning this (ground?), we shall not be able to render account?’
Perhaps the town clerk means, ‘If we do not at once disperse quietly,
we shall not ...’ Another possibility is that μηδενός αίτιου
ὑπάρχοντος, which was taken above to refer to what precedes, may
be taken in close association with what follows: there being no cause
(or even crime) concerning which we shall not be able to give an
account. This however becomes impossible in view of the next
phrase, περί τής συστροφῆς ταύτης, concerning this concourse (cf.
23.12, the only other use of συστροφή in the NT, where the meaning
is conspiracy). The omission of περί makes no difference: an
account of is the same as an account concerning. The omission of οὑ
(by P74 D E 33 36 453 945 1175 1891 al gig vg sa bomss) does ease the
sentence: There being no reason concerning which we shall be able
to give an account of this concourse. To this extent the omission must
be regarded as lectio facilior and unlikely to be correct; but the
possibility of accidental dittography or haplography (OYOY) is so
strong that no confidence can be felt in either the longer or the shorter
reading.
Hort’s conjecture of primitive corruption is understandable. ‘The
difficulty is ... too great to allow acquiescence in any of the
transmitted texts as free from error. Probably αίτιοι ὑπάρχοντες
should be read for αιτίου ὑπάρχοντος, with the construction
μηδενὸς αίτιοι ὑπάρχοντες περί ου ού κ.τ.λ. (“although we are
guilty of nothing concerning which’’ etc.). The usage of the NT
admits this use of μή with a participle, and the interchanges of I and
Y, E and O, in uncials are of the commonest’ (Introduction: Notes
97). Unfortunately Hort does not explain how he takes περί τής
συστροφής ταύτης. There is very much to be said for C. F. D.
Moule’s view (ExpT 65 (1954), 221) that the author wrote the
sentence in several forms, and omitted to delete the rejected words
before publication. Even this however will hardly account for the two
fundamentally different impressions (not translations) which it is
possible to derive from the verse. The town clerk is evidently seeking
to achieve the result described in the last words of the verse: he
wishes to send everyone home in peace, but he may be arguing: (1)
We are running the risk of being accused of riot and insurrection, and
940 COMMENTARY ON ACTS

there will be no defence, for there is no rational ground or excuse for

what has happened; that is, we shall get what we deserve; or (2) We
are running the risk of being accused of riot and insurrection, even
though we have not in fact done anything for which we cannot plead
that there was just cause; punishment inflicted by the proconsul will
then be undeserved, but this will make it no less unpleasant.
It may be (1) that there has been corruption of the text (Hort); (2)
that Luke never revised his original draft (Moule); (3) that Luke was
not clear in his own mind how the situation should be viewed; (4)
that the town clerk wanted to ‘have it both ways’ and used
ambiguous arguments designed to appeal to different sections of the
crowd.
Whichever view we take there is an ‘atmosphere of indignant
defence of the city’s privileges and reputation’ (Sherwin-White 84).
Sherwin-White compares the description in Acts with the civic
speeches of Dion of Prusa [= Dio Chrysostom] (Orations 46.14;
48.1-3; 38.38). He adds, ‘There is something rather ominous in this
tone. This was the last age of civic autonomy in the ancient world.
Civic politics in the old pattern of the city-state, with its assemblies
and councils, expired in the course of the later second and early third
centuries AD. The city councils became closed hereditary oligarchies
... The scene belongs unmistakably to an era that did not survive the
age of the Antonines ... The evidence of Acts not only agrees in
general with the civic situation in Asia Minor in the first and early
second centuries AD, but falls into place in the earlier rather than the
later phase of development’ (84f.). These observations are interesting
and important but do not greatly affect the dating of Acts; no one
would wish to make it later than the age of the Antonines.
An additional aspect of the narrative is suggested by R. F. Stoops
in an important article (JBL 108 (1989), 73-91). Its main substance
may be conveyed by the following paragraph. ‘The speech of the
town clerk reinforces the apologetic thrust of the passage. This
speech identifies opposition to the Way, rather than the Way itself, as
the source of trouble and the threat to the established order. The
Gentile residents of Ephesus are cautioned against acting the part of a
political assembly without being properly constituted. The Way and
its local assembly (ἐκκλησία) are not condemned. The silence of the
town clerk on this point implies that the ἐκκλησία of believers,
unlike Demetrius’s mob, can coexist with the legitimate assembly of
citizens in Ephesus. The story suggests that the assembly of believers
ought to be left alone and allowed to conduct its own affairs. That
seemingly small request was an extraordinary privilege in the Roman
world, one that had set the Jews apart from other groups’ (88f.).
Stoops makes the good point that the apologetic element in Acts was
not addressed primarily to Roman officials but to Christians. ‘Those
readers had a concrete need for self-definition and reassurance. Acts
 49. RIOT AT EPHESUS. 19.21-40. 941

is not a handbook of debating points but a presentation of Luke’s

understanding of the world and the place of the community of
believers within that world’ (90).
ἀπολύειν is used in a variety of senses in the NT and elsewhere;
here to dismiss; cf. Josephus, Ant. 11.337, απέλυσε τό πλήθος.
 50. BACK TO PALESTINE, THROUGH MACEDONIA,
GREECE, AND TROAS 20.1-16

(1) After the uproar had ceased Paul sent for the disciples,1 gave them an
exhortation, said farewell, and set out to travel to Macedonia. (2) When he
had passed through those parts and2 exhorted them in much preaching he
came into Greece; (3) when he had spent three months [there]3 and a plot was
made against him by the Jews as he was about to leave for Syria, he made up
his mind to return through Macedonia. (4) There were associated with him
Sopater the son of Pyrrhus, a Beroean, of the Thessalonians Aristarchus and
Secundus, Gaius4 of Derbe and Timothy, and the Asians Tychicus and
Trophimus. (5) These5 went on ahead and waited for us in Troas. (6) We
sailed from Philippi after the Days of Unleavened Loaves6 and came to them
at Troas after five days; there we stayed7 seven days.
(7) On the first day of the week,8 when we had gathered to break bread,
Paul, since he was about to leave on the morrow, discoursed with them and
prolonged his speech till midnight. (8) There were many lamps9 in the upper
room in which we had met, (9) and a young man, Eutychus by name, sitting10
by the window, was being gradually overcome by deep sleep11 as Paul
discoursed longer and longer, until having been finally overcome by sleep he
fell12 from the13 second floor, and was picked up dead. (10) Paul went
down,14 fell on him, embraced him, and said, ‘Stop making a disturbance, his
life is in him.’15 (11) He went up, broke bread, and ate; he conversed further
until dawn, and so left. (12) They took up the boy alive, and were16 no little
comforted.
(13) We went on ahead to the ship and set sail for Assos, where we were to
take up Paul, for so he had17 given orders, since he himself intended to go by

1NEB, after encouraging them; NJB, after speaking words of encouragement to

them.
2RSV, given them much encouragement; NEB, after speaking words of encourage-
ment; NJB, he said many words of encouragement.
3Greek omits.
4NEB, the Dobenan.
5NJB, they all.
6NEB, Passover season.
7NEB; NJB, a week.
8NEB, Saturday night.
9RSV, lights.
10NEB, on the window-ledge; NJB, on the window-sill.
11RSV, sank into a deep sleep.
12NJB, to the ground three floors below.
13RSV, third story; NEB, third storey.
14RSV, bent over him; NEB, threw himself upon him.
15NEB; NJB, there is still life in him.
16NEB, immensely; NJB, greatly encouraged.
17RSV; NJB, arranged; NEB, made this arrangement.
942
 50. BACK ΤΟ PALESTINE. 20.1-16. 943

land. (14) When he met us at Assos we took him up and came to Mitylene.
(15) Thence we sailed, and on the following day came opposite Chios; on the
next day we reached Samos, and18 the day after that we came to Miletus. (16)
For Paul had chosen to sail past Ephesus in order that he might not have to
spend time in Asia, for he was making haste so as to be, if possible, in
Jerusalem on the Day of Pentecost.

Bibliography
Μ. Black, as in (46).
H. Conzelmann, ZNW 45 (1954), 266.
E. Delebecque, Bib 64 (1983), 556-64.
C. J. Hemer, FS Bruce (1980), 9-12.
P. W. van der Horst, NovT 17 (1975), 158.
B. Lindars, in C. F. D. Moule (ed.), Miracles, London 1965, 61-79.
S. Μ. Praeder, NovT 29 (1978), 193-218.
H. Riesenfeld, FST. W. Manson, 210-17.
W. Rordorf, ZNW 68 (1977), 138-41.
R. Staats, ZNW 66 (1975), 242-63.
G. Zuntz, Gnomon 30 (1958), 26.

Commentary
The structure of this section is clear. The first six verses describe the
opening stages of Paul’s roundabout return from Ephesus to Pales-
tine. When the uproar in Ephesus is over, he leaves for Macedonia,
addresses the Christians in those parts, and makes his way to Greece.
After spending three months there he determines to sail for Syria; his
plan however is changed by a Jewish plot, and he decides to return by
the way he came, through Macedonia. In v. 4 we learn the names of
men accompanying Paul, but no sooner are they named then they (or
some of them) separate from him, go ahead, and wait for us—for
now the first person plural is re-introduced—in Troas. We sailed
from Philippi (reached, it seems, though this is not quite definite, by
land).
Paul and his companions are thus reunited with one another in
Troas. Here there are Christians (not mentioned in v. 6, but the fact
becomes evident as the story proceeds), and the travellers join them
in the common meal. This becomes the occasion of a miracle. There
is nothing in the story of Eutychus to connect the event specifically
with this time and place; it could have happened anywhere, and it has
no necessary connection with the sequence of events described in the
preceding and following verses. It gives the impression of being a
18NJB adds: after stopping at Trogyilium.
944 COMMENTARY ON ACTS

free piece of tradition which Luke had some reason to connect with
Troas (perhaps he heard it there) and interpolated into the record of
the journey. This is resumed in v. 13. Up to the end of v. 15 there is
nothing but geography, as the party, now united and travelling by
sea, touch at Assos and Mitylene, sail near Chios, and touch again at
Samos and Miletus. This means that Ephesus is by-passed, appar-
ently because Paul had decided (κεκρίκει, v. 16) on this course.
Whether Paul as a passenger could decide the ship’s route and
stopping-places seems questionable; he may well have been glad to
find that the ship did not intend to call at Ephesus, though whether
the motive for haste that is supplied is convincing raises further
questions; see the note on v. 16.
Of the opening sub-paragraph Conzelmann (115) writes, ‘Die
Liste zeigt, dass Lk altes Material besitzt. Aber die Ereignisse sind
wieder vereinfacht.’ Conzelmann does not say by whom—by Luke
or at some earlier stage in the tradition—they were simplified. Nor is
it clear exactly what this means. A writer intent on simplification
could have achieved a better result than appears in vv. 1-6, where
some points are by no means clear. Conzelmann refers to relevant
passages in the Corinthian letters (1 Cor. 16.5; 2. Cor. 2.12ff., 7.6ff.),
but here we have not so much simplification as simple omission.
Greece, an imprecise term, is mentioned, but Corinth is not, and
Paul’s complicated and often agonized dealings with the church there
after the close of his first visit are not mentioned. One must suppose
that in the three months of v. 3 the Corinthian problems were solved
at least to the extent that Achaea contributed to the collection (Rom.
15.26; see 2 Corinthians 21, 25-28). Notwithstanding the first
person plural in vv. 5, 6, 7, 8, 13, 14, 15, these passages are often
distinguished from the other We-passages, and it is at least a
plausible suggestion that We in vv. 7, 8 has been brought in from the
surrounding travel notes. It is rightly observed that in the Eutychus
story as a whole interest focuses upon Paul, his companions having
no part to play. This is inherent in the form and substance of the
story, but it is probably correct to deduce from the absence of We and
the fact that the Eutychus story interrupts the connection between vv.
6 and 13, that it comes from a different source (probably local
tradition; see above). Weiser (557), modifying and developing
Conzelmann’s view, thinks that behind the travel notices lies ‘Uber-
lieferungsgut ... wahrscheinlich in Form eines Itinerars’. It is not
easy to see why these first person plurals (except perhaps those in vv.
7 and 8) should be distinguished from those in chs. 16 and 27.
‘Auszuschliessen ist auch nicht die Vermutung, dass Titus das
Reisetagebuch führte; dieser Vermutung mag man zuneigen wenn
man die Differenzen der “Wir-Berichte” von der zweiten Mis-
sionsreise und der Jerusalemreise entsprechend bewertet’ (Pesch
2.187). This is hardly convincing, and it is a striking fact that in vv. 5,
 50. BACK TO PALESTINE. 20.1-16. 945

6 the new We is taken up at Philippi, where the previous We ceased.

It is no new suggestion, but there is something to be said for the view
that the writer of the Itinerary had remained in Philippi in the
interval. For this part of Paul’s travels see especially Thornton
(146-8, 255-8).
The concluding sub-paragraph (vv. 13-16) sees the party down
the west coast of Asia Minor as far as Miletus, Paul, apparently alone
and for no given reason, making the first part of the journey by land.
For Paul’s decision not to visit Ephesus see the note on v. 16. It may
be that Luke is concealing the fact that Paul did not dare to show his
face in Ephesus, and that the verse is part of the editor’s attempt to
play down the measure of opposition to Paul’s work there, though it
is hard to see why he should wish to minimize the perils endured by
his hero. It is certainly true that the motive given for by-passing
Ephesus—a strong desire to observe the feast of Pentecost in
Jerusalem—can hardly fail to evoke scepticism. At least the conclud-
ing note (v. 16b) seems clearly to be an editorial supplement to a
travel document, which at this point is little more than a list of
names.

1. θόρυβος properly denotes sound—not merely tumult but

clamour, and there is no reason why it should not have this meaning
here. After the uproar had ceased Paul took up again the intention
mentioned in 19.21.
μεταπεμψάμενος is the reading of P74 Β E 33 36 453 1175 pc,
προσκαλεσάμενος of A D Ψ <a> latt sy, μεταστειλάμενος of 945
1739 1891 pc. The words all have substantially the same meaning.
μεταπέμπεσθαι occurs 9 times in Acts, nowhere else in the NT—a
Lucan word; μεταστέλλεσθαι occurs here only; προσκαλεισθαι
occurs 9 times in Acts, 4 times in Lk., but 16 times in the rest of the
NT. One is inclined to accept μεταστειλάμενος as inviting assimila-
tion to the more usual words, but the support is very slight.
μεταπεμπ. is more likely to have suggested μεταστειλ. than
προσκαλ. and should probably be accepted.
παρακαλέσας: see on 2.40 and elsewhere. There was no need for
Luke to add a clause explaining what Paul exhorted them to do; he
gave them a Christian exhortation. D*vid, πολλά παρακελεύσας, is
probably not more than a slip in writing, d has multo exhortatus, and
according to Ropes (Begs. 3.191) the first hand of D has
παρακελε[ύ]σας; doubt about the υ supports παρακαλέσας as the
intended reading.
ἀσπασάμενος must here mean said Goodbye (that is, saluted in
farewell, as at Euripides, Trojan Women 1276, ὡς ἀσπάσωμαι τήν
ταλαίπωρον πόλιν), though the word much more often means a
greeting or welcome. The three participles μεταπ ... παρακαλ.,
ἀσπασ. all precede the finite verb ἐξῆλθεν in time.
946 COMMENTARY ON ACTS

πορεύεσθαι is omitted by D E 323 945 1739 2495 pc gig bomss; no

doubt it seemed unnecessary to the narrative. There are however two
distinct thoughts: he went out of the city of Ephesus, and he set out
on the long journey to Macedonia. For Macedonia see 16.9.
2. διελθών. Cf. 19.21.
(πὰντα, D) τὰ μέρη ἐκεῖνα. Luke does not indicate what districts
he had in mind; presumably those of Macedonia through which Paul
would pass on his way to Greece.
παρακαλέσας αυτούς (χρησάμενος, Dvid; an unusual word in
such a setting in Acts; possibly therefore original?) λόγω πολλῷ. If a
single occasion in one place were in question, one would say, with a
long speech; since work in various areas (μέρη) is in mind the sense
must be much speaking, much preaching.
Moving round the north-western angle of the Aegean, Paul came
into Greece, εις τήν Ελλάδα. The article is used (contrast
Μακεδονίαν, anarthrous, in v. 1) because Ελλάς was originally
adjectival (supply γη or χώρα)—BDR § 261.4. “Ελλάς steht ...
volkstümlich für “Achaia” ’—Pausanias 7.16 (Schneider 2.280).
3. ποιήσας τε μήνας τρεις. For this use of ποιεῖv cf. 15.33; 18.23;
also 2 Cor. 11.25; Jas 4.13. It occurs outside the NT, though mostly
in later Greek (e.g. Josephus, Ant. 6.18 (ποιήσασα ... μήνας
τέσσαρας); also PIand 6 (1934) 97.7, quoted in ND 4.63-67). This
seems to be the occasion when Paul paid his third visit to Corinth (2
Cor. 12.14; 13.1) and settled the Corinthian problems at least to the
extent that Achaea joined in the collection that Paul was organizing
(Rom. 15.26). He was now on his way to Jerusalem, bearing, it
seems, the proceeds of his appeal. The visit to Corinth seems to have
been a difficult one, difficult and stormy even if in the end
triumphant. Problems and success are alike passed over in Acts.
Because Luke thought them unedifying and suppressed them?
Because he was not aware of them? These are questions to which
only hesitant answers can be given, and that when all such questions
have been collected and reviewed. The silences of Acts are an
important factor in any discussion of its authorship.
γενομένης ἐπιβουλῆς ... ύπό τών ’Ιουδαίων. The participle
follows upon the preceding one without connecting particle. Cf.
20.19, ἐν ταις ἐπιβουλαῖς τών ’Ιουδαίων; and other passages—
9.24; 21.27; 23.12. There is no means of knowing what the Jewish
plot was. Bornkamm (4.136) thinks that the Jews, probably on
pilgrimage to Jerusalem for Passover, were travelling on the ship that
Paul was intending to use and that it was for this reason that Paul,
changing his plans, decided to travel overland through Macedonia,
and eventually picked up a ship in Philippi or Troas (vv. 5f.). This is
an ingenious suggestion, possibly correct. Certainly Paul was about
to sail to Syria (see 21.3) when ἐγένετο γνώμης ... τού ὑποστρέφειν
 50. BACK TO PALESTINE. 20.1-16. 947

διἀ Μακεδονίας. For the genitive γνώμης see BDR § 162.7, n. 9:

‘att. τής αυτής γνώμης ἦσαν udgl.’ LS 354 refer to Thucydides
1.113.2 (ὃσοι τής αυτής γνώμης ἦσαν); 1.140.1 (τής μἐν γνώμης ...
αεί τής αυτής ἔχομαι. We may add Plutarch, Phocion 23.4 (752),
τήν μὲν πόλιν ἐλπίδος μεγάλης γενομένην; Josephus, War 6.287.
Paul’s change of plan is thus represented as purely rational: con-
sideration of the Jewish plot leads him to travel by a different route
from that which he had intended to take. The Western text (D (gig)
syhmg) emphasizes the supernatural guidance afforded by the Holy
Spirit. Instead of μέλλοντι τοῦ it reads ἠθέλησαν ἀναχθήναι εις
Συρίαν εἶπεν δὲ τὸ πνεύμα αῦτῷ ...
It was at this time that Romans was written (cf. Rom. 15.23-28).

4. This verse is remarkable for a considerable measure of textual

confusion; there is an excellent discussion (covering vv. 3-5) by
Ropes (Begs. 3.190f.).
In the Old Uncial text the verse begins συνείπετο δὲ αὑτῷ, which
Ropes rightly translates there were associated with him, and under-
stands as a general association of persons who were assembling from
various places to share in the journey to Syria, which, according to v.
3, was planned, but changed as far as the immediate future was
concerned into a return to Asia by way of Macedonia. The named
companions went ahead to Troas; Paul and at least one companion
joined them there (v. 6). D however omits συνειπ. δέ αὐ. and has
instead μέλλοντος οὖν ἐξιέναι αὐτοῦ; syhmg has this clause but adds
συνείποντο αὐτῷ. At this point there is a further addition, by A (D)
E Ψ <a> gig vgmss sy, of ἄχρι (μέχρι) τῆς ’Ασίας. This implies, with
some confusion, that the named group accompanied Paul throughout
the journey, from Corinth to Syria. It must be added that if the
Western editor did not grasp clearly what Luke meant in these verses
he was not without excuse.
It is certain that the journey undertaken here had as at least one of
its purposes, probably its primary purpose, the conveying to Jerusa-
lem of the proceeds of Paul’s collection. That Luke does not mention
it is a surprising fact; see Dibelius (211) and many subsequent
writers. Luke’s only hint at this event, of evident importance to Paul,
is in 24.17; reasons for his silence can only be conjectural; see pp.
1001, 1107f. According to 1 Cor. 16.3f., it was Paul’s intention that
local representatives should convey the money to Jerusalem, perhaps
in his company. It is reasonable to infer that those mentioned in this
verse (except perhaps Timothy) were such local representatives and
not assistant missionaries (so e.g. Roloff 296).
There are textual problems also in the list of names.
Σώπατρος Πύρρου, Sopater the son of Pyrrhus. For both names
see Hemer (236). Πύρρου is anarthrous, as in classical use: BDR §
162.2, n. 4; Μ. 3.168. This word is omitted by <a> sy; Ropes (Begs.
948 COMMENTARY ON ACTS

3.191) thinks that ΠYPPOC arose through confused repetition of the

preceding six letters (ΠATPOC); this does not seem probable, but it
is hard to see why Pyrrhus should be omitted if the name stood
originally in the text. Neither Sopater nor Pyrrhus appears elsewhere
in the NT; Sosipater (whose name appears here in 104 (1175) pc gig
pc vgs co) is mentioned at Rom. 16.21. Hemer identifies the two.
Sopater came from Beroea (Βεροιαῖος); Hemer (124) shows that
this is the form used in local inscriptions.
There follow two who came from Thessalonica (see 17.1), Aris-
tarchus and Secundus. The former is presumably the Aristarchus
mentioned at 19.29 as a Macedonian (see the note on this verse). He
reappears at 27.2; see also Col. 4.10; Philemon 24. Secundus (the
Greek form of the name is derived from Latin) is not mentioned
elsewhere. For the accentuation (Σέκουνδος or Σεκοῦνδος) see
BDR §41.3.
The next pair is Gaius Derbaeus (that is, Gaius of Derbe; see 14.6
and the note) and Timothy. See 16.1; this verse suggests but does not
positively affirm that Timothy’s home was in Lystra whereas the
association with Gaius might suggest that Timothy came from Derbe.
Both were Lycaonian towns, though some distance apart. The
connection of Gaius with Derbe may however be in error. (1) In
19.29 along with Aristarchus there is mentioned a Gaius who was a
Macedonian (see the note on the text of 19.29). (2) In the present
verse a variant reading gives instead of Δερβαῖος, Δουβ[έ]ριος (so
D* d (doverius) gig). This would refer to a town Doberos (Δοβήρος
in Thucydides 2.98-100) in Paeonia (see Der kleine Pauly 2.110), at
this time a northern division of Macedonia, and give us a Mace-
donian Gaius in agreement with 19.29. This variant, however, though
accepted by not a few (Williams 229; but see especially Clark xlix, 1,
374-6) is no more convincing than the singular (Μακεδόνα) in
19.29; in each case the more difficult reading must be accepted. As
Metzger (475f.) points out in a detailed note, it is easier, in a second-
century context, to understand the change from Δερβαῖος to
Δούβριος than vice versa, and the pairing of the names suggests that
Gaius and Timothy came from the same region.
There follow the Asian pair, Tychicus (see Hemer 236) and
Trophimus (see Hemer 236f.; also ND 3.91-3—the name was
common, non-Jewish, and suggests servile status, in the past or in the
present). For ’Ασιανοί, D (syhmg sa) have Έφέσιοι, by assimilation to
21.29, Trophimus the Ephesian (cf. 2 Tim. 4.20). The reading is to be
rejected, though the statement may be correct; Ephesians were
Asians. For Tychicus, D has Eutychus—an odd slip, since there is no
reason to think the Eutychus of v. 9 an Ephesian. Tychicus is
mentioned at Eph. 6.21; Col. 4.7; 2 Tim. 4.12; Titus 3.12. The
references are sufficient to show that both men were companions and
colleagues of Paul’s.
 50. BACK TO PALESTINE. 20.1-16. 949

On this verse Bultmann (Exegetica 418) rightly notes that lists

such as this are more likely to have been preserved in writing than in
oral tradition.
5. οὗτοι, the persons named in v. 4 (though Begs. 4.253 thinks that
probably only Tychicus and Trophimus are intended) went on ahead
and waited for us in Troas (see 16.8). It is implied that whoever was
responsible, in whatever way, for the We source joined Paul at
Philippi (v. 6). ‘We’, which persists to v. 15, has not been used since
ch. 16, and it is possible that we should think of the writer as having
remained in Philippi through the whole of the intervening period. It
seems that we should think of the group of travellers as collecting in
Greece (v. 2) but then splitting up for the first stage of the journey
east, five (or two?) sailing direct across the Aegean to Troas and
waiting there, while Paul went, perhaps by land, to Philippi and
thence sailed to Troas. For opinions on the question who the
companion was who drew up the record, see Weiser (559).
This account of events assumes the reading προελθόντες (P74 B2 D
36 104 323 614 1891 pm latt). προσελθόντες is read by Avid Β* E
H L P Ψ 945 1175 1241 1739 2495 (pm). This makes a less clear
story but has been held to harmonize with 1 Tim. 1.3 (cf. 1 Tim. 3.14;
4.13); it may have been introduced for this reason. According to
Conzelmann (115), προο- is better attested, but the sense requires
προ-. It is in fact doubtful whether προο- is better attested; the
agreement of B and D is striking.
For ἠμᾶς, D, alone, has αυτόν. This may have been intended to
take up μέλλοντος ... αὐτού in v. 4, but has little to commend it.
6. We sailed with Paul from Philippi; it may be presumed—though
it is not certain—that the writer joined Paul in Philippi. On the
question of a ‘We’ source see the Introduction, pp. xxv-xxx.
ἐξεπλεύσαμεν: in the NT only at Acts 15.39; 18.18; 20.6. But only
Luke has occasion to use the word (apart from a few short trips on
Galilee).
τῶν ἀζύμων. Luke still uses the Jewish calendar as a means of
dating events; there is no Christian ‘Easter’ that he can refer to. See
Begs. 4.254f. for a note on the origin of the Christian festival. Roloff
(296) takes it that Paul naturally observed Jewish festivals; Lüde-
mann (229) says that dating by Jewish festivals is Lucan; his addition
to a source? For Philippi see on 16.12.
We joined the rest (see vv. 4f.) at Troas (see 16.8) ἄχρι ήμερων
πέντε. This must be intended to mean that the journey from Philippi
to Troas lasted five days, but the use of ἄχρι seems to be without
parallel. We may think of the time up to which the voyage lasted
(Page 112). Begs. 4.255 cites Plutarch, Cicero 6 (863), ἄχρι παντός,
continually, and Hermas, Mandate 4.1.5, ἄχρι τής ἀγνοιας, so far as
he is ignorant; but these are not close parallels. Luke uses the word in
950 COMMENTARY ON ACTS

its normal sense elsewhere, e.g. at v. 11 (ἄχρι αυγής). The strange-

ness of the construction is marked by D’s elegant πεμπταῖοι, but the
reading of P74 A E 33, ἀπό ἡμ. π., is hardly an improvement. This
journey took five days, the journey in the opposite direction (16.11)
only two. Commentators agree that an easterly wind could be
responsible for so great a difference; so e.g. Beyer (122).
ημέρας επτά. It is often said that the next verse implies that Paul
arrived and left on a Monday; it would be better to say that if he left
on a Monday he must have arrived on a Tuesday; Tuesday to
Monday, counting inclusively, is seven days.
For ὅπου, Β (Ψ) <a> have ου, D has ἐν ᾗ καί. There is no
difference in meaning, διατρίβειν occurs twice (once) in Jn, but in
Acts see 12.19; 14.3, 28; 15.35; 16.12; 20.6; 25.6, 14.

7. A new narrative begins which is entirely separable from the

context of the journey from Greece to Jerusalem. It is a straightfor-
ward miracle story, the intention of which is to show the supernatural
power attendant upon the Gospel and manifested in the work of Paul
(cf. e.g. Peter in 9.36-42). Attempts to find an allegorical intention in
the story (even Weiser’s, 564) are unconvincing.
’Εν δέ τῇ μία τῶν σαββάτων. For σάββατα see 17.2 and the note.
The cardinal μία is used where the ordinal would be expected; so also
at Mt. 28.1; Mk 16.2; Lk. 24.1; 1 Cor. 16.2. Tt is Jewish Greek’
(Black, AA 124), though not without some analogy in Greek idiom
(Μ. 1.96; 2.174, 439). Begs. 4.255 is probably right in taking the
reference to be to what we should call Sunday evening; it is however
not impossible that Luke means Saturday evening, when by Jewish
reckoning the first day of the week began. See H. Riesenfeld in FS
T. W. Manson, 210-17; also R. Staats in ZNW 66 (1975), 242-63,
especially 247f. Staats chooses Saturday night.
συνηγμένων ἡμῶν. ‘We’ joined Paul at v. 6; the first person plural
continues. For a general discussion of this see Introduction. Bult-
mann (Exegetica 420) thinks that the narrative was drawn from a
written source (see on v. 8), and that the first person plural was
introduced into it when it was incorporated into the itinerary. See
above, p. 949.
κλάσαι άρτον. See 2.42, 46; 27.35, and the notes. There is no
indication in the present passage, unless it is implied by the phrase
itself, that the meal was other or more than a church fellowship meal,
accompanied by religious discourse (ό Παύλος διελέγετο) and
conversation (v. 11). That the event took place in a private house (v.
8) does not in itself prove that it was not an action of the whole
church; at this period there were few, perhaps no, special church
buildings. Reicke (DFZ 75f.), makes a distinction (perhaps without
foundation) between eucharistic agape (Gemeindegottesdienst) and
private agape, but notes that the eucharistic agape will often have
 50. BACK TO PALESTINE. 20.1-16. 951

been held in private houses, and that later the private agape was
sometimes held in church. Whether the ἐκκλησία ever in fact had
anywhere in which to meet except private houses is very doubtful.
For a general consideration of eucharist, agape, and the like in Acts,
with the recognition that Acts provides little or no basis for such
terminology or for any developed set of practices, see Introduction,
pp. xciif.
διελέγετο: see 17.2, 17; 18.4, 19; 19.8, 9; 20.7, 9; 24.12, 25;
always of Paul; elsewhere in the NT only Mk 9.34; Heb. 12.5; Jude
9. The meaning varies between dialogue, discussion between two or
more persons, and discussion in which one person discusses a matter,
as in a sermon or lecture. Paul had already spent seven days in Troas,
and was to depart on the next day; no doubt for this reason, since he
could hardly hope to have another opportunity of speaking to these
Christians, he extended his sermon till midnight, μέχρι μεσονυκτίου.
The omission of the article with a word denoting time is classical (Μ.
3.179), but cf. 16.25, where the article is used. There is no indication
at what hour the gathering had begun; presumably not till after the
day’s work had ended. The λαμπάδες in v. 8 show that it was dark.
Nor is it said at what point in the proceedings the discourse began.
Probably the meeting began with supper, but even if this went on for
some time Paul must have preached at very much greater length than
any of the sermons recorded in Acts; these are of course no more
than the barest outlines. The time was early May (if we build on v.
16). Workmen (apart from indoor household servants) should have
been free from their work by 7 or 8 p.m.
It was the first day of the week—the Christian Sunday (even if it
was Saturday evening), but of course a working day. Observance of
the Jewish Sabbath is already abandoned in the NT (Col. 2.16), and
early in the second century the two observances were taken to be
typical of the Jewish and Christian religions respectively (Ignatius,
Magnesians 9.1, μηκέτι σαββατίζοντες ἀλλὰ κατὰ κυριακήν
ζῶντες). 1 Cor. 16.2 refers not to a Christian gathering but to
Christian action on the first day of the week; Rev. 1.10 refers to the
Lord’s Day as an occasion when the Seer was in the Spirit. Cf. Pliny,
Epistles 10.96.7: ... quod essent soliti stato die ante lucem con-
venire. This was a morning gathering. The passage continues: quibus
peractis morem sibi discendendi fuisse, rursusque ad capiendum
cibum. The latter gathering, for food, corresponds more to what is
described as taking place at Troas. It is often supposed, but without
good reason, that the morning gathering mentioned by Pliny was a
eucharist, the later (evening?) meeting a non-sacramental meal (an
agape); so e.g. Lightfoot (Ignatius 1.52). For meetings on the Lord’s
Day see also Didache 14.1; Barnabas 15.9.
The theological significance of the replacement of the Sabbath by
the ‘Lord’s Day’ is discussed by Barth (CD 3.1.228; 2.458f. 4.53)
952 COMMENTARY ON ACTS

and by Calvin (Institutes 2.8.33f.)—not convincingly. Luke’s refer-

ence to the first day of the week is made in passing, as a natural
explanation of the fact that the Christians were taking supper
together. It does not appear that he is pressing the observance of the
day as something that he wishes to commend to his readers; rather he
assumes that they will fully understand what is going on. Commenta-
tors are apt to add the assumption that Luke understood the contents
of Christian worship in the terms in which it was later practised.
Bengel (467): Itaque credibile est, fractione panis hic denotari
convivium discipulorum cum eucharistia conjunctum; Pesch (2.193):
The ‘urchristliche Gottesdienst... findet am Sonntag statt, in einem
Privathaus, mit Wortgottesdienst und eucharistischem Mahl’. These
are relatively cautious statements.
The reference here to the first day of the week has been made the
basis of a conjectural attempt to find a precise date in the chronology
of Paul’s life; see especially Hemer in FS Bruce (1980), 9-12; also
Hemer (169, 216) and R. Jewett (Dating Paul's Life (1979), 47-50).
The argument in outline is that Paul left Troas on Monday (vv. 7,
11); he therefore began his seven-day stay in Troas on Tuesday. He
arrived after a five-day journey (v. 6) from Philippi, which he must
have left on Friday. But he left Philippi after the days of unleavened
bread (v. 6), therefore Passover fell in that year on a Thursday. But
‘the Passover of 57, as calculated from the full moon, fell probably
on Thursday 7 April, and the Passover of neighbouring years could
not have fallen on that day of the week’ (Hemer 169). It follows that
Paul’s journey took place in AD 57. Unfortunately this is by no
means convincing. (1) Lk. 22.7 is sufficient to show Luke’s quite
imprecise knowledge, or careless statement, of the relation between
Passover and Unleavened Bread; (2) Passover in fact coincided with
the first of the days of unleavened bread, so that if Unleavened Bread
ended on a Thursday Passover must have been on a Friday; (3) μετά
τὰς ἡμέρας τῶν ἀζυμων does not necessarily mean ‘on the next day
after the Feast of Unleavened Bread’; (4) the times of full moon in
the first century can be calculated now with greater accuracy than
they could be observed at the time; (5) even if the ‘five’ and ‘seven’
of ν. 6 were drawn from the ‘We’ source, or Itinerary, it is impossible
to place complete confidence in their precise accuracy. And such
questions as, ‘Did the last day of the five coincide with the first day
of the seven, or not?’ are unanswerable.

8. Instead of λαμπάδες, D (not d, which has faculae) has

ὐπολαμπάδες, a very rare word (see LS 1887; MM 658; BA 1684;
also Μ. 2.328). The best that LS can offer is ‘part of a στοά,
possessing ... a roof, and tiles’. The evidence in MM suffices for the
meaning window. Moulton also gives window, but adds ‘apparently a
screen under which the light shines’. The most helpful parallel to the
 50. BACK TO PALESTINE. 20.1-16. 953

present use is Joseph and Aseneth 14.9, ὐπολαμπαδός καιομένης,

which would make it a synonym of λαμπάς. It may be that the
copyist responsible for the word thought of the meaning window, the
thought being put into his mind by v. 9. G. Zuntz (Gnomon 30
(1958), 26) prefers the reading of D (cf. Metzger 477), but it seems
unlikely that the original reading would survive in one MS only. The
lamps (which Bultmann, Exegetica 420, sees as a pointer to the use
of a written source) are probably mentioned to account for the
drowsiness of Eutychus (though there are some who think that E. fell
asleep in spite of the lamps, which were intended to promote
wakefulness). It is most unlikely that Luke had any further motive in
mentioning them, though they are said by opponents whom Justin
answers to be used in lovefeasts (1 Apology 26: εἰ δέ καί τἀ
δύσφημα ἐκεῖνα μυθολογούμενα ἔργα πράττουσι, λυχνίας μὲν
ανατροπήν καί τὰς ἀνέδην μίξεις καί ἀνθρωπείων σαρκών βόρας,
οὐ γινώσκομεν). Cf. Minucius Felix 9.9; Tertullian, Apology 7. In a
meeting that lasted all night (v. 11) lamps would be needed.
ἐν τω ὐπερῴω. See on 1.13; the word is used also at 9.37, 39. At 1
Clement 12.3 it is used of the flat roof of a house; in earlier Greek
(see LS 1871) it described the women’s part of a house; in Acts
(nowhere else in the NT) it is simply an upper storey (cf. v. 9).
For ἦμεν, 1 pc bo have ἦσαν, which must be a slip; copyists had
forgotten the first person plural of v. 7—though Conzelmann (115)
regards the ‘we’ in this verse as ‘offensichtlich sekundär’. It may
well be that this incident has been incorporated into an Itinerary, but
this can hardly be said to be obvious.
Rackham (379) mentions Jewish Sabbath lamps, the use of which
was known to Gentiles (Persius, Satire 5.180-184). There is no
reason to suppose that the lamps in this story were other than
functional.

9. νεανίας ὀνόματι Εὔτυχος. For the name see Hemer (237); it

means Fortunate, and a young man brought back to life might well be
thought fortunate. But the name was a not uncommon one (note the
story of Augustus and the peasant Eutychus with his ass Nicon,
Suetonius, Augustus 96.2). The story in Acts may or may not be
historical, but there is no need to suspect its historicity simply
because of the occurrence of this name.
θυρίς (cf. 2 Cor 11.33) is, unlike ὐπολαμπάς (v. 8) a common
word for window.
καταφερόμενος ... κατενεχθείς. The change of tense brings out
vividly the gradually increasing drowsiness and its climax at the
point where Eutychus finally falls asleep. The only biblical parallel to
this use of καταφέρεσθαι appears to be Ps. 75(76).7 (Aquila; LXX
ἐνυσταξαν, Hebrew but cf. Aristotle, De Generatione Animal-
ium 5.1 [779a] (Hatch, Essays 25). Though the verb is repeated it is
954 COMMENTARY ON ACTS

followed in the first use by the dative (ὕπνω βαθεῖ), in the second by
ἀπό and the genitive (ἀπὸ τοῦ ὔπνου). ἀπό is used for ὐπό (see Μ.
1.102,237,246; BDR § 210.2, n. 2) at 10.33 (v.1.); 15.4; use with the
instrument is less usual than with the agent, but there seems to be no
other interest here than a desire for variation. There is yet another
construction in Lucian, Dialogi Meretricii 2, ἐς ὕπνον κατηνέχθην.
Page (212) points out that ὕπνω βαθεῖ is not deep sleep but strong
drowsiness, which ends in sound sleep. P. W. van der Horst (NovT 17
(1975), 158) quotes Hierocles the Stoic, p. 25, col. 5,1 ... βαθυτάτῳ
πεπιεσμένοι... ὕπνφ; p. 27, col. 5,27 βαθέων ὕπνων.
For τής θυρίδος, D has τῇ θυρίδι; for καταφερόμενος,
κατεχόμενος; for βαθεῖ, βαρεῖ: ἐπί τῇ θυρίδι κατεχόμενος ὕπνω
βαρεῖ. It would be difficult to find different translations for the two
readings; one—almost certainly that of D—is due to loose copying.
Some editors connect ἐπἰ πλεῖον not with διαλεγομένου but with
κατενεχθείς—wrongly, since the adverbial phrase is more suitable
to the present than to the aorist participle.
In the Qumran community falling asleep in the assembly was
punishable by exclusion for 30 days (1QS 7.10). There is nothing to
suggest that Luke’s interest in Eutychus was of a disciplinary kind;
Eutychus’ accident provides the occasion for a miracle by Paul and
serves no other purpose. See Braun (ThR 29 (1963), 173).
από του τριστέγου, from the third storey, that is, in English usage,
from the second floor. The adjective τρίστεγος appears to denote a
building on three levels (e.g. Josephus, War 5.220); τό τρίστεγον (sc.
οίκημα) meant the highest of three. The better houses of antiquity
seem for the most part to have been one-storey buildings. The
dwellings of the poor were apartment houses of several storeys.
καί ήρθη (καί ὃς ήρθη (sic), D*) νεκρός. In view of ν. 10 one
might expect ώς νεκρός. Luke however was capable of expressing
the thought that observers mistakenly supposed someone to be dead
(14.19), and probably means that Eutychus was truly dead. V. 10
must be interpreted accordingly. Luke thinks of a miraculous
resurrection (Schneider 2.283). ‘Im Vergleich mit anderen Toten-
erweckungserzählungen ist leicht zu erkennen, wie wenig schema-
tisch find wie sehr konkret hier erzählt ist’ (Pesch 2.189). From this
concreteness Pesch (2.193) concludes that we have here a tradition of
high historical probability. But concreteness, which may be fictional,
is not a criterion of probability.

10. Luke uses ‘almost technical language in describing miracles’

(Knox, Hell. El. 8). There may be special allusions to the Elijah and
Elisha stories of the OT: 3 Kdms 17.17-24 ( ... ὐπερφον ...
ἐκάθητο ... ἐπιστραφήτω δὴ ἡ ψυχή του παιδαρίου τούτου εις
αυτόν ... ἔδωκεν αὐτόν τή μητρὶ αυτοῦ ... ζῇ ὁ υιός σου); 4 Kdms
4.18-37 (ἐκοιμήθη ... καί ἀπέθανεν ... τεθνηκός ... διέκαμψεν
 50. BACK TO PALESTINE. 20.1-16. 955

ἐπ’ αυτόν ... Λάβε τόν υιόν σου). Thus ή ψυχή αὐτοῦ ἐν αὐτῷ
έστιν probably means, His life is now, in virtue of my action, within
him; this is consistent with the statement of v. 9, strictly understood,
that the boy was (not, appeared to be) dead.
μή θoρυβεῖσθε, present imperative, that is, Stop this disturbance.
The meaning of συμπεριλαβών is unmistakable, but LS 1682 give
no other example of the literal meaning, to embrace. BA 1555 quotes
Xenophon, Anabasis 7.4.10, περιλαβών τόν παῖδα. [Editions of
Xenophon, giving no MS variation or conjectural note have περι-
βαλών, but OCT has –λαβ-.] D has συμπεριλαβών καὶ εἶπεν. This
un-Greek construction may be an Aramaism (see Black, AA 68f., and
cf. 4.3; 5.21; 7.4; 8.2; 10.27; 12.16; 13.7; 14.6, 14), but is more
probably due either to the influence of d (circumplexit et dixit), or to
a mere slip.
Clark (liii, 130f., 377), who thinks the text to have been written
originally in sense lines sets out vv. 10-13 as follows:

καταβὰς δέ ό Παύλος ἔπεσεν ἐπ’ αὐτῷ

καί συνπεριλαβών καί εἶπεν μή θορυβεῖσθαι
ή γάρ ψυχή αὐτοῦ ἐν αὐτῷ ἔστιν
ἀναβὰς δέ καί κλάσας τόν άρτον
5 καί γευσάμενος ἐφ’ Ικανόν δέ ὁμειλήσας
ἄχρις αὐγῆς ούτως ἐξῆλθεν
ἀσπαζομένων δέ αὐτῶν
ἤγαγεν τόν νεανίσκον ζῶντα
καί παρεκλήθησαν οὐ μετρίως
10 ήμεῖς δέ κατελθόντες εἰς τό πλοῖον

and thinks that lines 8 and 9 (with ήγαγεν—singular) should be read

after line 3.

11. άναβάς—from the street to the second floor.

κλάσας τόν άρτον, the purpose of the gathering, v. 7; see the note
there.
Γευσάμενος. γεύεσθαι normally means to taste, but in Luke’s use
it is to eat, to take a meal (see Lk. 14.24; Acts 10.10; 23.14; the only
exception, Lk. 9.27, is taken directly from Mk and is in any case
metaphorical). It would therefore be mistaken to infer that the
reference was to a sacramental meal, in which a mere fragment of
bread was tasted. Luke means that Paul joined his fellow Christians
in eating a meal; their eating is not specifically mentioned because
for Luke Paul is the centre of interest. If Luke had meant that
Eutychus now ate a meal, thereby proving his complete restoration,
he would have been obliged to express himself differently.
Among NT writers only Luke uses the verb ὁμιλεῖν. At Lk. 24.14,
15; Acts 24.26 it means not to address a meeting but to engage in
956 COMMENTARY ON ACTS

conversation. This is a normal meaning of the word and probably

applies here also. No doubt on such occasions Paul did the lion’s
share of the talking, but it was not an entirely one-sided engagement.
οὕτως sums up the preceding participles—ἀναβάς, κλάσας,
γευσάμενος, ὁμιλήσας: a classical use (BDR § 425.6) which occurs
in the NT only here and at 27.17; but cf. 20.35. ἐξῆλθεν could mean
he went out of the house or he left Troas, probably the former; but
Luke’s main intention is to indicate that the incident is now ended as
far as Paul is concerned.
αυγή, daybreak; cf. Isa. 59.9. Begs. 4.257 gives ‘the daylight
before the sunrise’ in the note but not in the translation.

12. Dibelius 87, noting that D begins the verse ἀσπαζομένων δέ

αυτών ἤγαγεν τον νεανίσκον ζῶντα, says that D smoothes out the
transition. This is correct; but see Clark, on v. 10.
All that remains is to make explicit the happy ending of the story
for the Christians in Troas. Presumably they had brought the boy
upstairs before Paul restarted the meal and the discourse. He had
been dead (v. 9) and was now alive (ζώντα)—an astounding miracle,
which Luke passes over almost as if it were an everyday occurrence.
He does not add that the people were astounded, only that
παρεκλήθησαν. In Acts this word usually means to exhort or to ask:
here it must mean they were comforted, οὐ μετρίως is a characteristic
Lucan litotes; cf. 12.18; 19.11; 21.39; 28.2; also e.g. Plutarch, Titus 9
(373), τὸν δέ Τίτον ... ού μετρίως παρώξυνε τά τοιαῦτα.
The subject of ἣγαγον is not really indefinite (unbestimmt—
Haenchen 561); it must be the Christians of Troas who had
assembled for their meeting; it could be that they took him to his
home, but the interpretation given above seems more probable. D
(see above) makes the subject clear; the subject of the singular
ἤγαγεν can only be Paul: while they were saying their farewells Paul
brought... For ἀσπάζεσθαι, to say goodbye, see v. 1.13

13. ἡμεῖς: the ‘We’ passage continues, but on this occasion Paul is
not included; see below. We means ‘alle Begleiter des Paulus’
(Haenchen, 562). The Commentary of Ephraem and the Armenian
Catena, which often give old Syriac readings (Begs. 3.442f.), have
Ego Lucas et qui mecum. This is at any rate correct interpretation (if
Luke is the author of the We passages). See further Metzger (477).
προελθόντες is thereadingof P41vid P74 B2CLΨ 33 36 323 614
945 1739 2495 al; A B* E <a> have προσελθόντες; D gig syp have
κατελθόντες. The different sense given to the text by these readings,
and the reasons for preferring προελθ., will be discussed below.
ἀνήχθημεν, we set sail; for this word see 13.13.
ἐπί τήν ’Άσσον, a town on the mainland, founded from Lesbos
(Mitylene; see on v. 14) in the eighth century BC. From 133 BC it
 50. BACK TO PALESTINE. 20.1-16. 957

was under Roman rule, presumably in the province of Asia. Aristotle

lived in Assos from 348 to 345 BC.
On this voyage Paul was not with ‘us’; he travelled by land
(πεζεύειν—not necessarily on foot; cf. Strabo 6.35, τινὰ τών
φορτίων πεζεύεται ταῖς ἀρμαμάξαις) from Alexandria Troas to
Assos. Why Paul chose to travel alone in this way we do not know.
The suggestion that he was liable to sea sickness is exegetical
despair. There is nothing to suggest that he made an evangelistic tour
through the district.
’We’ sailed to Assos with the intention (μέλλοντες) of taking up
Paul there, ούτως γὰρ διατεταγμένος ἦν, for so he had arranged;
but διατάσσεσθαι has often a stronger sense in Acts (see 7.44; 18.2;
23.31; 24.23)—perhaps, he had given orders to this effect, μέλλων
αυτός πεζεύειν, since he himself was intending to go (was about to
go) by land. D has ὡς μέλλων. See BDR § 425.3: ὡς gives the clause
a subjective sense, indicating that μέλλων expresses not necessarily
the fact, but the mind of the speaker. It is doubtful whether D can be
followed here. It should be noted that D reverses the order of the
preceding words, reading ἦν διατεταγμένος; ὡς may have originated
in the accidental repetition of the last two letters of the participle.
μέλλων αυτός πεζεύειν, with or without ὡς, fits admirably with
the reading προελθόντες at the beginning of the verse. The main
party sailed on ahead, and picked Paul up at Assos, where he was
obliged to take to the sea. κατελθ. has been defended on the ground
that it is lectio dijficilior: in Greek one does not go down to a ship but
up to it, so that κατελθ. would invite correction, but it would hardly
be changed to προελθ. or προσελθ., and movement up is covered by
ἀνήχθημεν—one goes down through the town to the harbour, then
up to the boat, and on to the high seas. προσελθ. is possible and
could have been assimilated to v. 5 (but B* there has προσελθ.!), but
προελθ. fits the context so well that it is probably what Luke
intended to write, even if he wrote προσελθ. by mistake.

14. Where he met us (συνέβαλεν, C D <a> is an improvement on

the imperfect συνέβαλλεν, which is therefore probably original,
though it is hard to know why Luke wrote it; συνέβαλλον, must
be a slip). At Assos (see v. 13; εἰς with the accusative for ἐν with the
dative; cf. 2.5), we took him up, that is, into the ship, though
ἀναλαμβάνειν can be used generally (as at 2 Tim. 4.11).
Mitylene. So the MSS of Acts (Μιτυλήνη), though the earlier
(before 300 BC) spelling was Μυτιλήνη. Μ. was the largest town—
and a very large one, almost as great in superficial area as Athens—
on the island of Lesbos, for centuries an important centre of Greek
life, commerce (Μ. was a notable port), and art (the home of Alcaeus
and Sappho, among others). The cult of Augustus (or Augustus and
Roma) was established in Μ. as early as 27 BC (CAH 10.486;
958 COMMENTARY ON ACTS

Dittenberger, OGIS 2.456), and the island enjoyed, on the whole,

imperial favour.

15. ἀποπλεύσαντες. The verb occurs in the NT only in Acts (13.4;

14.26; 20.15; 27.1). It is hardly to be distinguished from ἐκπλεῖν
(15.39; 18.18; 20.6).
τῇ ἐπιούση. The distance from Mitylene to Samos was too great
for one day.
ἄντικρυς occurs here only in the NT. For the distinction between
ἄντικρυς and αντίκρυ see Rutherford (Phrynichus 500f.). Luke
probably means that they sailed between the island of Chios and the
mainland (‘right through’). The town Chios was situated on the east
coast of the island, and it may be the town rather than the island that
is in mind. The island was prosperous (though the words of
Hermocrates in Thucydides 8.45.4, οἱ μὲν Χῖοι ... πλουσιώτατοι
ὄντες τῶν Ελλήνων, were not an unbiased economic estimate), and
the town had been made a civitas libera by Sulla (86 BC).
τῇ δὲ ἐτέρᾳ (ἐσπέρα in Β 36 453 1175 1241 pc bopt probably goes
back to an accidental slip) ... τή δέ ἐχομένη (ερχoμενη in D* 614
1175 1891 al): Luke chooses variants for τή ἐπιούση (above).
Samos, another of the Ionian islands, separated from the mainland
by a channel only just over a mile broad. The main harbour lay on the
south east coast. Augustus was there in the winter of 31/30 BC
(Suetonius, Augustus 17) and again in 21/20 BC, when he declared
Samos a civitas libera (Dio Cassius, 54.9.7), a privilege withdrawn
by Vespasian (Suetonius, Vespasian 8).
The meaning of παρεβάλομεν (the first person plural returns) is
uncertain. It may mean to pass by, here more probably to reach; cf.
Thucydides 3.32.2, ... ἐς 'Ιωνίαν παραβαλεῖν; Josephus, Ant.
18.161, Άγρίππας δέ εις Ποτιόλους παραβαλών. It is probable that
Luke used the word in this sense and that the Western editor
(followed by the Byzantine text) misunderstood it in its alternative
sense, and by adding the words καί μείναντες ἐν Τρωγολ(λ) ίῳ (so D
(Ψ) <a> gig sy sa, also omitting the following δέ) provided an
alternative port of call. An answer is thus provided to the question
(Ropes, Begs. 3.195; Hanson 202), Why, if these words were not
original, were they added? Trogyllium lay on the mainland, approx-
imately opposite the port of Samos. Μ. 2.362 refers to WH,
Introduction: Notes 98; for the spelling see WS 47; Ramsay (Church
155).
It was between Chios and Samos that the ship sailed past Ephesus
(see ν. 16).
Miletus was an ancient, prosperous, and influential city; loniae
caput... super XC urbium per cuncta maria genetrix (Pliny, Natural
History 5.112). Here Paul halted; see below. It may not be a matter of
coincidence that there is evidence of a Jewish element in the
 50. BACK TO PALESTINE. 20.1-16. 959

population (see Trebilco, 12, 56. It is here that the well-known

theatre inscription occurs: τόπος Εἰουδέων τών καί θεοσεβίον (sic)
(CU 2.748; Deissmann, LAE 446f.; discussed by Trebilco, 159-62).
See NS 3.24,25,167,168; G. Kleiner, Das römische Milet (Sitzungs-
berichte der Wissenschaftlichen Gesellschaft, Frankfurt/Main, 8.5;
1970).
A journey the reverse of Paul’s was made by Herod: Josephus,
Ant. 16.16—20.

16. κεκρίκει: for the dropping of the augment in the pluperfect see
Μ. 2.190 (though strangely this example is not given along with
4.22; 14.23). Paul had chosen; for choose as the meaning of κρίνειν
see LS s.v. II 1 (996). His choice (of route) was to sail past Ephesus
(see on v. 15; and on Ephesus see on 19.1) in order to save time; he
did not wish χρονοτριβήσαι ἐν τῇ ’Ασίᾳ, as no doubt he would have
been obliged to do if he had made direct local contact with the church
in Ephesus. For χρονοτριβήσαι, to waste time, to loiter, though
given by LS 2009, are probably a little strong for this passage; Paul
would not have wasted his time in Ephesus—there is no doubt that
he could and would have been usefully employed, but he would
probably have found it difficult not to spend longer with the church
there than suited his plans. This assumes the details of the story as
given by Luke in the second part of the verse. It is possible that
Paul’s plans and motives may have been different; see below. D (gig)
vg reword: μηπότε γενηθῇ αὐτῷ κατάσχεσις.
The use of γένηται is unusual, and is probably not to be connected
with the Hebrew construction ... ... which does seem to
underlie some of Luke’s expressions (though seldom if ever in the
later chapters). Μ. 1.17, considering the question of Hebraism,
compares the English idiom, 'It happened I was at home that day’.
This is not a precise parallel, but it is relevant to the use of γίνεσθαι
meaning to happen. Paul did not wish it to happen to him that
(infinitive of result) he should spend more time ... Cf. BDR §
409.4.
The reference to time (χρονοτριβήσαι) makes it probable that
ἔσπευδεν has the intransitive meaning, to make haste. Paul wished to
be in Jerusalem for the Day of Pentecost; this is no doubt what the
Greek means, but the translation veils certain difficulties.
εις before 'Ιεροσόλυμα (so B C D <a> latt; ’Ιερουσαλήμ in P74 A
Ε Ψ 1704 1739 2495 pc) stands for ἐν with the dative (cf. v. 14).
τὴν ἡμέραν (unless we follow D in prefixing εις) is an unusual
accusative of point of time. On the growing use of this construction
in the papyri see Μ. 1.63 and Μ. 3.248 (adding also Demosthenes
54.10 (1260), ἐκείνην τήν εσπέραν).
Paul’s desire is qualified by εἰ δυνατόν εἴη αὐτω (omitted,
accidentally according to Ropes, Begs. 3.195, by D*), which
960 COMMENTARY ON ACTS

introduces a measure of uncertainty; he recognized that he might

come too late. See BDR § 385.2, n. 4; Radermacher (131) puts the
matter neatly: ‘Die Partikel εἰ leitet, parallel unserm “ob” einen Satz
ein, der dem erwarteten Erfolg eine Überlegung angibt.’
Whether Paul reached Jerusalem in time for the feast we do not
know. According to Conzelmann (116) it was possible, given
favourable weather. Stahlin (266) thinks that the crowds of 21.27
show that Paul was in Jerusalem at the time of the feast. In any case
we hear no more of it. Would Paul have been so keen to attend a
Jewish feast (there was as yet no Christian Pentecost)? Gal. 4.10
suggests a disparagement of calendrical observances; cf. Rom. 14.5f.
Would he have saved much time by not calling at Ephesus? He went
on to Miletus (v. 15); from Miletus he sent to the church of Ephesus
and summoned the elders; addressed them, and sent them home.
Haenchen (564) estimates the distance as 50 km. (31 miles) as the
crow flies and that Paul’s messengers will have taken two days, the
elders longer (but elders were not necessarily old) to cover it. If these
estimates are correct the proceedings will have taken not less than
five days, possibly six. Hemer (125) however thinks that two or three
days would be sufficient. It is unlikely that Paul would have been
obliged to spend more than five or six days in Ephesus. If we think
that Paul was not seeking to save time and wish to guess at a different
reason for his movements we may suggest that he was (it seems)
carrying a considerable sum of money, the product of his collection
for the saints in Jerusalem. He might well feel safer in Miletus than in
the great city of Ephesus. Conzelmann (116) suggests that Paul had
made Ephesus too hot to hold him, a fact that Luke suppresses for
apologetic reasons. Bengel (468) observes, ‘In festo, magni con-
ventus: magna benefaciendo occasio’; Calvin (2.172) similarly, but
at greater length. Tajra (61) thinks that Paul wished to be in
Jerusalem for Pentecost (a) to show loyalty to Jewish tradition and
practice, (b) to deliver the collection. It is more probable that we
should think on other lines. The story is not all fiction; writers do not
often voluntarily concoct the sort of problem that we see here. It
might have seemed worth while to an imaginative writer to invent a
great farewell speech in Ephesus; he would not of his own accord
place such a speech in Miletus. If however he found the tradition of a
speech in Miletus (and none in Ephesus), he might ferry the Ephesian
elders across in order to provide a partly Ephesian audience.
 51. PAUL’S SPEECH AT MILETUS 20.17-38

(17) From Miletus Paul sent to Ephesus and summoned the elders of the
church. (18) When they reached him he said to them, ‘You [elders]1 know
how, from the first day I set foot in Asia, I conducted myself towards you for
the whole time I was with you, (19) serving the Lord with all humble-
mindedness and tears and afflictions that befell me through the plots of the
Jews. (20) [You know]2 that I3 kept back none of the things that were4
profitable for you, so as not to declare them to you and teach you in public
and5 in private, (21) testifying to both Jews and Greeks repentance towards
God and faith in our Lord Jesus. (22) And now behold I am going to
Jerusalem,6 bound in the Spirit, not knowing the things that will befall me
there, (23) only that the Holy Spirit in city after city testifies to me that bonds
and afflictions await me. (24) But7 I take my life to be of no account as
valuable to myself8 so that I may finish my course and the ministry that I
received from the Lord Jesus—the ministry that consists in9 testifying the
Gospel of the grace of God. (25) And see now, I know that all of you, among
whom I went about proclaiming the kingdom, will see my face no more. (26)
Therefore I affirm to you this day that I am clear from the blood of all men,
(27) for I kept nothing back so as not to proclaim to you the whole counsel of
God. (28) 10Take thought for yourselves and all the flock in which the Holy
Spirit has11 appointed you bishops, to shepherd12 the church of13 God, which
he14 saved and acquired through15 his own blood. (29) I know that after my
departure fearsome wolves will come into your company, not sparing the
flock, (30) and from among yourselves men will rise up16 speaking perverse
things so as17 to draw away disciples18 as their followers. (31) Watch
1Elders is not in the Greek.
2You know is not in the Greek.
3RSV, did not shrink from declaring to you; NJB, I have not hesitated to do
anything.
4NEB, for your good.
5RSV, from house to house; NEB, NJB, in your homes.
6NEB, under the constraint of; NJB, in captivity to.
7NEB, I set no store by my life, I only want to finish; NJB, I do not place any value
on my own life, provided that.
8RSV, if only I may.
9the ministry that consists in is not in the Greek.
10NEB, keep watch over; NJB, be on your guard.
11RSV, made you guardians; NEB, given you charge; NJB, the overseers.
12RSV, feed.
13RSV, NEB, the Lord.
14RSV, obtained; NEB, won for himself.
15NJB, the blood of his own Son.
16NJB, with a travesty of the truth on their lips.
17NEB, to induce the disciples to break away and follow them; NJB, to induce the
disciples to follow them.
18RSV, after them.
961
962 COMMENTARY ON ACTS

therefore, remembering that night and day for three years I did not cease
admonishing each one with tears. (32) And now I commit you to God and to
the word of his grace,19 to him who is able to build you up and to give you
the inheritance among all those who have been sanctified. (33) I desired no
one’s20 silver or gold or clothing. (34) You yourselves know that these
hands21 ministered to my own needs and to those who were with me. (35) In
all ways I showed you that it is necessary22 to work like this and to help
the weak,23 and to remember the words of the Lord Jesus,2425 [to remember]
that he himself said, '25 It is more blessed to give than to receive.’
(36) When he had said these things he knelt down with them all and
prayed. (37) A good deal of lamentation arose on the part of all, and they fell
on Paul’s neck and kissed him, (38) grieving most of all at the prediction he
had uttered, that they should see his face no more. They saw turn off to the
ship.

Bibliography
L. Aejmelaeus, Die Reception der Paulusbriefe in der Miletrede (Apg
20.18-35), Helsinki 1987.
C. K. Barrett, FS Mitton, 2-5.
C. K. Barrett, FS Dahl, 107-21.
C. K. Barrett, FS Dupont, 686f.
H. W. Beyer, TWNT2.606f.
F. Bovon, Kremer, Actes, 339-58.
H. Braun, ThR 29 (1963), 173.
T. L. Budesheim, HThR 69 (1976), 9-30.
C. F. Devine, CBQ 9 (1947), 381-408.
J. Dupont, Le Discours de Milet, Paris 1962.
J. Dupont, Nouvelles Études, 424-45.
C. Exum and C. Talbert, CBQ 29 (1967), 233-6.
K. N. Giles, NTS 31 (1985), 135-42.
Μ. D. Hooker, NTS 35 (1989), 331-42.
P. W. van der Horst, NovT 16 (1974), 309.
Μ. Karrer, NovT 32 (1990), 152-88.
J. J. Kilgallen, JBL 112 (1993), 312-14.
W. S. Kurz, JBL 109 (1985), 251-8.
J. Lambrecht, Kremer, Actes, 307-37.
F. Lövestam, StTh 41 (1987), 1-10.

19RSV, NEB, which is; NJB, that has power.

20NEB, NJB, money.
21NEB, earned enough for; NJB, earned enough to meet.
22RSV, by so toiling; NEB, in this way, by hard work.
23RSV, remembering.
24[to remember] is not in the Greek; RSV, how he said.
25NEB, happiness lies more in; NJB, there is more happiness in.
 51. PAUL’S SPEECH AT MILETUS. 20.17-38. 963

H. J. Michel, Die Abschiedsrede des Paulus an die Kirche Apg 20.17-35,

München 1973.
C. F. D. Moule, BJRL 47 (1965), 430-52.
J. Munck, FS Goguel, 155-70.
E. Plümacher, ZNW 83 (1992), 270-5.
W. Pratscher, NTS 25 (1979), 284-98.
H. Sahlin, NovT 24 (1982), 188.
W. Schmithals, FS Schneider, 307-22.
H. Schürmann, FS Jaeger, 108-16.
H, Seesemann, TWNT 6.23-37.
K. Stowers, NovT 26 (1984), 59-82.
A. Strobel, NTS 15 (1969), 191-210.
B. E. Thiering, JBL 100 (1981), 59-74.
F. Μ. Young, JTS 45 (1994), 142-8.

Commentary
A brief introduction (vv. 17, 18a) and a concluding note (vv. 36-38)
provide a framework for a long address by Paul. The framework may
certainly be considered Luke’s own writing. Verse 17 provides a
suitable audience for the speech, which (see below) forms an
important part of Luke’s presentation of the figure of Paul. That Paul
(for whatever reason) did not on this journey visit Ephesus was a
datum of tradition; that he made a speech in Miletus may have been.
In any case, Miletus was the nearest spot touched on in his journey to
the great centre of population. Since Paul could not, or did not, go to
Ephesus, Ephesus was brought to him, represented by the elders of
the church, who could on behalf of the church as a whole hear Paul’s
farewell and learn what he had to teach about pastoral responsibility.
The speech completed, there must be leave-taking, the more serious a
matter because Luke makes it clear that this is a final farewell. The
matter is repeated in similar (Lucan) language in 21.5, 6. For stylistic
details see the notes.
The speech attributed to Paul raises a number of problems. It is
analysed by Pesch (2.199) as follows: Rückblick I (18b—21), Vor-
biick I (22-24b), Rückblick II (24cd), Vorblick II (25ab), Rückblick
III (25c), Unschuldsbeteuerung (26), Rückblick IV (27), Mahnung I
(28), Vorblick III (29-30), Mahnung II (31a), mit Rückblick V
(31b), Verabschiedung (32), Rückblick VI (33-35). What this list
means is that the speech is really incapable of analysis. In fact it is
not really quite as bad as this; see Dibelius (157) for a simpler and
more convincing analysis. It proceeds from point to point, with
numerous repetitions, as various ideas occur to the speaker (writer).
Its form is that of the farewell address: for the main biblical and some
non-biblical parallels to this form see J. Munck in FS Goguel
964 COMMENTARY ON ACTS

(155-70). Rackham (383) rightly points out that the themes of this
address are those of the farewell address of Lk. 21: Take heed; watch.
They are however put in an entirely different setting. The eschato-
logical basis and construction of Lk. 21 have virtually disappeared;
there is a warning of problems in the church (vv. 29, 30) but none of
a coming judgement. And there is apologetic reference to the
speaker’s example, which is not found in the gospel. Roloff (301)
takes the speech to be Luke’s work, on the ground of (a) the Lucan
style, (b) the fact that the church order presupposed is that of Luke’s
(rather than Paul’s) period, and (c) the perspective from which Paul
himself is viewed, which is that of a later age. In particular, v. 25 (cf.
vv. 37, 38) shows that at the time of composition Paul was already
dead. The origin of the speech is well discussed by Bauemfeind
(239). It appears ‘dass der Geist dieser Rede der Geist des Paulus ist.
Aber nicht der Buchstabe der uns bekannten paulinischen Briefe.’
‘Die Vorgeschichte der Rede werden wir uns also nicht anders zu
denken haben, als die Vorgeschichte der meisten anderen Reden in
der AG: Lk hat sorgsam auf ältere Predigtüberlieferung geachtet und
aus dem, was er vorfand, selbständig aufgebaut.’ This does not mean
that a primitive form of the speech, on which Luke worked, can be
found (though one has been suggested in vv. 18b, 19, 21, 24-28, 29,
30, 31, 32).
Two matters call for special attention. Notwithstanding the view
quoted above from Bauemfeind, a stronger case for Luke’s knowl-
edge of the Pauline epistles, or some of them, can be made here than
anywhere else in Acts. A modest list (see further NT parallels in
Weiser 571f.) of Pauline passages to which allusions can with some
plausibility be found may be given: Rom. 1.1; 15.14; 1 Cor. 1.2; 3.5,
9, 12; 4.12, 14; 10.13; 14.3, 4, 5, 12, 24; 15.9; 2 Cor. 1.1; 2.4; 4.5;
12.14f.; Gal. 1.13; Phil. 1.1; (Col. 1.20; 3.12, 24;) 1 Thess. 2.9, 14;
5.6, 10, 11, 12. This is an impressive list, but on examination it
proves to contain nothing that is truly convincing as a literary
allusion to the written Pauline corpus. If even here Luke shows no
acquaintance with Paul’s letters (and he gives nowhere any hint that
Paul ever wrote a letter) a considerable problem results. How could a
devoted admirer of Paul, writing up to a generation after Paul’s
death, be unaware of Paul’s literary legacy to the church? And if he
was aware of it, how is it that he never mentions it? For provisional
answers see FS Mitton 2-5.
The second matter to be considered is the relation between this
speech and the Pastoral Epistles (see the note on v. 17). That Luke
(that is, the author of Acts) wrote also the Pastorals is argued in
general terms by S. G. Wilson, Luke and the Pastoral Epistles,
London, 1979), by C. F. D. Moule (in BJRL 47 (1965),
430-52—with the view that Luke wrote them during Paul’s life and
under his supervision), and by A. Strobel (NTS 15 (1969), 191-210).
 51. PAUL’S SPEECH AT MILETUS. 20.17-38. 965

For the general question of the authorship of Acts see Introduction,

pp. xliv-liv. The parallels between this passage and the Pastorals are
so well summed up by Wilson that his account may be quoted
(117f.).

1. Paul looks back on his past career with some confidence,

believing that he has fulfilled the tasks designated for him (Acts
20.18-21, 25—6; 2 Tim. 4.6f.). Moreover, the striking metaphor
of an athlete finishing his race is used in both Acts 20.24 ... and
2 Tim. 4.7 ... At the same time he is deeply concerned with the
fate of the Church in his absence. This is indicated by the whole
of Acts 20.17-35 and each of the Pastoral letters.
2. The problem Paul foresees and warns of is heresy, which will
assault the Church from within and without (Acts 20.29-30; 1
Tim. 1.3f.; 3.1f.; 6.20f.; 2 Tim. 2.14f.; 3.1f.). The heresy
appears to be an early form of gnosticism and its centre is in
Ephesus (Acts 20.17f.; 1 Tim. 1.3). Paul urges constant alert-
ness (Acts 20.31; 2 Tim. 4.2f.).
3. The responsibility for resisting the false teaching is placed on
the church leaders or on Paul’s assistants. The church leaders
are, in both cases, elder-bishops (Acts 20.17-28; 1 Tim. 5.17; 2
Tim. 2.2; Tit. 1.5f.), and it is Paul’s example and instruction
which will be their chief weapon (Acts 20.27, 30-5; 1 Tim.
3.14; 4.11f.; 6.20; 2 Tim. 1.8f., 13-14; 3.10f.; Tit. 1.5).
4. Paul speaks of his own suffering for the sake of the gospel
(Acts 20.19-24; 2 Tim. 1.11-12; 2.3; 3.11) and indicates that
for him a martyr’s death lies ahead (Acts 20.25, 37; 2 Tim.
4.6f.).
5. The ministers whom Paul appoints and exhorts are warned of
the dangers of the love of money (Acts 20.33-5; 1 Tim. 6.9-10;
Tit. 1.11).
6. Paul commits his successors to the Lord and his grace (Acts
20.32; 2 Tim. 4.22).

The parallels are real and substantial, and there can be little doubt
that Acts and the Pastorals were produced in similar circumstances
and at times not very remote from one another. In Acts 20.17-34
(unlike some other parts of Acts) the motivation is very similar to
that of the Pastorals, and the resemblance is therefore greater than
elsewhere. There is however little in the Pastorals to parallel the most
explicit theological verse (28) in the speech, and when both works
are considered as wholes it must be remembered that the Pastorals
treat Paul as the pre-eminent apostle, whereas Acts is very reticent in
the use of the word απόστολος to describe him. But it remains true
that there is a clear relation between Acts and the Pastorals.
There is a very full account of earlier discussions of this speech,
966 COMMENTARY ON ACTS

and an important treatment of it, by J. Lambrecht (‘Paul’s Farewell

Address at Miletus (Acts 20.17-38)’) in Kremer, Actes 307-37.
17. From Miletus (v. 15) Paul sent to Ephesus (v. 16) and
summoned (μετεκαλέσατο; D has μετεπέμψατο, sent for, there is no
significant difference in meaning) the elders (τούς πρεσβυτέρους;
see on 14.23) of the church (τῆς ἐκκλησίας; see on 5.11, and
Introduction pp. lxxxviif.). Stählin (267) thinks that vv. 18, 25 show
that elders from other churches also were present; it seems a doubtful
inference. At v. 28 the elders are called επίσκοποι; see on that verse.
It is probable that we see here the constitution of the church as
known to Luke: a community led by a group known indifferently as
πρεσβύτεροι and ἐπίσκοποι. Paul never in the epistles uses the word
πρεσβύτερος; ἐπίσκοπος occurs once, in Phil. 1.1. Both occur in the
Pastoral Epistles—one of many contacts between these epistles and
this part of Acts (‘Almost every detail of Acts 20.17-33 can be found
in the Pastorals’—Wilson, Pastorals 117).
There was in the town of Ephesus not only a βουλή (Council) but a
γερουσία which might be thought to consist of πρεσβύτεροι (Hicks,
AGIBM, 3.74-78). This is not important; there were Christian
πρεσβύτεροι in many other places; there was no need to model them
on a particular form of local government.
18. παραγίνεσθαι is a Lucan word: Mt. 3 times; Mk, once; Lk., 8
times; Jn, twice; Acts 20 times; rest of the NT, 3 times. Only here in
Acts is it constructed with πρός; its frequency is of course connected
with the number of journeys described in Acts. After πρός αυτόν, P74
(A) D lat have ὁμόσε ὄντων αυτών, when they were all in one place;
a fairly obvious condition for speech-making. ὁμόσε occurs nowhere
else in the NT. E 2464 pc have ὁμοθυμαδόν, an Acts word.
υμείς is presumably emphatic (cf. 10.28; 15.7): You elders know,
if no one else does. ἐπιστᾶσθαι is characteristic of Acts: 8 times, in
the rest of the NT, 5 times. After ἐπίστασθε, D 2464 pc sa have
ἀδελφοί, evidently feeling that some form of address was called for.
It is surprising that we do not have ἄνδρες ἀδελφοί; cf. 1.16. Paul’s
work in Asia began in Ephesus (18.19; 19.10), and it is likely that
many of the elders would have been chosen from among the earliest
converts. They had had the opportunity of observing Paul’s ministry,
which he describes in the following verses, at first hand and from an
early time. A very early time: από πρώτης ἡμέρας. There is no
article. According to Μ. 3.179 this is the effect of the preposition.
BDR § 256 n. 3 makes a useful distinction. When an ordinal and a
word of time occur the article seems to be present only when the
reference is to a particular definite time; it would be implied that here
first day really means early days.
ἀφ9 ης: the preposition is repeated. Contrast 1.22 and the classical
use. BDR § 293.3e, n. 14 suggest that repetition is due to the
 51. PAUL’S SPEECH AT MILETUS. 20.17-38. 967

separation of the initial preposition from the relative, but only two
words intervene. According to Blass (219) ἀπό πρ. ἡμ. ἀφ’ ἦς =
ἀφ’ ἧς ημέρας τό πρώτον.
ἐπιβαίνειν is characteristic of the later part of Acts: 20.18; 21.2,4;
25.1; 27.2; elsewhere in the NT only Mt. 21.5.
For πώς with γίνεσθαι see BDR § 434.2, n. 5: how I conducted
myself. Cf. the reading of D (see below): ποταπῶς ... ήν.
The accusative τόν πάντα χρόνον now rightly indicates extent of
time (contract 20.16). πάντα preceded by the article means the whole
time; see Μ. 3.201 (all that time); BDR § 275.3, n. 6.
At the end of the verse, instead of πώς ... ἐγενόμην, D has ὡς
τριετίαν ή καί πλεῖον ποταπῶς μεθ’ ὐμῶν ἧν (the only occurrence
in the NT of this form of the first person singular; see Μ. 1.56; Μ.
2.203; BDR § 98, n. 1) παντός χρόνου. The three year period may
be derived from 19.8, 10; and see v. 31; the genitive (which usually
expresses time within which) is less apt than the accusative.
The speech begins in the manner of a farewell, with a defence of
the speaker’s conduct; cf. e.g. 1 Sam. 12.2f.
19. This verse contains a striking number of parallels with the
Pauline epistles. They are not quotations; they do not prove that Luke
had read the epistles (on this question see Introduction, pp. xxviif.,
xlivf. but they do show contact with the Pauline tradition, so that this
verse may be said to depict a man who could have written the
letters.
δουλεύειν τῷ κυρίφ: see especially Rom. 12.11 (unless τῷ καιρῷ
is read); also Rom. 16.18; 1 Thess. 1.9; in addition, passages in
which Paul describes himself as δούλος 'Ιησοῦ Χριστού, or the like
(e.g. Rom. 1.1; 1 Cor. 7.22; Gal. 1.10; Phil. 1.1).
He serves μετὰ πάσης ταπεινοφροσύνης: cf. Rom. 12.16; 2 Cor.
7.6; 10.1; 11.7; 12.21; Phil. 2.3; 4.12. The use of πας also is Pauline:
2 Cor. 8.7; 12.12.
δακρύων: cf. 2 Cor. 2.4.
πειρασμών: for the biblical history of this word, from its primary
Greek sense of trying or proving to affliction, disaster, punishment,
see Hatch (Essays 71f.); H. Seesemann (TWNT 6.23-37). It is clear
that Paul (Luke) is here thinking of afflictions; that these arose
mainly through ἐπιβουλαις τών 'Ιουδαίων is attested (possibly with
exaggeration) in Acts repeatedly (e.g. 13.45); see also 1 Thess. 2.15.
Titus 1.10, 14 are to be put with the evidence of Acts. In the NT
ἐπιβουλή is used only in Acts, always in relation to Jewish
opposition to Paul (9.24; 20.3, 19; 23.30). Paul speaks of the Jews
almost as if he were not one: Bengel (468), ‘Gentium apostolus jam
quasi de alienis loquitur.’ Yet so he does in 1 Cor. 9.20.
20. ὡς is still dependent on ἐπίστασθε (v. 18; in this verse note
πώς). See BDR § 396.1, n. 4.
968 COMMENTARY ON ACTS

Here and at v. 27 Paul emphasizes that he has withheld nothing;

the elders and indeed the whole church at Ephesus have received
from him the whole of Christian truth, ὑποστέλλεσθαι (middle) is
taken by BA 1688 to include always the element of fear or cowardice
that is certainly present at Heb. 10.38: in the present verse ‘feige
verschweigen’, in v. 27 ‘sich schauen vor, aus Angst vermeiden oder
unterlassen’. This does not seem to be justified; see LS 1895f. There
is a good parallel in Plato, Apology 24a: ούτε μέγα ούτε μικρόν
ἀποκρυψάμενος ἐγὠ λέγω οὐδ’ ὑποστειλάμενος. Luke’s motive
here does not seem to be the desire to point out Paul’s courage in
speaking unwelcome truths but rather to emphasize the completeness
of his Gospel. He probably has in mind the secret teaching of the
gnostics and their claim that they possessed a secret tradition from
Paul which he had withheld from the church at large. ‘Hier endlich
entdecken wir geringfügige Spuren der Abwehr einer der späteren
Gnosis vergleichbaren Haltung’ (Schille 402). So also Conzelmann
(117), adding, ‘Über die Gnostisierung des paulinischen Missionsge-
bietes: Bauer, Rechtgläubigkeit bes. 235ff.’ Against this Luke asserts
that the public tradition of such great Christian centres as Ephesus
went back to the apostle and contained the whole of his teaching. Cf.
Col. 1.28, where the same motivation is to be seen. Paul withheld
ούδέν τών συμφερόντων; it is implied that anything beyond the
known and publicly recognized teaching of Paul was not profitable.
Paul made his proclamation to you and taught you δημοσία και κατ’
οἴκους, that is, in public gatherings, and in the private meetings that
Christians held in their own houses. ‘... publicly (as at Athens) and
privately (as at Troas)’ (Hanson 203). For the adverbial use of
δημοσία cf. 16.37; 18.28; perhaps 5.18.
The infinitives ἀναγγεῖλαι and διδάξαι are introduced by τοῦ μή.
They have a consecutive sense (Μ. 3.142) and μή is usual after verbs
of negative meaning (as ὑποστέλλεσθαι). In the present verse (cf.
also ν. 27) μή is omitted by D. The negative is not always used (e.g.
Xenophon, Memorabilia 1.2.34, δῆλον ὅτι ἀφεκτέον ἄν εἵη τού
ὀρθώς λέγειν) and it must be left open whether the Old Uncial or the
Western text was making a stylistic improvement.
ὑμᾶς is also omitted by D syp Lucifer.

21. διαμαρτυρόμενος, a Lucan word (Lk. once; Acts 9 times; the

rest of the NT, 5 times), repeated in a similar way at v. 24 (cf. v.
23).
Ίουδαίοις τε και Έλλησιν is a common Pauline thought and
formula; see Rom. 1.16; 10.12; 1 Cor. 1.24; 10.32; 12.13; Gal. 3.28;
Col. 3.11. In Ephesus Paul had begun his work among Jews (19.8)
but had moved beyond the synagogue (19.9).
The content of Paul’s testimony (cf. v. 24, where it has a related
but different content) is described as τήν εις (P74 A D Ψ <a> add τόν)
 51. PAUL’S SPEECH AT MILETUS. 20.17-38. 969

θεόν μετάνοιαν καί πίστιν εις τὸν κύριον ήμῶν Ίησοῦν. This
phrase has only a superficially Pauline sound. Paul very seldom
refers to repentance, and for faith in Christ prefers the genitive πίστις
Ι' ησοῦ Χριστού (but see on this M. D. Hooker, NTS 35 (1989),
321-42, and works cited there). Conzelmann (117) thinks that the
phrase recalls the first and second articles of the creed; it might be
more apt (though not altogether different) to recall Rom. 10.9.
Repentance is called for because men have not treated Jesus as Lord;
faith is summed up in acceptance of the resurrection. Paul himself
prefers the more objective statement; Luke, like the Pastorals, is apt
to give a subjectivized version of Paulinism. The two elements in
conversion, repentance and faith, are introduced by one article (τήν),
which has the effect of binding them closely together. Zerwick § 184:
‘... subsumptio plurium sub uno articulo eorum aliquam unionem si
non identitatem indicat... Pulchre A 20.21 obiectum praedicationis
dicitur esse “conversio ad Deum et fides in Christum”, sed sub uno
articulo, ut fere intellegas “conversionem ad Deum per fidem in
Christum”.’ It is interesting in the light of this to note the reading of
D. The text quoted above (that of NA26) is contained in B H L P Ψ
614 pm gig syh; P74 A C E 33 36 323 945 1175 1241 1739 2495 pm
vg syp bopt add Χριστόν (E omitting ἡμῶν); but D has διά τού
κυρίου 'Ιησοῦ Χριστού. One cannot think that the scribe of D was
evaluating the single article in the manner of a modern grammarian.
If there was a serious motive for the change it will have been the
conviction that faith as well as repentance was directed towards God
and that Jesus was the mediator rather than the object of faith. More
probably the Western editor was simply indulging, as he seems
frequently to do, a love of variation for its own sake.
A further suggestion (mentioned but not accepted by Marshall
331) is that the verse is expressed in chiasmus: Jews needed faith in
Jesus Christ, Greeks needed to turn to God. Marshall is right to
question this: ‘Faith was necessary for all converts... and repentance
was also needed in the case of Jews’. For the latter point it would
suffice to refer to 2.38.

22. From this point (to which he will return; v. 25) Paul turns to
the future and the lot that awaits him. For καὶ νῦν ιδού cf. 13.11 and
the note. It is probably correct to describe the expression as a
Septuagintalism; cf. Gen 12.19. Knox, Hell. El. 17, observes that in
the ‘We’ passages the phrase καί (...) ιδού occurs only here and at
20.25; 27.24, all on Paul’s lips. This need not mean more than that it
belongs to spoken style.
The construction and precise meaning of δεδεμένος ἐγὼ τφ
πνεύματι are not clear. In a literal sense Paul was not yet bound (as
later he would be); he was however under divine constraint (and
the verb can refer to inward constraint and compulsion: Apollonius
970 COMMENTARY ON ACTS

Rhodius, Argonautica 4.880, τῶ μιν ἀμηχανίη δῆσεν φρένας; other

examples in LS 383) to continue on a course that was certain to lead
sooner or later to his arrest. Moreover, as the next verse states expli-
citly, the Holy Spirit was constantly witnessing to him that this would
be the result of his actions, τω πνεύματι could refer to Paul’s own
spirit or to the Holy Spirit. Blass (219) thinks (in view of v. 23) that the
Holy Spirit cannot be intended; Bauemfeind (239), cf. Weiser (576),
is convinced that it is. Bound in the Spirit seems a suitable English
translation since it too is imprecise, but suggests control by the Spirit
(especially if we write Spirit with a capital S) and is at least not
inconsistent with spiritual foreshadowing of what is to come.
Paul must go to Jerusalem (19.21), at whatever cost; this (what-
ever we make of the present verse) is represented as not only Paul’s
intention but as God’s (the Spirit’s) will. As he goes he does not
know (εἰδώς) what will befall (συναντάν, normally as at 10.25, to
meet (of persons)) him there. That is, he does not know in detail what
form imprisonment and affliction (v. 23) will take.
Instead of εἰδώς, P41 D have γινώσκων. There is no difference in
meaning. Behind the variant is simply the freedom an editor felt in
changing the text before him. See Introduction, p. xxi, and many notes.

23. πλὴν ὅτι, except that, is classical; so Knox (Hell. El. 12), also
BDR § 449.1, n. 3, who point out that this is the only place in Acts
where πλήν is used as a conjunction.
Paul does not know (v. 22) what will happen to him in Jerusalem,
only that the Holy Spirit διαμαρτύρεται (in v. 21 Paul uses this word
of his testimony to the Gospel) that bonds and affliction await him.
The visit to Jerusalem will be a painful one; he must face imprison-
ment again.
The masculine singular δεσμός takes here a neuter plural in –a. See
Μ. 2.121f.: ‘Δεσμός has plural δεσμά in Lk. 8.29; Acts 16.26; 20.23,
δεσμοί Phil. 1.13, the rest being ambiguous. Thackeray (p. 154)
observes that in LXX as in NT δεσμό is literary. Rutherford (Gram. 9)
asserts after Cobet, that δεσμά = actual bonds, δεσμοί = bondage.
The distinction cannot be pressed for the NT, though it would suit very
well: Acts 20.23 gains vividness from it.’ For examples of the Spirit’s
testimony see 21.4 (in Tyre), 10-14 (in Caesarea).
The Holy Spirit bears witness κατὰ πόλιν, in city after city, κατά
being used, as frequently, distributively. The addition of πάσαν (D
gig (vg) sy) adds nothing to the sense. Some editors or copyists took
ὅτι to be recitativum, and changed με into σε, thus: ὅτι δ. καί θ. σε
μένουσιν ἐν ’Ιερουσαλήμ (P41 (D 614 gig vgcl syh**) sa).
Betz (110, 119) remarks that the θεῖος ἀνήρ knows his approach-
ing end. On apostles as θειοι ἄνδρες see e.g. 3.12: what is to be
observed here is that Paul professes ignorance of exactly what will
happen, that he does not know anything in himself but is given
 51. PAUL’S SPEECH AT MILETUS. 20.17-38. 971

limited information by the Holy Spirit, and that the Holy Spirit gives
the same information to all who happen to be present at the time.

24. The text and the construction of the opening clause are alike
obscure. The text of ΝΑ26, οὐδενὸς λόγου ποιούμαι τήν ψυχήν
τιμίαν έμαυτφ, is read by P41 Β C 1175 pc (gig syp). There are
several ways in which it may be construed. We may take the first five
words as virtually complete in themselves: I take my life to be of no
account. τιμίαν ἐμαυτῷ is then an epexegetical addition: as being of
value to myself. Or, τιμίαν may be taken as equivalent to αξίαν. We
should then have: I take my life to be worth to myself not a single
word, that is, to be worth nothing. A further suggestion is that two
clauses, each of which could have stood alone, have been combined:
οὐδενός λόγου ποιούμαι, I take no account of anything (i.e. of any
danger), and οὐκ ἔχω τιμίαν ἐμαυτῷ τήν ψυχήν, I do not hold my
life precious to myself. The variants similarly divide the clause in
two: (a) οὐδενός λόγον ἔχω (+ μοι, D) οὐδἐ ποιούμαι τήν ψυχήν
τιμίαν έμαυτφ (P74 A D(c) 33 pc) and (b) οὐδενὸς (+ τούτων
1891 pc) λόγον ποιούμαι οὐδἐ έχω τήν ψυχήν μου τιμίαν ἐμαυτῷ
(Ε Ψ <a> (syh)). (a) will be rendered, I take account of nothing nor do
I count my life as valuable to myself, (b) I take account of none of
these things nor do I hold my life precious to myself. NA26 rightly
choose the most difficult of the three texts; and though it is difficult
the general sense of the text is clear. Life itself is worth less to Paul
than the fulfilment of his apostolic calling. Metzger (479) says that
the text approved here is ‘awkward, yet idiomatic’; similarly Ropes
(Begs. 3.196). Field conjectures ἀλλ’ οὐδενὸς λόγου ποιούμαι ουδέ
ἡγοῦμαι τὴν ψυχήν μου τιμίαν ἐμαυτῷ.
The next two words are marked by further textual and linguistic
problems, ὡς is read by P41 P74vid A B* C Ψ <a>, but 2 B2 vg have
ἕως; E 33 323 614 945 1739 2495 al have ώστε; D gig Lucifer have
τοῦ (quam). τελειῶσαι is read by P41vid A (C) D Ε Ψ HR gig Lucifer,
but X B pc vg have τελειώσω. Among these variants έως may be
discounted as an easy complement to τελειώσω. ὥστε, with, on the
whole, late attestation, should probably be regarded as an easy
companion to the infinitive τελειῶσαι though Blass (220) points out
that it would be easy for τε to drop out before τελειῶσαι. The
comparative contained in the Western text ( ... than completing ... )
is again a relatively easy reading. We are left with ὡς τελειῶσαι and
ὡς τελειώσω. Each is unusual, but neither is impossible. If ὡς
τελειῶσω is read it is, according to Μ. 3.105, the only final ὡς in the
NT. Turner adds that it is rare also in the Koine. [He refers also to a
variant ὅπως, but I do not know where this is to be found.] ὡς may
also however be read with the infinitive, again in the final sense, and
this is presumably intended by NA26. There is a parallel in Lk. 9.52,
where ώς έτοιμάσαι is read by P45 P75 B, but ὥστε ἐτ. by all other
972 COMMENTARY ON ACTS

MSS. See BDR § 369, n. 1; § 391.1, n. 2; BA 1793 (s.v. ὡς, IV 3).

The meaning, whether τελειῶσω or τελειῶσαι is read, is, My
purpose in discounting the value of my life is that I may ...
For τελειοῦν τον δρόμον cf. 2 Tim. 4.7 (τετέλεκα). Paul frequently
uses images drawn from the games, including the verb τρέχειν
(δραμ-), but never the word δρόμος. Ε Ψ <a> syh add μετά χαράς.
After the image of the race Paul speaks more directly of the work
he is concerned to complete. It is the διακονία ἣν ἔλαβον. P41vid P74vid
614 2495 have ἣν παρέλαβον; D(c) gig vgcl have τού λόγου ὅν
παρέλαβον. The variants make little difference to the sense, unless
διακονία is taken to refer (as it does at Rom. 15.31; 2 Cor. 8.4; 9.1,
12, 13) to Paul’s collection for the poor in Jerusalem. This is very
unlikely in view of the fact that Acts has made no reference to this
collection. The διακονία that Paul had received from the Lord Jesus
was to testify (διαμαρτύρεσθαι, as at vv. 21, 23) τό εὐαγγέλιον τής
χάριτος τοῦ θεοῦ. τής χάριτος must be taken as a genitive of
content. The good news was that God was bestowing his unmerited
favour upon, was being gracious to, the human race. The addition
'Iουδαίοις τε καὶ Έλλησιν (p41vid D gig samss) is a secondary
assimilation to v. 21, but is clearly in harmony with the speaker’s
intention. The content of Paul’s preaching is given in other allusive
summaries in vv. 25, 27, 28, 32. For the Gospel of the grace of God
cf. the word of his (the Lord’s) grace at 14.3; 20.32.

25. With καί νῦν ἰδού (see ν. 22) Paul again turns briefly from the
past to the future. This is a farewell speech; see above.
οὐκέτι ὂψεσθε τὸ πρόσωπόν μου is clear, and it is surprising that
there seem to be few or no parallels to this mode of expression. Cf.
Gen. 48.11 (Hebrew: Greek different). Whether
οὐκέτι is translated no longer or no more it seems clear (cf. v. 38)
that Paul is predicting that personal contact is at an end. It is
impossible to draw inferences (but see Hamack, The Acts of the
Apostles 293f.) from this for the date of Acts; we do not know
whether Paul was released from a first imprisonment in Rome and
returned to the East. Cf. Phil. 1.25, where Paul himself uses οἶδα in
forecasting release. But Luke could hardly have written this verse
(and ν. 38) if he had known that Paul returned to Asia. So also
Haenchen (566).
υμάς is not so much emphatic as intended to take up the link
between the second person plural ὄψεσθε and Paul’s recollection, in
the second part of the verse, of the manner and content of his ministry.
Since the πάντες that follows is so closely attached to this υμάς it
cannot be concluded (as it is by Haenchen 566) that it is intended to
refer to all the Pauline communities, of whom he now takes leave. The
words do indeed have this secondary reference: the speech serves as a
farewell to all; but it is only a secondary reference.
 51. PAUL’S SPEECH AT MILETUS. 20.17-38. 973

For διῆλθον cf. 9.32; if the whole of Asia is in mind (cf. 19.10)
there may be a reference to missionary tours. But probably the writer
is thinking only of Paul’s long residence in Ephesus during which he
went about proclaiming the kingdom. This short text (of P74 A B C
Ψ 33 36 453 pc sy bopt) is almost certainly correct. Its brevity
attracted ‘improvements’: τὴν βασ. τοῦ (+ κυρίου, gig) Ίησοῦ (D
gig sa Lucifer); τὴν βασ. του θεοῦ (Ε <a> vg bopt Theodoret); τό
εὐαγγέλιον τού θεού (323 1891 pc). For the use of βασιλεία (τού
θεού) in Acts see 8.12; 19.8; 28.23, 31. It means in effect the
recognized content of Christian preaching, and is so expressed in
order to bring out the continuity between the preaching of Jesus and
the preaching of the post-resurrection church.
26. Looking back to his ministry in Ephesus (or Asia) Paul affirms
in the strongest terms the testimony of his conscience to his
blameless behaviour and in particular to his complete openness in
declaring the truth about God and his purpose for mankind.
μαρτύρομαι, a strong word: asseverate. Cf. 26.22; Gal. 5.3; Eph.
4.17; 1 Thess. 2.12.
ἐν τῇ σήμερον ήμερα also is emphatic, σήμερον occurs 21 times
in Acts, here only with ἡμέρα (Acts 27.33 is not parallel). Paul uses
the combination at Rom. 11.8 (quoting the OT); 2 Cor. 3.14. This
very day, however, would perhaps be too strong.
At the beginning of the verse D* has ἄχρι οὖν τής σήμερον
ἡμέρας and C D2 H L al have διό (instead of διότι; not noticed in
NA26). These variants may be due to the fact that διότι (probably
original) does not give the required sense. ‘διό means and so (cf.
άρα) and διότι means because, since. See MM s.vv. [163f.; 164f.].
But in Acts 20.26, where the context certainly requires and so, there
is considerable MS support for διότι’ (Moule, IB 164).
For καθαρός cf. 18.6; also, for the sense, Ezek. 3.18-21; 1 Thess.
2.10. The formulation is Lucan rather than Pauline. The construction
with από is not a Hebraism; see Deissmann (BS 196), with reference
to ten Fayyum papyri as well as Demosthenes 59.78 (1371) (... εἰμί
καθαρὰ καί ἁγνή ἀπό <τε> τών άλλων ...) and a Pergamum
inscription. The meaning is clear: Paul is responsible for no man’s
(eternal) death through neglecting to preach the Gospel to all and to
deliver it in all its fullness. H. Sahlin (NovT 24 (1982), 188) proposes
instead of τού αἵμ. π., αἰτιώματος παντός. It does not seem a
necessary conjecture.
27. For ὑποστέλλεσθαι see v. 20; also on μή (omitted here by D*
pc). Here ἀναγγέλλειν, at v. 20 left without an object (but see v. 21),
has an object: πάσαν τήν βουλήν τοῦ θεοῶἐῦ. βουλή is a Lucan, and
especially an Acts (2.23; 4.28; 5.38; 13.36; 19.1 (v.1.); 20.27; 27.12,
42), rather than a Pauline (1 Cor. 4.5 only) word. Here it must refer to
the saving purpose of God for the human race. Paul has proclaimed
 COMMENTARY ON ACTS

this in its entirety. See J. T. Squires, The Plan of God in Luke-Acts,

SNTSMS 76 (1993).
For Luke’s motivation in emphasizing this point see on v. 20. At
this point Conzelmann (118) adds a reference to ‘die Esoterik der
Qumran-Sekte, 1 QpHab 7.1ff.; 1 QS 3.13f.; 11.5ff.’. He continues,
‘Zu dieser polemischen Zuspitzung vgl Iren[aeus] Haer[eses] 1.1;
Tert[ullian] Praescr[iptione] Haer[eticorum] 22f.; W. Bauer bei
Hennecke2 139.’

28. This verse is both the practical and the theological centre of the
speech; the practical centre, because Paul’s primary intention is to
urge the Ephesian elders to do their duty effectively—or rather,
Luke’s primary intention is to convey the same exhortation to his
contemporaries, and the theological centre, because here only in Acts
is there an attempt to state the significance of the death of Christ and
at the same time to bring out the ground of the church’s ministry in
the work of the Holy Spirit.
προσέχειν (sc. τόν νουν) in itself means to direct the mind
towards, to give thought to, but this naturally extends along the line
of with a view to the advantage of i.e. take thought for. Here the
elders are bidden in the first instance to take thought for themselves,
that is, to maintain the quality and integrity of their own Christian
life, and in the second place to take thought for the church in which
they hold a responsible position. Their position has so far (v. 17)
been described as that of elders; another word will shortly be used,
but before that the image of the flock, and therewith by implication
that of the shepherd, is introduced. The noun τω ποιμνίῳ is taken up
in the verb ποιμαίνειν, but this is preceded by another noun,
ἐπίσκοπος. The language of shepherding is—perhaps surprisingly—
not Pauline, but later the image became common; see Jn 21.15-17
(cf. 10.11, etc.); 1 Pet. 2.25; 5.2-4; Eph. 4.11; Heb. 13.20: Jude 12. It
rests upon familiar OT passages; e.g. the story of David; Ps. 23; Jer.
3.15; 23.1-4; Ezek. 34.1-24; but it is also used of rulers and leaders
in the non-biblical world. See St John 373f. The shepherd directs his
sheep, knowing where they ought to go in their own interests, and
also protects them, against wild beasts and robbers. The Christian
shepherd is one who is able to guide and also to protect against the
agencies that mislead and endanger Christians.
It was the Holy Spirit who appointed (τίθεσθαι, middle, not
differing in meaning from the active, and taking a double accusative)
the elders. As a procedure this does not necessarily differ from 14.23,
but throws the process further back. Paul may appoint elders, but
only those whom the Holy Spirit has already singled out and has thus
himself appointed. The ministry is not appointed from below, nor
from above if this means by those already ministers; the Holy Spirit
is at work in the church choosing and preparing by his gifts those
 51. PAUL’S SPEECH AT MILETUS. 20.17-38. 975

who are to be ministers. The Holy Spirit appointed them in order that
they might shepherd (ποιμαίνειν, infinitive of purpose; Radermacher
(153); BDR § 390.2, n. 5) God’s flock. In this sense it is true that
there is ‘im Sinn des Lukas ein gewisses Gegenüber von Amt und
Gemeinde’ (Weiser 584).
It is clear that the same persons, who act as shepherds, are
described both as πρεσβύτεροι (v. 17) and as ἐπίσκοποι (v. 28). That
the two terms are applied equally to the same persons does not mean
that they are identical in meaning. Thus ‘ “Bischöfe” bezeichnet hier
die Aufgabe der Ältesten’ (Conzelmann 119). Similarly Weiser (578)
says that ἐπίσκοπος is not an Amtstitel but a Funktionsbezeichnung.
Cf. H. von Campenhausen (Amt 87f.). Not quite the same is the view
of Schneider (2.296) (and others) that we have here a combination of
the ‘elder’ pattern of church order with the Pauline (Phil. 1.1) pattern
of bishops and deacons. This view is not helped by the absence of
deacons. Cf. also Roloff (305): Jewish based elders were combined
with Hellenistic ἐπίσκοποι. It is broadly speaking true that the one
designation describes ministers from a sociological, the other from a
theological angle. It seems that up to the time when Acts was written
elders were for the most part found among the older members of the
church, those at least who had been Christians longest, and that they
exercised the kind of leadership that older members may be expected
to give, guiding the younger by their experience and accumulated
wisdom (cf. 1 Pet. 5.1,5; and the Pastorals passim). Yet the church is
not a school or club, and the noun ἐπίσκοπος is related to the verb
ἐπισκέπτεσθαι, which suggests the saving act in which God visited
and redeemed his people, and the kind of ministry in which this
redeeming visitation is constantly applied and brought before the
minds of church members. This is not a simple function of age. Here
it is closely connected with the work of a shepherd. Elsewhere in the
NT the word ἐπίσκοπος is used at 1 Peter 2.25 (of Christ or God, in
parallel with ποιμήν) and at Titus 1.7 (where it is certainly used of
elders); 1 Tim. 3.2 (where it is probably used of elders; see Pastoral
Epistles 32, 58); Phil. 1.1 (along with διάκονοι, and with no
reference in the epistle to πρεσβύτεροι). Its general equivalence (but
not synonymity) with elder is clear. It is used of officials in various
social groups (see evidence in BA 606); it is used by Philo of Moses
(Quis rerum divinarum Haeres? 30, ὁ ἐπίσκοπος Μωυσῆς, who
keeps watch over men and affairs; it is used of the Cynic-Stoic
preacher (see H. W. Beyer in TWNT 2.606f.). The background of the
NT ἐπίσκοπος has also been found in the of the Qumran
community; see especially CD 13.7-9: This is the rule for the
overseer of the camp he shall instruct the Many
in the works of God and shall give them understanding of his
marvellous acts of power and shall recount before them the eternal
events. And he shall have pity on them as a father with his sons and
976 COMMENTARY ON ACTS

shall bring them back as a shepherd his flock (omitting

two unintelligible words). On this see Haenchen (567); but the best
discussion is that of Fitzmyer (Essays 293f.). He concludes, ‘Grant-
ing then the common etymological meaning of episkopos and mbqr,
and certain similar functions, it is nevertheless difficult to set up any
direct connection between the Essene “Overseer” and the institution
of the early Jewish Christian church in Palestine.’ He thinks that
there may possibly have been such a connection outside Palestine; it
is however difficult to envisage this. It is probable that the word was
picked up by Christians from secular institutions but that they were
encouraged to use it by associations which came mainly through the
verb in biblical Greek. See also B. E. Thiering (JBL 100 (1981),
59-74, ‘Mebaqqer and Episkopos in the Light of the Temple Scroll’.
There is no suggestion of any succession in episcopal office; there is
no need for succession (in the commonly understood sense); the
Holy Spirit will provide ministers as they are required.
The elders or bishops are to act as shepherds for the church of
God, τήν έκκλησίαν (for this word in Acts see 5.11 and Introduction,
pp. lxxxviif.) τοῦ θεοῦ. τοῦ θεοῦ is the reading of X B 614 1175
2495 al vg sy boms Cyril. Other readings are τού κυρίου (P74 A C* D
E Ψ 33 36 453 945 1739 1891 al gig p syhmß co Irenaeuslat Lucifer)
and τού κυρίου και (τοῦ pm) θεού (C3 <a>). The reading τού θεού
causes so much difficulty when taken with the following words (see
below) that it must be accepted as original, the other two readings
being attempts to ease the difficulty.
τού θεού provides the subject of the relative clause that follows,
ἥν περιεποιήσατο. The verb is often taken to mean simply to acquire
(as at 1 Tim. 3.13). Begs. 4.261f. argues for the meaning rescued.
Its meaning in the OT and NT seems to be prevailingly ‘save alive’,
or ‘rescue from destruction’. This observation is justified; see e.g.
Isa. 31.5, περιποιήσεται καὶ σώσει; 43.21, λαοῦ μου ὅν
περιεποιησάμην; also Luke’s own use at Lk. 17.33. Marshall adds
Ps. 74 (73).2, ‘which significantly follows a verse in which Israel is
likened to a flock’. The two ideas are very close together: God
acquired a people by saving them.
He did so διά τού αίματος τού ίδιου; so P74 A B C D E Ψ 33 36
945 1175 1739 1891 al Cyril. <a> has διά τού ἰδίου αίματος. The
latter reading can only be translated, through his own blood, which is
very difficult theologically unless in the preceding clause we have (as
does) κυρίου. Salvation through the blood of the Lord Jesus
Christ raises no special problems, but it is probably incorrect to read
anything other than θεού. The reading of B etc., which must be
accepted, may, like that of <a>, be translated through his own blood,
but can also be translated, through the blood of his Own, his Own
being taken as a title of Christ, God’s own Son. This is rejected by
Turner (Insights 14f.) who thinks, mistakenly, that it leads to a denial
 51. PAUL’S SPEECH AT MILETUS. 20.17-38. 977

of the deity of the Son. In any case, of course, this would not be a
reason for rejecting a reading required by grammar. Μ. 1.90f. is more
inclined to defend it, noting that ‘in the papyri we find the singular
[of ίδιος] used thus as a term of endearment to near relatives, e.g. ὁ
δεῖνα τω ἰδίω χαίρειν.’ Schneider (2.297) similarly thinks that ὁ
ίδιος corresponds to ὁ αγαπητός and ό μονογενής. Begs. 4.262
adopts ‘of his own’ since ‘it is inconceivable that “his own blood” is
right’. See also Hort, Introduction: Notes 98-100 (perhaps υἱοῦ
should be inserted); Ropes in Begs. 3.197-9. It is very unlikely that a
trained theologian would write ‘his own blood’; but Luke was not
such a theologian, and the natural way of reading the Greek should
probably be adopted. It was enough for Luke that when Jesus Christ
shed his blood on the cross he was acting as the representative of
God; he was God’s way of giving life, blood, for the world. Roloff
(306) makes the good suggestion that Luke was putting together two
formulas: the church of God; Christ has acquired his people by blood
(that is, by atoning death). But he was not careful to change the
subject. An explanation of this kind is more convincing than
Wettstein’s long (2.596-9) note on patristic references to blood and
Calvin’s theological explanation (2.184): ‘... because ... the two
natures are so united in Christ as to constitute one Person, what
properly belongs to the one is sometimes improperly transferred to
the other. For example in this verse Paul attributes the blood to God,
because the man Jesus Christ, who shed His blood for us, was also
God. The figure of speech was called the communicatio idiomatum
by the Fathers, because the property of one nature is applied to the
other.’ He goes on to warn against the errors of Nestorius and
Eutyches. Cf. Rackham (393), with references to Ignatius, Ephesians
1.1; Tertullian, Ad Uxorem 2.3. Communicatio idiomatum will not
really serve here because we are dealing not with the two natures of
Christ but with two Persons of the Trinity.
How the blood shed on the cross saved a people for God is a question
which it would be wrong to pursue at great depth because it probably
did not occur to Luke to inquire deeply into it. Derrett (Law 403) is
right in making a distinction: the blood shed is not a price paid, though
the redeemer does acquire rights over the redeemed. And Delling
(Kreuzestod 94) is right in saying that the shedding of blood does not
(or at least does not necessarily) imply the idea of sacrifice. Stephen
shed his blood (22.20), but his death was not a sacrifice. Luke nowhere
else asserts so plainly that it was the death of Jesus that brought the
redeemed people into existence, but the form—and indeed the exist-
ence—of his two-volume work, hinging on crucifixion and resurrec-
tion, implies so much. And with that Luke seems to be content.

29. Again the speaker turns to the future, this time with definite
prophecy of what will happen μετά τήν ἄφιξίν μου. The usual
978 COMMENTARY ON ACTS

meaning of ἄφιξις is arrival, and attempts have been made to give it

this meaning here—arrival in Jerusalem, in Rome, or in heaven
(after death). These all seem wildly improbable; a simpler and better
suggestion, though often overlooked, is that of Bengel (469): ‘Sensus
igitur est: primum venit Paulus; deinde venient lupi.' It seems
however that Paul (Luke) must mean departure; so already Chry-
sostom, also e.g. Moule (IB 89). Some parallels to this use of the
word (probably suggested by the compounded ἀπό) are adduced; not
all are quite convincing. BA 254 reduce the unambiguous cases to
Demosthenes, Epistles 1.2 (1463); 3.39 (1484) (even in these one
wonders if ἔνθαδε ἄφιξις might not have been intended to mean
arrival at home; Aelius Aristides 48.7k = 24p.467D; Josephus, Ant.
2.18 (τήν ἐκεισε ἄφιξιν; departure thither (Thackeray)); 4.315 (τής
πρὸς εκείνους ἀφίξεως; departure to them (Thackeray)); 7.247 (τήν
πρός τὸν Δανίδην ἄφίξιν; the journey to D. (Thackeray)); PMich
497.12. There seems to be ro example of the use of άφιξις in the
sense of departure in death (cf. Lk. 9.31, ἔξοδος), which may be but
is not necessarily intended here. Possibly some ambiguity is
intended: ‘As soon as I am out of the way.’
Paul’s departure will be followed by the entry of λύκοι βαρείς,
fearsome wolves. In D, the adjective is used at 20.9 of heavy sleep;
the common idea is oppressiveness. This is the meaning of the word
in Josephus, Ant. 15.354, where Herod the Great is warned of being
βαρύν ... καί τυραννικόν. The adjective is used with οργή and
χόλος at 3 Macc. 5.1, 30,47, but it remains somewhat surprising as a
description of wolves. Luporum ... rapacium (Horace, Carmina
4.4.50, cited by Schweizer, Beiträge 63) is more natural. G. W. H.
Lampe’s essay ‘Grievous Wolves’ (FS Moule 253-68), though
instructive in regard to early Christian heresy, casts no light on the
use of βαρεῖς. βαρύς can also mean powerful (e.g. Homer, Iliad
1.89; Xenophon, Agesilaus 11.12—but this may be an example of
βαρύτατος = fiercest) and it may be Luke’s point that the heretics
are formidable rather than ferocious.
It is the description of the church as a flock (v. 28, repeated in v.
29) that invites the image of wolves. The wolves will not spare the
flock; the trouble-makers will pursue their own ends regardless of
what becomes of the church. For the imagery, used in various ways
in the early church, see inter alia Didache 16.3; Ignatius, Phil-
adelphians 2.1,2; 2 Clement 5.2-4; Justin, I Apology 16,58; Trypho
35. For wolves in the OT see Ezek. 22.27; Zeph. 3.3.
The wolves come εις υμάς, that is, they come in from without.
They may be Jewish or Gentile teachers, importing their beliefs into
the church, or possibly oppressive authorities, though this would fit
less well with εἰς, which suggests that the persons in question made
their way into the church’s fellowship. Contrast v. 30.
For the fulfilment of this verse and of v. 30 see the Pastorals and
 51. PAUL’S SPEECH AT MILETUS. 20.17-38. 979

Revelation. On the chronological relation between this verse and the

Pastorals see Wilson (Pastorals 123). For the thought cf. Acts of
Thomas 67 (L.-B. 2.2.184f.), διατήρησον αυτήν από τών λύκων
τών διαρπαζόντων.
At the beginning of the verse some authorities give a closer
connection with what precedes: for ἐγῴ οἶδα, C3 Ε Ψ p syh have
ἐγώ γὰρ οἶδα τοῦτο; Β alone has ὅτι ἐγώ οἶδα.

30. In contrast with those who invade the church from without
men will rise up ἐξ υμῶν αὐτῶν (αυτών is omitted by B, but the
emphasis was probably intended), from among yourselves, out of
your own number. Heresy and schism will arise from within. Strictly,
υμείς should refer to the Ephesian elders, who are being addressed,
but Luke is probably now thinking of the church at large. ‘Ambition
is the mother of all heresies’ (Calvin 2.185); not true, but Luke might
have agreed.
Heresy is represented by the διεστραμμένα, the perverse things
that are spoken, schism by the desire αποσπάν, to draw away,
disciples. For διαστρέφειν cf. 13.10. For αποσπάν cf. e.g. Josephus,
Life 321, βουλόμενος δ’ αυτούς ἀποσπάσαιτῶν Τιβεριέων; Lucian,
Lepithae 26,... Δίφιλος ... δύο ἤδη μαθητάς μου ἀποσπάσας ... ;
Bengel (469): ‘abstrahere, a simplicitate erga Christum et ab unitate
corporis.’
There is no evidence in the present passage to suggest in which
ways the truth would be perverted or who would attempt to set up
rival communities over against what Luke regarded as the true
church. It is clear that a time is contemplated when orthodoxy and
heresy would be clearly differentiated, and when the church would
be an institution with clearly defined boundaries. It did not always
appear so simple. Gnostic teachers, for example, emerging within the
church would often regard themselves as its most faithful, or at least
as its most intelligent and advanced members. See Bultmann (The-
ologie 174); and on the whole question, W. Bauer (Rechtgläubigkeit
und Ketzerei).
For ἀποσπᾶν, D has ἀποστρέφειν, without difference in meaning.
For αυτών (at the end of the verse), A B have εαυτών; again there
is no difference in meaning.

31. From the future the speaker turns back to the remembered past.
The two—past and future—are deliberately mingled in this speech
and that which unites them is expressed in the imperative
γρηγορεῖτε. The whole is an exhortation to responsible pastoral
service, and this is reinforced both by the example given in the past
and by warnings of future perils.
γρηγορεῖτε recalls apocalyptic passages in the gospels (especially
Mk 13.34, 35, 37, and parallels), in Paul (especially 1 Thess. 5.6),
980 COMMENTARY ON ACTS

and elsewhere (e.g. Rev. 3.2, 3; 16.15); it is significant that here it

refers not to watchful preparedness for the coming of the Son of man
at the End but to vigilance in caring for the continuing life of the
church. For Vigilate as a component in early Christian catechesis see
especially E. G. Selwyn, The First Epistle of Peter (1946), Essay II,
363-466. Of such watchfulness Paul had given an example in a
ministry (in Ephesus) that lasted three years (see 19.8, 10; BDR §
5.3c, n. 17) and was maintained ceaselessly night and day. This claim
must be taken with the statement in v. 34 that Paul worked with his
hands to supply his own needs and those of others. Unless Luke was
very careless and forgetful he meant that Paul was ceaselessly
occupied; whenever he was not engaged in working for his living he
was working as evangelist and pastor.
It was work over which he shed tears; cf. 2 Cor. 2.4. It included
νουθετεῖν (Rom. 15.14; 1 Cor. 4.14; 1 Thess. 5.12, 14; also Col.
1.28; 3.16; 2 Thess. 3.15—found only in these passages: a Pauline
word), admonition (note also the noun, νουθεσία, 1 Cor. 10.11; Eph.
6.4; Titus 3.10), setting their minds in order, that is, so that they
should think, and consequently act, as Christians. He admonished
each one (ὑμῶν, D E 323 614 945 1739 1891 2495 al latt sy, is a
thoughtless addition; Paul admonished each member of the church,
not only each one of the elders there assembled); cf. Col. 1.28. This
may well be an anti-gnostic point: each one, not only the
πνευματικοί. Cf. vv. 20f.

32. Paul exhorts his hearers; he also commends them to God, that
is, to God’s protection and guidance.
For τὰ νῦν cf. 4.29; 17.30 and the notes. For παρατίθεσθαι
(middle) cf. Lk. 23.46; Acts 14.23; (17.3 for a different use). Paul has
watched over these elders; he now hands over the responsibility to
God, as Timothy must hand on the responsibility for Gospel truths to
faithful men (2 Tim. 2.2). But ‘der lukanische Paulus übergibt bei
seinem Abschied den Amtsträgem nicht das Wort als festes Lehrde-
positum, er unterstellt sie vielmehr dem Wort als der heilvoll in der
Geschichte wirkenden Kraft Gottes’ (Roloff 306f.).
παρατ. τῷ θεῷ is the text of P74 A C D Ε Ψ <a> vg sy samss; B
326 pc gig sams bo have π. τω κυρίῳ. This recalls the words of
Stephen at 7.59, and is probably due to assimilation. It is doubtful
whether Luke would have felt strongly about either reading as
against the other.
To God; καὶ τῷ λόγω τῆς χάριτος αὐτοῦ. This word is the
Gospel; cf. v. 24. This is an example of hendiadyoin; Luke means,
‘to God, who is active in the word of grace, which you proclaim and
by which you yourselves live’. So Blass (222): ‘ ... quasi una notio
sunt; agit enim Deus per verbum suum’.
If this phrase does in its two parts represent one concept, the
 51. PAUL’S SPEECH AT MILETUS. 20.17-38. 981

grammatical question of the reference of τῷ δυναμένῳ becomes less

important. The nearer antecedent is λόγος; θεός (or κύριος, if read)
might possibly be thought more appropriate. In fact it is the word of
God, God in his word, who is able to build up the elders themselves
and the church that they serve. For οἰκοδομεῖν (said of the church)
see 9.31; 1 Thess. 5.11 and other Pauline passages. For the church as
οικοδομή see 1 Cor. 3.9 (and 1 Corinthians 86-92).
God is able also to give a κληρονομιά (another Pauline word,
though Paul more frequently has κληρονoμεῖv and κληρονόμος).
The building up of the church belongs to the present age; the
inheritance is probably thought of as received in the age to come. It is
an inheritance that God will give and they will receive.
ἐν τοῖς ἡγιασμένοις: ἐν with the dative is not, as it sometimes is,
a substitute for the simple dative but means among; the Ephesian
elders take their place among a larger company (πᾶσιν) of those
who have been sanctified (Μ. 3.264f.). It is not easy to find a reason
why the perfect participle passive of ἀγιάζειν should be used (as at
26.18) instead of the adjective άγιος. In Acts the adjective is used
most frequently of the Holy Spirit, but not exclusively so; see 3.14,
21; 4.27, 30; 6.13; 7.33; 9.13, 32, 41; 10.22; 21.28; 26.10. A
number of these passages refer to Christians. The question is
discussed in relation to OT passages by Wilcox (35-7), but without
adequate ground for suspecting the use of an aberrant LXX text.
The OT background of inheritance is of course important; see Deut.
33.3, 4; Ps. 15(16).5; Wisdom 5.5. Also important are some
Qumran passages. See 1QS 11.7, 8: To those whom God chose he
had given them as an eternal possession, and he has made them to
inherit in the lot of the holy ones 1QH 11.11,12

Polycarp 12.2 (Deus ... aedificet vos ... et det vobis sortem et
partem inter sanctos suos) may show knowledge of this verse; see
1.37.
At the end of the verse, instead of πᾶσιν, D has τῶν πάντων; after
πᾶσιν, 614 pc syh** add αὐτῷ ή δόξα εις τούς αἰῶνας. αμήν. The
addition is certainly secondary; it is hard to know what τών πάντων
means. Ropes (Begs. 3.199) says that it ‘seems to be a survival of
τών αιώνων from’ the addition (which he gives in the form αὐτῷ ή
δ. εις τ. αἰῶνας τ. αἰῶνων. αμήν). This is not very convincing, but
may serve till a better guess is made.
Schille (404) writes, ‘Übrigens wird nachträglich sichtbar, warum
Paulus nicht nach Ephesus gereist ist: weil Lukas nur die Ältesten in
die Sukzession des Paulus stellen wollte.’ But Luke gave his reason
for the omission of Ephesus in 20.16. He may have been wrong;
Paul’s motive may have been different (see on the verse); but Luke
has given his opinion, the opinion he wishes to communicate to his
readers, and it is not Schille’s.
982 COMMENTARY ON ACTS

33. Paul, again looking into the past, justifies his conduct in
Ephesus (and, no doubt, Luke means to imply, elsewhere), using
language similar to that of Samuel in 1 Sam. 12.3. What he had done
he had not done for money (had he been accused of this?), or other
material advantage.
For οὐδενός, P74 A E have οὐθενός. For this spelling see on
19.27.
ιματισμός: for additions to the references in MM 304 see ND
3.69f.; the meaning of course is not in doubt.
34. Not only did Paul show no desire for other men’s property (v.
33); not only did he work for his own living; he worked also for the
benefit of his companions.
αυτοί is correctly used with the second person plural of the verb:
you yourselves know. It was unnecessary for him to inform them; we
might add, But it was useful for Luke to inform his readers. A
speaker however will often find it desirable to remind his readers of
what they may have forgotten, or be in danger of forgetting.
These hands—the speaker shows them to his hearers, ὑπηρετεῖν,
the verb, is used in the NT only in Acts; the cognate noun, υπηρέτης,
is characteristic of John, but is used at Acts 5.22, 26; 26.16; the use
of this word of John Mark at 13.5 may but need not mean that John
Mark was engaged in secular paid work in order to maintain Paul and
Barnabas. Paul now had no such assistance but worked with his own
hands; see 18.3 (cf. 1 Cor 4.12). χρεία has been used to mean need,
or the service of need, at 2.45; 4.35; 6.3; so also 28.10. Here we
should expect, after ταῖς χρείαις μου, καί ταῖς τῶν ὄντων μετ’
ἐμοῦ. But ὑπηρετεῖν is more naturally used with persons than with
needs, and with the verb coming into sight Luke takes the opportu-
nity of shortening his sentence by one word. See also Begs. 4.263,
where the sentence as Luke wrote it is taken as an example of
' “comparatio compendiaria” in the old grammars’.
In 1 Corinthians 9 Paul (using a saying of Jesus different from that
quoted in v. 35) argues that it is right that preachers of the Gospel
should be maintained by those to whom they minister, though he
himself chooses to make no use of the principle in relation to the
church at Corinth. He did accept gifts from other churches (1 Cor.
9.6-18; 2 Cor. 11.7-11; 12.16-18).
Begs. 4.263 attaches πάντα (v. 35) to this verse: These hands
ministered all things (or perhaps, always).
35. In D Speculum the verse begins not with πάντα but with
πᾶσιν, an easier, and secondary, text: I showed you all. If πάντα is
read it must be taken as an accusative of respect: In all ways, in all
respects, I showed you. Begs. 4.263 (but see on v. 34) translates, I
gave you a thorough example, or I showed you always.
Again 1 Cor. 4.12 is recalled, this time by the word κοπιᾶν, which
 51. PAUL’S SPEECH AT MILETUS. 20.17-38. 983

probably needs a somewhat stronger translation than working. When

the verb does not simply mean to be tired it still carries with it the
association of weariness—to wear oneself out with toil. The con-
struction is rightly taken by BDR § 425.6, n. 10, ‘nicht etwa “so,
nämlich arbeitend”, sondern lebhaft mit Gebäude, “seht, so muss
man arbeiten und There is a reference to Mayser 2.3.73f. for
papyrus evidence. One must work, as Paul did, not simply to secure
one’s independence but also to help the weak (who presumably are
not able to work for themselves). Blass (223) takes ‘the weak’
differently. ‘Ut ap. Paul. ἀσθενοῦντες sunt parum firmi, qui facile
offensionem capiant; quibus ut parceret, ne suspicionibus de avaritia
apostoli laborarent, licita quoque mercede ille abstinuit. Cf. 1 Cor.
9.12, et de re Act. 18.3.’ Cf. Eph. 4.28, which is very close to the
present passage, except that it is addressed specifically to ὁ
κλέπτων.
Just as in 1 Cor. 9.14 Paul clinches his argument that preachers
ought to be supported by the congregation by quoting a saying of
Jesus, so here he confirms his advice to the elders to support
themselves in the same way.
μνημονεύειν is, like ἀντιλαμβάνεσθαι, dependent on δεῖ: You
must remember. The formula of remembering (used with λόγοι) is
discussed by J. Μ. Robinson in J. Μ. Robinson and H. Koester
Trajectories through Early Christianity, Philadelphia, 1971, 96f.; see
not only Lk. 22.61; Jn 15.20 but also 1 Clement 13.1, 2; 46.7, 8; cf.
2.1. The present saying may have been drawn from a collection of
such λόγοι. There is no parallel in the canonical gospels. The
question of authenticity is left open by Knox (Hell. El. 29) and was
affirmed by J. Jeremias in the first edition of his Unknown Sayings of
Jesus (ET, 1957), 77-81. Jeremias notes parallels in Aristotle, Ethica
Nicomachea 4.1.7 (1120a); Plutarch, Maxime cum principibus Viris
3 (778c) Pseudo-Plutarch, Regum Apophthegmata: Artaxerxes 1
(173d); Seneca, Epistles 81.17; Sirach 4.31; Didache 1.5; 4.5;
Barnabas 19.9; Apostolic Constitutions 7.12.1, but surprisingly does
not mention Thucydides 2.97.4, where it is said that the Thracians
thought it better λαμβάνειν μάλλον ἤ διδόναι, therein being opposite
to the Persians (who thus must have thought it better διδόναι μάλλον
ἤ λαμβάνειν, which is virtually the Lucan saying). It is true, as J. J.
Kilgallen points out (JBL 112 (1993), 312-14), that Thucydides
expresses the matter less clearly than could be desired, but the
parallel is valid. See Haenchen (569f.). If the saying is taken, as by
Jeremias, in an exclusive sense (‘giving is blessed, not receiving’),
and even if it is taken comparatively, and if logic is strictly pressed, it
is unfair to the weak, who have no choice but to receive. But logic
must not be pressed in this way. The saying addresses a single
person, for whom giving and receiving are both possibilities; he will
do better in giving than in receiving. In later editions Jeremias’s view
984 COMMENTARY ON ACTS

was changed. In the third edition (with O. Hofius as collaborator) we

read (37), ‘Auch das Apg 20.35 zitierte Agraphon ist wahrscheinlich
ein aus der griechisch-römischen Welt stammendes Sprichwort, das
Jesus in den Mund gelegt worden ist.’ See FS Dupont 686f. See
further Conzelmann (119): ‘μᾶλλον ist nicht semitisierend exklusiv
zu fassen ... , sondern ist echt Komparativ ... Vgl den Komparativ
von 5.29!’ Conzelmann also refers to other passages mentioned
above.
D* has the meaningless μακάριος; this comes from Latin—gig
has beatus, in error for beatius, an easy mistake. syp has μακάριος ὁ
διδοὺς μάλλον ή δ λαμβάνων. P. W. van der Horst (NovT 16 (1974),
309) adds to the parallels Musonius Rufus VI (56, 10f.L; 27,
14f.H).

36. Weiser (580) argues convincingly that vv. 36-38 are Luke’s
construction. For the farewell prayer cf. 21.5. For θεὶς τά γόνατα see
7.60 and the note.

37. ικανός is a Lucan word; see e.g. 18.18. A good deal of

lamentation arose on the part of all.
ἐπιπίπτειν ἐπὶ τόν τράχηλον occurs also at Lk. 15.20. It is
frequent in the OT (e.g. Gen. 33.4; 45.14; Tobit 7.6 ); also
προπίπτειν ἐπί τ. τ. It describes an emotional embrace, accompanied
by a kiss, κατεφίλουν αυτόν.

38. ὀδυνᾶσθαι is a Lucan word (Lk., 3 times; Acts, once), used of

both physical and mental pain; ‘poët., rarius prox. att.’ (Blass 224).
Nothing grieved them so much as Paul’s own prediction that they
would never see him again; v. 25. This is a final parting. No
difference is intended between ὄψεσθε and μέλλουσιν θεωρεῖν.
For προπέμπειν see 15.3; also 21.5. They saw him off to the ship.
The article (τὸ πλοῖον) suggests but does not prove that Paul
resumed his journey in the ship that had brought him to Miletus.
Ropes (Begs. 3.201): ‘In Codex Bezae Blass (St.Kr. 1898, p. 542)
reads μελλε[τ]αι for μελλει[..]ι [Rope’s reading of D.] This recalls
videbitis faciem meam gig sah, and the omission of αυτου in D
leaves the way open for this restoration. Scrivener’s conjecture was
μελλει[σο]ι.’
 XII
PAUL RETURNS TO JERUSALEM
(21.1-22.29)

52. JOURNEY TO JERUSALEM 21.1-14

(1) When we had separated from them and set sail we made a straight course
and came to Cos, and on the next day to Rhodes, and thence to Patara. (2) We
found a ship crossing to Phoenicia, embarked, and set sail. (3) Having raised
Cyprus and left it on our left we sailed to Syria and landed at Tyre, for there the
ship was discharging its cargo. (4) We sought out the disciples and stayed
there1 seven days. They told Paul through the Spirit that he should not go to
Jerusalem. (5) But it happened that, when we had completed the [seven]2 days,
we left and set out on our journey while they all, with their wives and children,
saw us off as far as outside the city. We knelt on the beach, prayed, (6) and said
farewell to one another. We embarked on the ship, while they returned home.
(7) 3We continued the voyage from Tyre; we arrived at Ptolemais, greeted
the brothers, and stayed with them one day. (8) On the next day we left and
came to Caesarea. We entered the house of Philip the evangelist, who was
one of the Seven, and stayed with him. (9) He had four daughters, virgins,
who prophesied. (10) We stayed on for many days, and there came down
from Judaea a prophet, Agabus by name. (11) He came to us, took Paul’s
belt, bound his own feet and hands, and said, ‘Thus speaks the Holy Spirit:
So shall the Jews in Jerusalem bind the man whose belt this is, and deliver
him into the hands of the Gentiles.’ (12) When we heard this, we and the
local residents begged [Paul]4 not to go up to Jerusalem. (13) Then answered
Paul, ‘What5 are you doing, as you weep and break my heart? for I am ready
not only to be bound but even to die in Jerusalem for the name of the Lord
Jesus.’ (14) Since he would not be persuaded we fell silent, with the words,
‘The Lord’s will6 prevail.’

Bibliography
F. Bovon, as in (51).
P. Corssen, ZNW 2 (1901), 289-99.
1NEB, NJB, a week.
2Seven is not in the Greek.
3RSV, when we had finished the voyage; NEB, we made the passage; NJB, the end
of our voyage from Tyre came.
4NEB, begged and implored Paul. Paul is not in the Greek.
5NEB, Why are you trying to break my resolution?
6RSV, NEB, NJB, be done.
985
986 COMMENTARY ON ACTS

G. Dautzenberg, BiK 38 (1983), 153-8.

H. Duensing, ZAW37 (1938), 42-6.
E. E. Ellis, FS Bruce (1970), 55-67.
G. D. Kilpatrick, JTS 6 (1955), 235-8.
H. Patsch, ThZ28 (1972), 228-32.
H. Sahlin, NovT24 (1982), 188.

Commentary
The journey continues: Miletus, Cos, Patata, Cyprus, Phoenicia,
Syria, Tyre, Ptolemais, Caesarea. Compare journeys in Lucan,
Bellum Civile 8.243-8: Ephesus, Samos, Cos, Gnidus, Rhodes; and
Livy 37.16: Miletus, Myndus, Halicarnassus, Cos, Cnidus, Rhodes,
Patara. A different kind of parallel is suggested by Ehrhardt (Acts
105): In 2 Kings 2.1-12 Elisha accompanies Elijah from place to
place, Elijah repeatedly predicting his imminent departure, to the
subdued lamentations of the prophets. There is nothing, beyond the
narrative itself, to suggest that Luke had the OT story in mind, or that
he was thinking of ‘S. Paul’s temptation’ (Rackham 397) as parallel
to ‘the temptations of Moses and Elijah, and, we may add, of the
Lord himself [Num. 20.7-13; 1 Kings 19.4; Lk. 22.40-44]’.
The source of the narrative is disputed. The most natural explana-
tion, especially in view of the first person plural in vv. 1,2, 3,4,5,6,
7, 8, 11, 12, 14, is that Luke is using the first person plural itinerary
used elsewhere in the second part of his book. Conzelmann (120)
disagrees; Luke has been following up traces of Paul and on the basis
of them constructing a story. Similarly Schille (407): ‘Die Stationen
sind nicht aus einem Itinerar abgeschrieben.’ Lüdemann (241) on the
other hand writes: ‘Dem Bericht [vv. 1-16] liegt eine Quelle
zugrunde, die eine Reise des Paulus von Milet nach Jerualem
enthielt.’ He adds that there has been a good deal of redaction, shown
by the Lucan vocabulary. It is important that Lüdemann (242) takes
vv. 8b, 9 to be part of Luke’s source and not his own elaboration of it.
If this is correct if means that there is at least a second-hand, and
conceivably a first-hand, connection between the author of Acts and
one of the Seven whose story runs back to ch. 6. It is interesting also
that Agabus, previously mentioned at 11.28, now reappears, though
introduced, and described as a prophet coming from Judaea, as
though he were a new character in the story. This suggests that
information about him may have reached Luke along two different
channels, Antiochene memories having supplied the reference in ch.
11, Caesarean that in ch. 21. If this is so the present reference must be
supposed to have been worked into the Itinerary so completely that
the first person plural was introduced into it. It is perhaps more likely
that the person responsible for We and the itinerary himself encoun-
tered Agabus here. The reference to Agabus was important to Luke
 52. JOURNEY TO JERUSALEM. 21.1-14. 987

because it provided him with yet another (cf. 20.37f.; 21.4) fore-
shadowing of the trouble that was to befall Paul in Jerusalem.
This paragraph presents a very clear example of what Luke
appears to have done throughout the journey narrative which begins
at 20.1 (cf. 19.21f.). Three components may be distinguished. There
is an itinerary, which gives in simple terms a list of places touched at,
probably noting a few additional matters that bear upon the journey,
and its route, e.g. the plot of 20.3, the halt in Miletus (20.15f.), and
the encounter with disciples at Tyre (21.3f.). To this travel source
Luke added incidents which (in all probability) he learned by local
inquiry: the miracle at Troas (20.7-12), the farewell at Tyre (21.5f.),
perhaps the meeting with Philip (21.8f.), and (but see below) the
prophecy of Agabus (21.10-14). The third component consists of
theological comment on the meaning of the journey. By far the most
important such comment is the speech at Miletus (20.18-35), which
shows Paul’s pastoral care for the whole church and foretells both his
approaching suffering and the perils of heresy and schism which lie
ahead. In addition to the speech there are repeated references to
Paul’s suffering and his determination to face and accept it. Here the
most important example is the prophecy of Agabus, with its pro-
phetic symbolism (21.10-14), which (if—see above—we may at the
same time regard it as a historical reminiscence) is important in that
it shows how event and interpretation originated and developed
together.
Haenchen (577f.) rightly rejects the objection to the historicity of
the narrative that Paul ignores the prophesying of the Christians in
Tyre, that we are told that the daughters of Philip prophesied but not
what they said, and that Agabus could not have bound his own hands
and feet. The story is sufficiently connected and makes sense; Luke is
not writing like a professor of history; this he neither was able nor
wished to do.

1. ἐγένετο with the infinitive: for Luke’s frequent use (in imitation
of the LXX?) of this construction see on 4.5.
ἀναχθήναι: see 20.13. The word recurs in v. 2.
αποσπάν is used at Mt. 26.51 for drawing a sword from its sheath,
but Luke uses it exclusively of persons. At Acts 20.30 it is used in the
active of those who draw away disciples; at Lk. 22.41 the passive is
used intransitively (ἀπεσπάσθη ἀπ’ αυτῶν); so also here. It suggests
a separation made with difficulty; they were loth to part. But ‘we tore
ourselves away’ might be too strong. For ὡς ... ἀποσπασθέντες, D*
sa have καί ἐπιβάντες ἀνήχθημεν ἀποσπασθέντων δέ ἡμῶν. The
sense is unchanged; the copyist, or editor, feels free to change the
wording.
εὐθυδρομήσαντες: see 16.11. The straight course to Cos implies a
north east wind, which was usual.
988 COMMENTARY ON ACTS

εἰς τὴν Κῶ. For the declension of Κῶς (cf. Άπολλῶς at 19.1) see
Μ. 2.121. After Samos, proceeding southwards, Cos was the next
sizeable island. For Cos at approximately the time Paul was there
(presumably only for an overnight stay) see Tacitus, Annals 12.61.
Claudius made it immunise it had fairly consistently taken the Roman
side in earlier wars. It was the home of Hippocrates and the site of the
medical school founded by him. For Jews on Cos see NS 3.69;
Trebilco 13, 134f. See also Strabo 14.2; Horace, Odes 4.13.13.
τῇ δὲ ἑξῆς, a Lucan term: Lk. 7.11; 9.37; Acts 21.1; 25.17; 27.18.
No other NT writer uses it, but it is fairly common elsewhere. Here
as in most places ημέρα must be supplied (at Lk. 7.11, χρόνῳ). Luke
likes to show variety in such words; cf. 20.15.
εἰς τὴν 'Ρόδον. The ship was following the chain of inshore
Aegean islands, most of which had satisfactory harbours. The same
name served for the island and for its chief city, a civitas libera
through most of the imperial period. ‘From Rhodes there is virtually
no significant evidence’ of Jewish residents (NS 3.69). See however
1 Macc. 15.23.
Patara (Herodotus 1.182.2; Horace, Odes 3.4.64—an oracle of
Apollo) was on the mainland, in Lycia, and the residence of the
legatus Augusti. The name is a neuter plural. It seems to have been a
likelier place for Paul to change ships (v. 2) than Myra, which is
introduced here by the Western text. Πάταρα stands alone in B E
Ψ <a> vg sy, Πατέρα in P74 A C. καὶ Μύρα is added by P41vid D (gig
vgmss sa). Myra may have been introduced from 27.5, but Clark (377)
defends the reading and so does Ropes (Begs. 3.201; followed by
Hanson 206, 208); both suggest that the second name may have been
omitted by homoeoteleuton. ‘For deliberate change, either by omis-
sion or by addition, no sufficient motive is easy to assign’ (Ropes,
loc. cit.). Ropes however refers to Paul’s presence in Myra accord-
ing to the Acts of Paul and Thecla (§ 40; L.-B. 1.266); it is perhaps
somewhat more probable than Ropes thinks that there was a tradition
that connected Paul with Myra and this may have affected the text.
See Metzger (482).

2. εὑρόντες πλοῖον. A change of ship was necessary; finding in

this case (not always) implies search. Stählin (273) says that Paul
changed ships ‘um rascher voranzukommen’. It may be that the first
ship was sailing no further. Another suggestion is that Paul was
evading a plot; cf. 20.3.
διαπερῶν, making the crossing. The present participle is used to
express the future. For other examples in Acts see 14.21f.; 15.27;
18.23; 21.3, 16; 26.17. See the notes; also Zerwick (§ 283); BDR §
339.2. But the idiom is as natural in Greek as in English. The new
ship must have been a bigger one, no longer hugging the coast and
the islands.
 52. JOURNEY TO JERUSALEM. 21.1-14. 989

εἰς Φοινίκην. Phoenicia was the strip of coast between Carmel in

the south and Nahr-el-Kelb in the north. The Libanus and Anti-
Libanus provided a curtain which shut off the interior and meant that
the district looked primarily towards the Mediterranean. The chief
towns lay on the coast, some of them (e.g. Tyre, v. 3) situated on
islands, which gave them great defensive strength.
ἐπιβάντες. Cf. 27.2; the word is synonymous with ἐμβαίνειν. See
e.g. Thucydides 1.111.2.
ἀνήχθημεν, as in v. 1; Luke’s usual word for putting to sea.

3. ἀναφάναντες (aorist participle active), P74 B* 614 1704 1739

pc d; ἀναφανέντες (aorist participle passive), A B2 C Ε Ψ <a> The
readings of B and its correctors are uncertain; see Ropes (Begs.
3.201). The active participle will have the sense of cause to become
visible, by coming near; cf. the English idiomatic use of raise, ‘to
come in sight of’ (SOD, s.v. III 7a). Lucian, Dialogi Marini 10.1 (τήν
νήσον ... ταύτην ... ἀνάφηνον) is not as close a parallel as is
sometimes suggested. The passive participle is often said to be mean-
ingless (Ropes: ‘yields no good natural sense’), but BDR § 159.4, n. 4
quote Theophanes, Chronicle 1.721, ἀναφανέντων αυτών τήν γῆν,
rendering it ‘als sie in Sicht des Landes kamen’. At § 309.1, n. 2 they
translate ἀναφάναντες τήν Κύπρον, ‘ “wir liessen Zypern uns sicht-
bar werden” ... nämlich durch Herankommen’. Referring again to
the passive they add, ‘ebenfalls wohl Schifferausdruck wie
ἀποκρύπτειν (abscondere) für das Gegenteil’. Cf. Delebecque (101):
ἀναφαίνω (ici à l’aoriste 1 actif) est pris dans un sens apparemment
nautique: “voir apparaître” (une terre, quand on est en mer).’
Cyprus: see 11.19. They left it on their left, passing to the right,
that is, to the south of it, following a direct line from Patara (Myra) to
Tyre. Wettstein (2.602), for the expression, compares Lucian, Nav-
igium 9, τήν Κρήτην δεξιάν λαβόντας ... They continued sailing in
this direction (ἐπλέομεν, imperfect) until they arrived (κατήλθομεν,
aorist) at their destination. 'Impf. ἐπλέομεν cursum, aor.
κατήλθομεν finem denotat’ (Blass 224). Similarly Μ. 3.66; BDR §
327.1, n. 1.
Syria: see 15.23; the word is used here in the sense of Phoenicia.
κατήλθομεν (P74 A Β E 33 326 1175 pc lat sy) is almost certainly
correct; κατήχθημεν (C Ψ <a> ) probably came in by analogy with
ἀνήχθημεν (v. 2).
Tyre was the most notable city on the Phoenician coast. It had
been destroyed by Alexander the Great in 332 BC (e.g. Arrian,
Anabasis 2.17-27) but speedily regained prosperity. There were
Jews here (Tcherikover 289; NS 3.14f.). For the earlier, biblical story
of Tyre see 2 Sam. 5.11; 1 Kings 5; Isa. 23; Ezek. 26-28.
For the duration of the voyage there are various estimates. Roloff
(309), without supplying evidence, says, with a favourable wind,
990 COMMENTARY ON ACTS

three days; Xenophon of Ephesus 1.14.6 three days (for Rhodes to

Tyre); Chrysostom, Homily 43.1, five days (see Bruce 1.385).
ἦν ἀποφορτιζόμενον; as in v. 2, a present participle where a
future might have been expected. See Zerwick § 283; BDR § 339.2
β. Again, the idiom is also English: That is where the boat was
unloading ... The past tense makes the choice of participle the more
natural.
γόμος, cargo occurs elsewhere in the NT only at Rev. 18.12f., an
oracle based on Ezek. 27, the dirge on Tyre. But this is significant
only of the fact that Tyre was a great trading city. Cf. Demosthenes
323.4 (883), τόν γόμον ... τής νεώς.
On this journey Antioch is not included. This may have been
simply through failure to catch a convenient boat; it may have been
that the church of Antioch was no longer, after the troubles of Gal.
2.11-21, open to Paul. The question is worth asking, though there is
no means of answering it.

4. ἀνευρόντες. Cf. v. 2 (εὑρόντες); the compound implies search.

Presumably the travellers had reason to think that they would find a
group of disciples (τούς μαθητάς, with article), and sought them out.
Search would be necessary, ‘erat enim urbs magna, Christiani pauci’
(Blass 225). Roloff (309) thinks that this group of disciples ‘auf das
Wirken der Hellenisten zurückging (vgl. 11.19; 15.3)’. This is
possible; more cannot be said.
ἐπεμειναμεν, constative aorist; the seven days are regarded as a
unit. For αυτού, there (cf. 18.19), A L have αὐτοῖς (cf. d, apud eos),
we stayed with them. For the seven-day stay cf. 20.6. Why did Paul
stay here for seven days when he was hurrying (20.16) to get to
Jerusalem? Was it because there was no ship available? Preuschen
thinks this hardly possible, but it is not clear why. The correct answer
is given in Begs. 4.265. τό πλοῖον (with article) in v. 6 shows that
the travellers continued in the same ship; the seven days were
therefore spent in unloading and loading the ship.
οἵτινες (equivalent to oἵ); it is characteristic of Acts to continue a
narrative with a relative. διὰ τού πνεύματος: presumably, showing
the phenomena of inspiration. Luke does not express himself clearly.
His words taken strictly would mean either that Paul was deliberately
disobedient to the will of God or that the Spirit was mistaken in the
guidance given. It is unthinkable that Luke intended either of these. It
is probable that what he meant but failed adequately to express was
something like what is written in vv. 10-14. The Spirit acting
through prophets foretold that the journey to Jerusalem would bring
Paul suffering, and his friends acting under the influence not of the
Spirit but of human concern sought to dissuade him from going there.
So, more or less, Calvin (2.193): ‘There are different gifts of the
Spirit, so that it is no wonder that those who are strong in the gift of
 52. JOURNEY TO JERUSALEM. 21.1-14. 991

prophecy are sometimes lacking in judgement or courage’; and

Bengel (470): ‘Spiritus significabat, Paulo imminere vincula: inde
rogabant discipuli eum, ne iret.’ Weiser (589f.) summarizes at length
Bovon’s explanation of the contrast with 20.23. ‘Lukas folge bei der
Gestaltung der Abschiedsszenen (20.36-21.16) dem Muster grie-
chischer Darstellungen des Abschieds berühmter Helden von ihren
Familien oder Freunden.’ See Bovon in Kremer (Actes 339-58).
5. The language of the verse is Lucan. For ἐγένετο and the
infinitive see v. 1; for προπέμπειν, 15.3; 20.38; for τιθέναι τὰ
γόνατα, 7.60; 9.40; 20.36.
ἐξαρτίσαι of time, where πληρῶσαι might have been expected,
‘sonst nicht belegt’ (Preuschen 125). LS 587 gives the meaning, but
with no other examples (there are examples of finishing buildings
and books). BA 553 quote Hippocrates, Epidemiae 2.180,
ἀπαρτίζειν τήν οκτάμηνον; see 2.7, ὀκταμήνῳ. τὰς ἡμέρας are the
seven days of v. 4.
αἰγιαλός ‘describes correctly the smooth beach at Tyre, as
opposed to ακτή, used of a rocky shore’ (Hemer 125). Stählin (273)
is more precise: the beach to the south of the mole built by Alexander
the Great to connect the islano of Tyre with the mainland. Hanson
(208): ‘... the beach at Tyre can still be identified. Is not this the
vivid touch of an eye-witness?’ Possibly; but not necessarily.
6. This verse continues in construction from the previous one—
προσευξάμενοι ἀπησπασάμεθα. For the verb ἀπασπάζεσθαι cf.
the simple form in 20.1; the combination with ἀπό makes it more
suitable for farewells. The leave-taking was mutual—άλλήλους. Cf.
Plutarch, De Genio Socratis 33 (598A), ἀσπασάμενοι ἀλλήλους.
The first person plural of ἀπησπασάμεθα is now divided; one group
embarks, the other goes home. The textual variations convey no
difference in meaning.
ἀνέβημεν P74 A C 36 453 614 1175 2495 pc
ἐνέβημεν Β Ε ( Ψ ) 945 1739 1891 pc
ἐπέβημεν <a>
Bruce (1.386) tends to favour ‘the less usual ἀνέβημεν’. But in Acts
ἐνεβ. occurs here only and thus may claim to be in Acts the less
usual.
The travelling party embarked εἰς τὸ πλοῖον, the ship, that is, the
ship in which they had arrived. See v. 4; the article is anaphoric.
εἰς τά ἴδια, to their own places, their own homes. Cf. Jn 16.32;
19.27.
7. τὸν πλοῦν διανύσαντες, having completed the voyage. For
διανύειν πλοῦν see Bruce (1.386); Conzelmann (121); and Xeno-
phon of Ephesus 1.11.2, 5; 1.14.6. There is no ground for the
992 COMMENTARY ON ACTS

inference that the voyage was completed in a different ship. It has

however been suggested that the sea voyage ended here (though it is
difficult to make this agree with v. 7). Page (220) does not take ἀπό
Τύρου with the participle but with the following verb, and renders,
We, having (thereby) completed our voyage, came from Tyre to
Ptolemais. See further on v. 8.
κατηντήσαμεν (an ‘Acts’ word—16.1; 18.19, 24; 20.15; 25.13;
26.7; 27.12; 28.13; so is κατέβημεν, read by P74 A Ε) εις
Πτολεμαΐδα where the sea voyage, as some think, may have ended.
Ptolemais was ‘einst Haupthafen für Palästina’ (Preuschen 125).
He cites Josephus, War 1.290, 394; Ant. 14.452; 15.199; 18.155.
These do not seem to prove more than that Pt. was an important
port. It was an ancient town, formerly Acco (Judges 1.13; 4 QpIsa
2.23), renamed after himself by Ptolemy V (285-246 BC). It was an
important centre of trade and had a Jewish population (see e.g.
Abodah Zarah 3.4—Rabban Gamaliel in the Bath of Aphrodite). It
was about 31 miles down the coast from Tyre. Here too were
Christians: ἀσπασάμενοι (the simplex; cf. v. 6) τούς αδελφούς
(for this term for Christians see 1.15 and many other passages).
The fact that ‘we’ stayed with them only one day (v. 8), com-
pared with the seven days at Tyre (v. 4) may mean that the
company were proceeding to Caesarea by ship and that the ship was
staying only one night in Ptolemais. Bauemfeind (241) however
suggests that nothing of note happened in Ptolemais and Caesarea
because these churches were founded by dependants of the Seven
(Acts 6).

8. τῇ δέ ἐπαύριον corresponds to the ἡμέραν μίαν of v. 7.

ἐξελθόντες: out of the city to the harbour or the ship; or, if the
travellers went by road, out of the city.
For ἤλθομεν (continuing the ‘we’ passage) <a> has οι περί τόν
Παύλον ἦλθον, Paul and his party went. This text however continues
with the first person plural ἐμείναμεν. D is missing at this point; for
the means by which the text of d in vv. 7-10 has been reconstructed
see in addition to Ropes (Begs. 3.201) G. D. Kilpatrick in JTS 6
(1955), 235-8. d appears to have read venimus, with the Greek
ἤλθομεν. One can only guess that the reading of arose because an
editor thought it a long time since Paul was last mentioned and that it
would be well to make clear to the reader (lest he should have
forgotten) that ‘we’ meant ‘Paul and his companions’. Cf. 13.13.
From Ptolemais to Caesarea was about 30 miles, ‘a suitable day’s
journey whether this stage was taken by sea or coastal road’ (Hemer
125f.) If however the party travelled by land and on foot one would
think two days a more likely time.
For Caesarea see 10.1 and the note.
Philip was last heard of at Caesarea (8.40). If the ‘we’ material is
 52. JOURNEY TO JERUSALEM. 21.1-14. 993

to be thought of as a source it can hardly be insignificant that its

author lodged with a Christian whose memories extended as far back
as Acts 6, and possibly earlier, since the persons appointed in Acts 6
can hardly have been at that time among the most recent converts. It
is unlikely (see I.53f.) that the material in chs. 1-5 depends on
written sources, equally unlikely that it is pure fiction. Luke must
have made inquiries from those who, he thought, could give him
trustworthy information. It seems not impossible that Philip was one
such person. See 1.51 and Introduction, p. xxiv.
Philip is described as the Evangelist, one of the Seven. The word
does not occur elsewhere in Acts (in the NT only at Eph. 4.11 ; 2 Tim.
4.5), but cf. 8.12,35,40, where Philip evangelizes (εὐαγγελίζεσθαι).
The article distinguishes this Philip from others of the same name;
contrast 10.6 (in this chapter it is the other Simon—Peter—who is
distinguished) and 21.16 (see BDR § 268.1, though this passage is
not mentioned there). It is however the title Evangelist itself that
distinguishes this Philip from Philip the Apostle (1.13). The word
was little used (see Hamack, The Mission and Expansion of Christi-
anity in the First Three Centuries, London21908, 1.321, 338, 348f.).
Eusebius seems to have regarded it as a term applicable to those who
assisted and followed the apostles in the work of mission preaching
and founding churches (see especially HE 3.37.2, τούς ...
προκαταβληθέντας ὑπό τών αποστόλων θεμελίους ἐπωκοδομοῦν
... ἔργον ἐπετέλουν ευαγγελιστών; also 2.3.1; 3.31.2-5; 3.37.4;
3.39.9; 5.10.2; 5.17.3). Bultmann is probably right in saying that the
word come into use as the term απόστολος came to be confined to
the Twelve, and in adding, ‘Als gebräuchlicher Titel hat sich
ευαγγελιστής nicht durchsetzen und halten können, weil allmählich
die Gemeindebeamten das Amt der Wortverkündigung übernahmen’
(Theologie 458). Knowling (444) is probably right with ‘a work
rather than an order’. Calvin (2.194): ‘In my opinion evangelists
were half-way between apostles and teachers.’ This is not really
helpful. Roloff (310): ‘Eher umschreibt sie [die Bezeichnung εὐαγγ.]
hier wie in 2 Tim 4.5 die Funktion des Gemeindeleiters’—a function
which in fact we never see Philip exercising. For possible non-
christian use of the word see MM 259; and cf. Clement of Alexan-
dria, Stromateis 3.52-54.
On the Seven see 6.1-6 and the notes. As Evangelist distinguishes
Philip from the Apostles, so ‘one of the Seven’ distinguishes him
from the Twelve. But oἱ ἑπτά seems even less than εὐαγγ. to have
remained in use as a technical term. The two Philips are often
confused; see Bruce (2.400) for the views of Zahn, Hamack, and
Chapman. Hengel (Between Jesus and Paul 14) thinks that Philip
was originally one of the Twelve and went over to the Seven. This
seems to imply more knowledge of the Seven, and of their difference
from the Twelve, than we possess.
994 COMMENTARY ON ACTS

9. Philip had four virgin daughters who prophesied. Not surpris-

ingly, though writers such as Eusebius (see most of the passages
cited on v. 8) refer to this statement, there is little confirmatory
evidence for it. At HE 3.39.9, however, Eusebius recalls information
which Papias claimed to have derived from the daughters of Philip.
For prophesying daughters cf. the three daughters of Job in Testa-
ment of Job 48-50.
That women should prophesy in Christian gatherings is confirmed
by 1 Cor. 11.5 and by the passage from Joel quoted in Acts 2.17; it
appears to be denied by 1 Cor. 14.34; 1 Tim. 2.11f. It is hardly to be
denied that the practice varied from part to part of the church (though
the apparent contradiction within 1 Corinthians is a special problem;
see 1 Corinthians 250f., 330-3), or that the tendency was for
attitudes to harden in the direction of the exclusion of women from
public participation in Christian meetings. Luke, however, towards
the end of the century, must have held that it was legitimate and
edifying for women to prophesy, even though he may have wished to
see this activity restrained by certain conditions (see below). What
Luke understood by prophecy is, at least in part, brought out in the
next paragraph, in which Agabus (that he makes the prophecy
suggests that the daughters were in the Itinerary—Weiser 591)
predicts the sufferings that are to happen to Paul (cf. 11.27-30). In
the Pauline epistles prophecy is evidently more than this, and
includes what may be called preaching; it is characteristic of Luke’s
turn of mind that the striking and verifiable predictive element in
prophecy should be prominent.
The four prophesying daughters (use of the participle
προφητεύουσαι suggests that for Luke prophecy was a function
rather than an office) were virgins. It is hard to tell whether Luke
relates this as a simple fact or sees a connection between their prophe-
sying and their virginity. If they had not been virgins would they have
prophesied? If they had been married would they have been allowed to
prophesy? It is noteworthy that in 1 Cor. 14.35, where women
(γυναίκες) are bidden to keep silence in the assemblies, they are told
that if they wish to learn anything they must ask their own husbands at
home. It is assumed that they are married; is it implied that different
regulations would apply to the unmarried? There is nothing else in
Acts that bears directly on this subject, but it may be recalled that
Priscilla, who taught Apollos (18.26), was married, and that the
prophetess Anna (Lk. 2.36) was a widow who had lived with her
husband seven years from her virginity. Luke provides no ground for
Calvin’s observation (2.195), 'One may well believe that they prophe-
sied at home, or in a private place, outside the public meeting.’ Bengel
(471) is more interesting: ‘Philippus evangelista: filiae prophetantes.
propheta major est, quam evangelista. Eph. 4.11.’
From Jerome, Epistles 108.8 we learn that his correspondent Paula
 52. JOURNEY TO JERUSALEM. 21.1-14. 995

saw in Caesarea the house of Cornelius, now turned into a church,

and the house of Philip, with the chamber of the daughters.
10. The genitive absolute participle ἐπιμενόντων must in the
context be supplemented by ἡμῶν (see v. 8, ἐμειναμεν, and v. 11,
πρὸς ἡμάς). Bultmann (Exegetica 420; cf. 20.7 and note) thinks that
the first person plural has been imported into this narrative (as into
that of the supper at Troas) from the Itinerary. Note however the
juxtaposition in v. 12 of ἡμεῖς and oἱ ἐντόπιοι. E <a> gig syhmg add
ἡμῶν, 1175 αυτών. But the omission of the pronoun in the
genitive absolute is classical: Zerwick (§ 50); BDR § 423, n. 3.
τις ... προφήτης ... Άγαβος. See 11.28. There is no indication
how Agabus has been occupied in the meantime, or whether his
prophesying constituted an office or was an occasional act, occurring
only when prompted by the Spirit. Marshall (340) observes that
Agabus is mentioned as if he were a new character, not previously
mentioned, and infers that the reference to him may have been taken
over from the ‘We’ source in which he had not previously appeared;
similarly Roloff (310), Bauemfeind (241), Lüdemann (242) (L.
argues that vv. 10-14 go back to tradition); otherwise Schille (408).
That Agabus is said to have come from Judaea suggests to Hemer
(126) the Jewish use of Judaea, since the district includes Jerusalem
but not Caesarea.
According to Ropes (Begs. 3.203) the text of d in this verse
(following upon IIII virgines prophetantes) opens with the words et
mansimus aput (sic) earn. It is hardly possible to make sense of this
except on the supposition that earn is an error for eum, and that the
Latin is the result of misreading ἐπιμενόντων as ἐπεμένομεν (μετ’
αὐτοῦ). Kilpatrick (see on v. 8), however, thinks this part of the
reconstruction of the text of d doubtful.
11. Instead of εαυτού the mass of late MSS have τε αὐτοῦ, which
gives a different picture, according to which Agabus bound (not his
own but) Paul’s hands and feet. ἑαυτοῦ is certainly correct; Agabus,
binding himself hands and feet, performs an acted parable, a
prophetic sign comparable with those of the OT prophets (for these
Roloff cites 1 Kings 11.29ff.; Isa. 8.1-4; 20.1ff.; Jer. 19. 1ff.; 27.2ff.;
Ezek. 4-5). His prophecy, so far as Acts records it, is of a predictive
kind. Its fulfilment appears in the following chapters. It is not
however strictly accurate. The Jews do not themselves bind Paul, nor
do they hand him over to the Gentiles. The Jews attack Paul by mob
violence; the Romans rescue him and then bind him, subsequently
not setting him free but treating him with some respect when they
learn that he is a Roman citizen. This lack of complete correspond-
ence between prophecy and fulfilment might be taken to argue for the
authenticity of the prophecy, but it would be unwise to take it in this
way. Luke is not a writer who notices or is concerned about neat
996 COMMENTARY ON ACTS

correspondences (witness his three accounts of Paul’s conversion)

and παραδώσουσιν εἰς χεῖρας εθνῶν may well have been formu-
lated so as to recall the arrest and trial of Jesus. So for example
Stählin (275); Pesch (2.214); ‘Das Leiden Jesu wird als vorbildliches
Martyrium gedeutet’ (Schmithals 193). See however Begs. 4.268,
where it is, rightly, pointed out that the Jews must have brought some
accusation against Paul or he would not have been kept prisoner so
long, or the case represented as ‘the Jews v. Paul’.
For καἰ ἐλθών, D* has ἀνελθὼν δέ; for ἐν, P74 D have εἰς. For the
latter variant is can be argued that εις is not the ‘correct’ preposition;
but in Acts it is often used where ἐν would be more suitable.
12. ‘We’, Paul’s travelling companions, continue to be the main
subordinate actors, but are here joined by οἱ ἐντόπιοι, the local
(Christian) residents. The word is sometimes said to be unusual, but
there is plenty of evidence for its use in MM 218, BA 544, LS 577.
More is added in ND 4.263f. It is used in opposition to ξένοι in IG
5.2.491 lines 77, 8 (Megalopolis, ii/iii AD).
παρακαλεῖν is used in several senses in Acts (see e.g. 2.40); here,
clearly, to beg, to entreat. It is followed (cf. 3.12) by τού and the
infinitive, where a simple accusative and infinitive would have
sufficed. Pleonastic τού is characteristic of the Lucan writings (Lk.,
20 times; Acts, 17 times), but Zerwick (§ 386) notes: ‘Ubi subest
idea impediendi, abstinendi etc. etiam classice adhibetur τού cum
infinitivo (genitivus separationis!), e.g. Lc 4.32; 24.16; A 10.47; R
15.22; 2 C 1.8.’ Cf. BDR § 409.5, n. 7.
Betz (118 n. 4) notes with examples how the way of the man of
God is beset with friends who warn him of trouble to come. In this
context perhaps the best example is that of Peter, Mt. 16.22.
For τοῦ μή ἀναβαίνειν, D gig have τόν Παύλον τού μη
ἐπιβαίνειν. There is no difference in meaning.
13. τότε ἀπεκρίθη is more forceful and solemn than the variants;
Then answered Paul... It was a memorable saying, expressing with
special clarity Paul’s devotion to the Lord and to his service. The text
of NA26 is τότε ἀπεκρίθη ό Παύλος, B(*om. ό) C(* + δέ) 36 pc bo.
The variants are
ἀπεκρίθη δέ (τε <a>) ὁ Π., Ψ <a>syh
ἀπεκρίθη δέ ό Π. καί εἶπεν, (373) 945 1739 1891 pc
εἶπεν δέ πρὀς ἡμάς ό Π., D (gig)
τότε ἀπεκρίθη ο (om. Ρ74) Π. καί εἶπεν, P74 A Ε (33) pc vg
syp sa boms
It has been suggested with plausibility that the origin of this
confusion was the mistake of attaching τότε to the end of the
preceding sentence. Fortunately the meaning is unchanged which-
ever reading is adopted.
 52. JOURNEY TO JERUSALEM. 21.1-14. 997

The participles κλαίοντες and συνθρύπτοντες appear to modify

the verb ποιείτε: What are you doing, as you weep...? συνθρυπτ. is
a rare word—to break in pieces, to reduce to powder. The general
but not the precise sense of the transferred use is clear. It must mean
breaking my heart, but might mean with sorrow or by weakening my
resolve. The latter suits the context better. D* gig p have
θορυβοῦντες. The question is rhetorical, and the implied sense,
grounded in the γάρ that follows, is either, You are weeping and
persuading in vain, for I am ready ...; or Cease to weep and
persuade, for I am ready ... For the expression cf. Plato, Republic
495de: οὕτω καἰ τὰς ψυχὰς ξυγκεκλασμένοι τε καἰ
ἀποτεθρυμμένοι διά τὰς βαναυσίας τυγχάνουσιν.
After δεθῆναι D (syp) add βούλομαι, thinking perhaps that
ἐτοίμως έχω was a long was off. For the infinitives with ἐτοίμως έχω
see BDR § 393.3, n. 4. Once more, εις is used for ἐν. For Paul’s
determination cf. Josephus, Ant. 13.6, ὁ Ίωνάθης φήσας ἐτοίμως
ἐχειν ἀποθνήσκειν υπέρ αυτών.
Paul is ready to die υπέρ τού ονόματος τού κυρίου Ιησού. Cf.
9.16; also 5.41; and for the use of όνομα in Acts see L182f. The
usage may have originated in the Hebrew (Aramaic ‘Es
fragt sich aber, ob hier nicht schon das Problem des “nomen
Christianum” anklingt’ (Haenchen 577).

14. Paul is not to be persuaded. It is clear to him that his apostolic

duty calls him to Jerusalem (why? not apparently to preach; to
convey the proceeds of his collection? to secure the unity of the
church? to act as well as speak his testimony to the Crucified?) and
nothing will deflect him from it. This is part of Luke’s picture of the
heroic missionary.
ήσυχάσαμεν εἰπόντες. For discussion of the time relation of the
aorist participle see M. 1.133f.; BDR § 420.3, n. 4; cf. 7.35; 22.24.
There is no need to suppose that we have here an aorist participle
describing subsequent action; at most the action is coincident. Μ.
1.134 n. 1 translates, ‘we ceased, with the words ...’. But even this is
not strong enough. The participle is strictly antecedent: ‘When we
had said ..., we fell silent’.
τού κυρίου τό θέλημα γινέσθω—according to Begs. 4.269, ‘The
Lord’s will prevail;’ ...be done would be γενέσθω. Cf. Lk. 11.2 (si
v.1.), γενηθήτω τό θέλημά σου; 22.42, πλὴν μὴ τό θέλημά μου
ἀλλὰ τὸ σὸν γινέσθω. Since Luke apparently did not have these
words in the version of the Lord’s Prayer familiar to him, it is
probable that, as at v. 11, he was alluding to, and drawing a parallel
with, the story of the Lord’s passion. His words could be interpreted
in a way very unfavourable to the speakers: it was only after they had
failed to secure their own will that they were prepared to accept the
Lord’s. But Luke does not mean this; rather, We are prepared at
998 COMMENTARY ON ACTS

length regretfully to recognize that it is the Lord’s will that Paul

should suffer; unwelcome as this is, may it nevertheless be done.
Schneider (2.305): ‘Das Schlusswort ist nicht Ausdruck von Resig-
nation, sondern bejaht den erkannten Willen Gottes.’ Similarly
Schmithals (193); Conzelmann (121) (comparing also Martyrdom of
Polycarp 7.1).
After εἰπόντες, D adds, πρὸς ἀλλήλους.
 53. PAUL AND THE CHURCH OF JERUSALEM 21.15-26

(15) After these days we1 packed up and set out on the journey up to
Jerusalem. (16) Some of the2 disciples went with us from Caesarea, bringing3
us to the man we were to lodge with, Mnason, a Cypriote and an early
disciple. (17) When we reached Jerusalem the brothers welcomed us
gladly.
(18) On the next day Paul went in with us to James, and all the elders were
present. (19) When he had greeted them he related in detail the things that
God had done among the Gentiles through his ministry. (20) When they
heard it they glorified God and said to him, ‘You see, brother, how many tens
of thousands there are of those who have believed among the Jews, and they
are all zealous for the Law. (21) They have been informed about you that you
teach defection from Moses, telling all the Jews who live among the Gentiles
that they should not circumcise their children or walk in accordance with the
customs. (22) What then? 4They will certainly hear that you have come. (23)
So do this that we tell you. We have four men who have a vow on them. (24)
Take these men with you, be purified with them, and pay their expenses that
they may shave their heads, and then all will know that there is nothing in the
things they have been told about you, but that you yourself also conform,
observing the Law. (25) But concerning those Gentiles who have believed
we ourselves5 wrote an injunction with the decision that they should be on
their guard against food sacrificed to idols, blood, strangled meat, and
fornication.’ (26) Then Paul on the next day took the men with him, was
purified with them, and entered the Temple, notifying the fulfilment of the
days of purification until the offering was offered for each of them.

Bibliography
K. Berger, NTS 23 (1977), 180-204.
Μ. Black, BJRL 23 (1939), 201-14.
F. F. Bruce, BJRL 67 (1984-85), 641-61.
C. Burchard, as in (43).
H. J. Cadbury, FS J. R. Harris, 51-3.
H. von Campenhausen, ZKG 63 (1950-51), 133-44.
R. P. Casey, HThR 16 (1923), 392-4.

1RSV made ready; NJB, made our preparations.

2RSV, NEB, NJB, disciples from Caesarea.
3NEB, a certain Mnason.
4NJB, a crowd is sure to gather, for they.
5RSV, sent a letter with our judgement; NEB, sent them our decision; NJB, have
written giving them our decision.
999
1000 COMMENTARY ON ACTS

H. Duensing, as in (52).
G. Jasper, Judaica 19 (1963), 147-62.
J. Jervell, Luke and the People of God (1972), 185-207.
G. Kittel, ZNW 30 (1931), 145-57.
J. H. Ropes, HThR 16 (1923), 162-8.

Commentary
This is a passage of great importance for the understanding and
evaluation of Acts, and of equally great difficulty. The opening
verses retain the first person plural of the travel narrative:
ἀνεβαίνομεν (v. 15); ἡμῖν, ξενισθώμεν (ν. 16); ἡμῶν, ημάς (ν. 17).
This may well have brought the We- narrative, or one section of it, to
a close. The warm welcome given to ‘us’ is a suitable conclusion,
and after the σὐν ἡμῖν of ν. 18 the narrative concentrates on Paul and
is written in the third person (vv. 19, 26). In vv. 15-17 there are (see
the notes) notable textual obscurities, but there is no reason for
thinking the reference to Mnason fictitious, and Paul did arrive in
Jerusalem, whether to a warm welcome or not. The story that follows
is also not without difficulty, but it is presented as the occasion that
leads to all the events in the rest of the book, up to and including
Paul’s arrival in Rome as a prisoner. It is however important to
observe that from 21.27a onwards there is no further reference to the
four who needed purification for the due fulfilment on their vows;
that is to say, the whole action against Paul, from 21.27b to the end of
the book, is totally independent of the proposal made by James in vv.
23, 24. Paul is accused of having brought Greeks into the Temple;
and it is certain that the four who had taken (Nazirite) vows were not
Greeks but Jews. This observation is important because of the serious
doubts that must be raised (see on vv. 21, 24, 25) regarding Paul’s
alleged action in proof of his alleged obedience to the Law. These
doubts are, as they must be, based on theological considerations; but
Paul was not only a radical theologian but one who allowed his
actions to be dictated by theological motivation. It is true that
probably a majority of students take the story to be essentially
historical: Paul did act in the way described. It is true also that Paul
accepted synagogue discipline to the extent of enduring five fearful
beatings (2 Cor. 11.24). The possibility cannot be excluded that in
the interests of peace he allowed himself to be persuaded, perhaps
against his better judgement, to take part in the legal requirements
laid upon those who had taken vows. If so, the outcome must have
speedily shown him the error of his decision. Those who wished to
misinterpret his action (and, perhaps we should add, were able to
perceive the truth behind the sham of James’s proposal) were quickly
able to do so and to find a ground for accusation.
There is a further possibility, suggested in outline by Haenchen
 53. PAUL AND THE CHURCH OF JERUSALEM. 21.15-26. 1001

(586-8), who brings the act of purification into relation with the
surprising silence of Luke’s narrative (apart from 24.17) regarding
the primary purpose of Paul’s visit to Jerusalem, which was (Rom.
15.25) to deliver the collection he had made among the Gentile
churches for the poor saints in Jerusalem. At what point did Paul
hand over the money to the leaders of the Jerusalem church? We do
not know, for Luke tells us nothing of the matter. If we are to go
further we can only guess. Certainly he would wish to have the
money out of his responsibility as soon as possible; did he deliver it
as early as v. 17, when ἀσμένως ἀπεδέξαντο ἡμᾶς οἱ αδελφοί? A
better guess might be vv. 18, 19, partly because this is represented as
something like an official reception (by James and πάντες οἱ
πρεσβύτεροι). There might even be a hint in the use of the word
διακονία: Paul recounts what God has done among the Gentiles διά
τῆς διακονίας αυτού—the word that he several times (Rom. 15.31;
2 Cor. 8.4; 9.1, 12, 13) uses to denote his collection. James, having
received the money, announced, perhaps with excellent intention,
perhaps to Paul’s dismay, ‘We shall use part of this gift to pay the
expenses of our four poor Nazirite brothers, and will do so in your
name, so as to still the rumours that you no longer care for the
ancestral religion and observe the Law.’ If this happened Paul could
hardly complain; the gift was presumably given with no strings
attached to it, and it was true that he understood the Gospel as the
fulfilment of God’s promise to his people and was prepared on
occasion to act ‘as if he were a Jew’ (1 Cor. 9.20).
Had James ulterior motives? Did he hope to discredit Paul in the
estimation of Gentile Christians? Did he even hope to ensnare him
into the Temple and provoke the riot that ensued—in which, as in
the whole legal process that followed, there is no indication that the
Christians of Jerusalem made any move to aid the apostle of the
Gentiles? We have no means of answering these questions, though
we cannot avoid asking them, and v. 25 remains ambiguous. It may
mean, as many suppose, ‘We are keeping faith with the Gentile
Christians by adhering to the Decree, which does not require
circumcision’ (cf. Stählin 278). It may mean, ‘This is as far as we
will go; no more concessions can be hoped for.’ James and his
colleagues may have feared for the success of their mission to Jews if
Paul should be allowed too much freedom.
Paul’s hope to secure the unity of the church by means of his
collection failed; it may be that this is why Luke omitted the
collection from his story (cf. Haenchen 586f.). According to Roloff
(312), ‘Nichts deutet daraufhin, dass Jakobus sein früheres grundsät-
zliches Ja zur gesetzesfreien Heidenmission (15.19; Gal. 2.9) und zur
Kirchengemeinschaft mit den Heidenchristen zurückgenommen
hätte.’ This needs considerable qualification in the light of Gal.
2.12.
1002 COMMENTARY ON ACTS

15. Μετὰ δὲ τὰς ἡμέρας ταύτας, presumably the days of 21.10,

though the expression (perhaps a Hebraism) sometimes means only,
After this.
ἐπισκευασάμενοι, here only in the NT; in the middle, as here, to
prepare for one's own benefit or use, so sometimes to pack up in
readiness for a journey. So Chrysostom, Homily 45.3, τουτέστι, τά
πρὸς τήν οδοιπορίαν λαβόντες. So also Begs. 4.269. At Xenophon,
Hellenica 5.3.1 the word is used for equipping horses; could it be that
Paul was travelling on horseback?
ἀνεβαίνομεν (the first person plural continues; C D E* 36 pc have
the present, ἀναβαίνομεν) εἰς Ιεροσόλυμα. The imperfect indicates
a process described in terms of its goal (introduced by εις). Cf. 21.3,
and see BDR § 327.1, n. 1.
Hanson (209) points out, with reference to vv. 15, 16, that from
Caesarea to Jerusalem was 67 miles (cf. 23.32). It is unlikely that so
great a distance could be covered in one day, even on horseback. See
the variant in v. 16.
In vv. 15-18 the text of D has been damaged by three mutilations.
For the reconstruction of the text see below and Ropes in Begs.
3.202-5; HThR 16 (1923), 162-8; and R. P. Casey (HThR 16 (1923),
392-4). Clark (377-9) disagrees with Rope’s evaluation of the text;
see below. The Western text is ‘ein Ergebnis exegetischer Überle-
gung auf der Basis unseres Textes’ (Bauernfeind 243). Rope’s
reconstruction is as follows:
Κεσα[ραίας σὐν ήμεῖν οὗτοι δέ ἤγαγον ημάς πρὀς οὓς
ξενισθῶμεν. καί παραγενόμενοι εἴς τινα κώμην ἐγενόμεθα
παρά Νάσωνί τινι Κυπρίῳ μαθητή ἀρχαίω. κἀκεῖθεν
ἐξιόντες ἤλθομεν εις Ιεροσόλυμα. ὑπεδέξαντο δέ ἡμᾶς
ασμένως οι αδελφοί. τή δέ ἐπιούση εἰσήει ό Παύλος σὐν ἡμῖν
πρὸς Ιάκωβον]
According to Clark it is doubtful whether ξενισθῶμεν should be
preceded by πρὸς οὕς or by παρ’ ᾦ.

16. The sense of this verse is on the whole clear, but it contains an
unusual number of grammatical uncertainties.
The partitive genitive τών μαθητών serves as subject of the verb
συνῆλθον; see Μ. 1.73 (also 223 for an interesting papyrus parallel).
BDR § 164. 2, n. 6 add that ‘hinter μαθητών könnte τινες τῶν
ausgefallen sein, da man die Wiederholung des Artikels vermisst.’
But από Καισαρείας could be taken adverbially with συνῆλθον
rather than adjectivally with the subject. Cf. Xenophon, Anabasis
3.5.16.
άγοντες is taken by grammarians as in place of a future participle,
used (in accordance with the familiar idiom) to express purpose.
Thus Μ. 3.80 translates, ‘went in with us ... in order to bring us’. So
 53. PAUL AND THE CHURCH OF JERUSALEM. 21.15-26. 1003

also BDR § 339.2, n. 8, who refer to Thucydides 7.25.7 [correcting

their reference], ἔπεμψαν ... πρέσβεις ... αγγέλλοντας. It should be
noted, however, that the sentence continues, τά τε άλλα αὖ
δηλώσαντας, δτι...; it seems probable either that we should emend
to ἀγγελοῦντας, or that Thucydides was distinguishing between the
present and future participles. The ambassadors handed on the news
as they went, and also had the purpose of... In fact it does not seem
necessary to find the sense of purpose in Luke’s clause; They
accompanied us, in doing so taking us to ... Delebecque (103) has a
somewhat different explanation, which however does not seem
preferable: ‘Le participe présent peut être l’équivalent d’un futur
après un verbe de mouvement; cf. 15.27.’ Cf. also Plato, Phaedo
116c, ἦλθον ἀγγέλλων.
In the relative clause (παρ’ ὧ ...) the noun has been attracted into
the case of the relative. We must understand πρὸς Μνάσονα, παρ’ ὧ
... So Moule (IB 130); and so already Bengel (471); Page (221) (also
Delebecque 103). παρά Μνάσονα does not differ essentially. The use
of the subjunctive however introduces the sense of purpose and the
whole is equivalent to a final clause which could be rephrased: ...
πρὸς Μνάσονα, ἵνα ξενισθῶμεν παρ’ αὐτῶ. Delebecque in the end
reaches a similar conclusion: ‘... chargés de nous mener chez Mnason
... pour être hébergés chez lui’. So Zerwick (§ 343) (cf. § 18); BDR §
378, n. 2; § 294.5, n. 8; Μ. 3.109. Luke’s way of expressing this
thought is significant in that it lays some emphasis on the verb
ξενισθώμεν: They brought us to the man we were to lodge with,
Mnason ... This corresponds to Luke’s interest in travel and lodging;
see Cadbury (Making 249-53). For παρά see BDR § 238.2, n. 2.
The name Mnason (there is no article; see BDR § 268.1, n. 3)
occurs nowhere else in the NT. It was not uncommon in Greek, often
in the form Μνήσων (the a appears to be Cypriote—Blass 227);
Hemer (237) cites a number of examples. It may however have been
a Greek name chosen for use in Greek circles by a Semite bearing
some such name as Menahem or Manasseh; see BDR § 53.3, n. 9,
and especially H. J. Cadbury in FS J. R. Harris, 51-3. The narrative
that follows makes it probable that Paul would stay with a Jew rather
than with a Gentile; Mnason was probably a Cypriote Jew. For Jews
in Cyprus see NS 3.68, 69; there were many Phoenicians in the
island—in an earlier century Zeno of Citium, whose father bore the
similar name Mnaseas). Mnason was also an αρχαίος μαθητής. For
the adjective cf. 15.7; here it means that his discipleship, his being a
Christian, went back to the earliest days, that is, either the earliest,
post-resurrection days of Christianity in Jerusalem, or to the begin-
nings of Paul’s mission in Cyprus (13.4-12)—possibly to neither of
these, since Barnabas the Cypriote had been a Christian before Paul’s
mission. Paul is represented as being in touch with the early days of
Christianity (Weiser 596)—if this is indeed what αρχαίος means.
1004 COMMENTARY ON ACTS

MM 80f. quote an inscription in which an αρχαίος μύστης inscribes

an ἀρχαῖος χρησμός; see also ND 1.136. The name and description
may be fictitious, but it is hard to see what Luke would gain—unless
perhaps a measure of verisimilitude—by introducing a fictitious
name, and Lüdemann is probably right in thinking that Mnason goes
back to tradition. Schille (412f.) however thinks that the description
of Mnason as an αρχαίος μαθητής shows only his Zeugnisfähigkeit;
but he does not appear in Acts to bear witness to anything. He may
have been connected with the Hellenists—the Seven—of Acts 6. So
e.g. Stählin (275); Bauemfeind (243).
The Western text in this verse gives a different account of events.
It is not of great interest that the name of Paul’s host is in D (d)
Νάσων and in gig vgmss bopt 'Iάσων. These variations could have
arisen accidentally and no known interest or purpose could have
occasioned them. The more interesting variation is best given by
simply transcribing the text of D. Ropes’s reconstruction (Begs.
3.203-205) of the broken text is quoted on v. 15. He comments
(3.204), ‘The “Western” text is inherently highly improbable. Its
indefinite reference to the “village” is futile and over-emphasized,
especially in view of the extreme interest and importance of the goal
of their journey. As their village-host, Mnason is wholly without
significance; whereas as a resident of Jerusalem this “old disciple”
was of real consequence to the narrative.’ Ropes apparently implies
that any editor could infer one or more overnight stops on a journey
of 67 miles (see on v. 15); this however cuts both ways; Paul and his
companions probably did spend one or two nights on the way, and
could have stayed with someone called Mnason. The Western text
could have preserved a recollection of this. If however this recollec-
tion had been embodied in the original text of Acts it is very doubtful
whether it would have been removed. The Western text is favoured
by Roloff (313), but must probably be rejected. Begs. 4.269 may be
right in the judgement that both texts are clumsy, and the original
may have been clumsier than either.

17. The Western text here connects with its reading in v. 16. In
Ropes’s restoration of D (see above) we have κἀκεῖθεν (from the
village of v. 16) ἐξιόντες ἤλθομεν. d has et inde exeuntes fuimus. Μ.
3.254 regards the Old Uncial text, γενομένων δέ ημών εις
Ιερουσαλήμ (εἰς for ἐν, as often in Acts, but εις may possibly have
influenced the Western text), as especially Semitic, along with Lk.
1.44; Acts 20.16; 25.15; Jn 17.23; 1 Jn 5.8. This is probably because
= both εις and ἐν. The genitive absolute (even though incorrect—
see BDR § 423.2, n. 8) does not suggest a Semitic source at this
point. It is clear that the meaning is, When we reached Jerusalem;
perhaps, When we came into Jerusalem.
Luke emphasizes once more (cf. 15.4) that Paul was welcome in
 53. PAUL AND THE CHURCH OF JERUSALEM. 21.15-26. 1005

Jerusalem: there were no serious differences between him and the

Jerusalem Christians, only false rumours (v. 21). Rom. 15.30, 31
suggest otherwise. So far as Luke’s account is correct it may mean
that Paul came first into contact with Hellenistic Jewish Christians.
See e.g. Beyer (129) (‘... ist... nicht die ganze Gemeinde gemeint,
sondern eben der Paulus nahestehende hellenistische Teil’); Marshall
(342). The historicity of vv. 17-26(36) has been variously estimated
(see further above, pp. 1000f.); thus Dibelius (8): ‘The conclusion is
best explained if from 21.17 onwards he [Luke] no longer has a
guiding “thread”’; Lüdemann (244f.): ‘Fragen wir nach der histor-
ischen Zuverlässigkeit der obigen Quelle, so muss die Antwort
hierauf positiv ausfallen.’ Thus James had indeed become head of the
Jerusalem church; the charge against Paul (v. 21) is historical; Luke
has removed the reference to the collection. Hengel (117) has the
interesting comment, ‘It becomes particularly clear at this point that
he [Luke] knew more than he wanted to say.’ Weiser (594) points out
that there are Lucan words in the verse (γίνεσθαι, ἀποδέχεσθαι,
αδελφός), but they are hardly sufficient to prove that Luke is writing
independently.
ασμένως occurs here only in the NT (except 2.41, Ε Ψ <a>). The
adverb is ‘common in later Greek’ (LS 258); cf. e.g. Demosthenes
18.36 (237).

18. τή δέ (τε, A Ε 945 1739 1891 pc gig—perhaps rightly)

ἐπιούση; cf. 16.11—one of Luke’s regular expressions.
εἰσήει ὁ Παύλος σὺν ἡμῖν πρὸς Ιάκωβον. The verb, with πρός,
conveys a hint of entering the presence of a great person (Xenophon,
Cyropaedia 2.4.5, ὁ δ’εἰσήει πρός τὸν Κυαξάρην; Herodotus 1.99.1,
εἰσιέναι παρὰ βασιλέα), or of coming into a law court. ‘We’ are now
distinguished from Paul as witnesses of the scene (though it may
be—has been—conjectured that ‘we’ were present to convey the
collection to the leaders of the Jerusalem church—Stählin 276). The
first person plural does not return till 27.1, but this is because now
‘Paulus steht beherrschend im Mitelpunkt der Szene’ (Haenchen
582). James must be the Lord’s brother, as in 15.13 (see the note).
There is no mention now of Simon Peter, and James is evidently the
leading man in Jerusalem. See H. von Campenhausen, ‘Die Nach-
folger des Jakobus’, ZKG 63 (1950/51), 133-44; A. Ehrhardt, The
Apostolic Succession (1953). He appears in the role of a president
accompanied by a group of elders (on πρεσβύτεροι see on 11.30;
14.23; 20.17, 28), or assessors. Easton (75, 130) compares the High
Priest, with elders, in the Sanhedrin; see also Ehrhardt (Acts 73). For
πάντες τε παρεγένοντο οι πρεσβύτεροι, D (sa) have ἦσαν δέ παρ’
αὐτῷ οι πρεσβύτεροι συνηγμένοι. Delebecque (103) finds this
reading ‘tentant’. See further on the text of v. 19.
Commentators and others note the absence of any references to
1006 COMMENTARY ON ACTS

Paul’s collection, here and in v. 19. See above. ‘Ein Begleiter des
Apostels konnte das nicht übersehen’ (Preuschen 126); but one must
add, unless he had good reason for doing so. Weiser (597), discuss-
ing the possibilities, thinks it possible that the fears of Rom. 15.30
may have been realized, but that we cannot know this; Luke omits
the collection because he is emphasizing ‘den Weg des Paulus als
Weg zum Leiden in der Nachfolge Jesu’. It is likely that both motives
were operative.
19. καί ἀσπασάμενος ... ὧν] οὕς ἀσπασάμενος διηγεῖτο ἕν (D*
has ένα) έκαστον ὡς, D (sa). ἐξηγεῖσθαι is a word of wider meaning
than διηγεῖσθαι, but it is impossible here to make a clear distinction
between them, since in this context each may—and must—mean to
relate, to describe in detail (though ἐξηγ. may do more to suggest
interpretative comment). Cf. 14.27; 15.4, 12. The use at 15.12 of
ἐξηγ. may give some support to διηγ. in the present verse. There is
no doubt that Luke does regard the mighty works done by Paul as
part of the justification of the Gentile mission (see 15.12); they
are not however his only means of supporting it, and it will be noted
that the present verse makes no clear reference to miracles at all. The
several events referred to comprehensively in καθ’ εν έκαστον
consist primarily in the conversion of Gentiles. For καθ’ ἕν as
the direct object accusative after the verb cf. Demosthenes 54.26
(1265).
διακονία is clearly not yet a technical term (see 6.1,4); it is simply
Paul’s service to God that has resulted in the conversion of the
Gentiles. It is clear that, in Luke’s view, it is what has been going on
ἐν τοῖς ἔθνεσιν that is of interest to the Jerusalem church.
20. The tenses in the opening sentence are to be noted:
ἀκούσαντες, they heard the news when it was brought to them;
ἐδόξαζον, they glorified God, and continued to do so; εἶπον, they
said. See BDR § 327.1; Μ. 3.66. See v. 19, with the note; the
Jerusalem authorities are (according to Luke) delighted to hear of the
success of the Gentile mission. They address Paul as αδελφέ.
πόσαι. In an indirect question the word should be ὅσαι. τών
πεπιστευκότων goes with μυριάδες...: how many tens of thousands
there are of those who have believed (and continue to believe—
perfect participle). These believers are ἐν τοῖς Ίουδαίοις, not simply
among the Jews, as converted Gentiles might be, but drawn from
the ranks of the Jews. Baur, Munck, and Nock would like to omit
τών πεπιστ.—those who complained were not Christians. See
Bruce (2.404f.), reporting but not expressing an opinion on this
conjecture.
πάντες ζηλωταί τοῦ νόμου (cf. Gal. 1.14; also 2 Macc. 4.2).
Clarke (NT Problems 135), followed by others, has suggested that at
this time all the members of the Jerusalem church were zealous for
 53. PAUL AND THE CHURCH OF JERUSALEM. 21.15-26. 1007

the Law because the Hellenists (6.1; 8.1, πλὴν τῶν αποστόλων) had
now been driven from the city. This does not appear to be the
meaning, (a) Luke’s statement that all except the apostles were
driven from the city is probably an exaggeration; (b) there had been
plenty of time since 6.1 for some Hellenist Christians to return; (c)
there had been time for other Jewish Hellenists to become Christians;
(d) it is not to be assumed that no Hellenist Jewish Christians were
enthusiasts for the Law. Luke is simply making a somewhat
exaggerated statement which may nevertheless be a not inaccurate
description of the Jerusalem church at the time.
θεόν (P74 A B C E L 33 36 323 945 1175 1739 1891 al vg syp
bo) is probably better than κύριον (D Ψ <a> gig vgmss syh sa). εἶπόν τε
(P74 X A B E L Ψ 36 453 945 al) is better than εἰπόντες (C D <a> syh),
which has grown out of it. αὐτῷ is omitted by D. ἐν τοῖς Ίουδαίοις
(A B C Ε 33 36 945 1175 1739 1891) is awkward and exposed to
alteration: P74 has τοίς Ίουδαίοις; Ψ <a> syh have ’Ιουδαίων; D gig p
syp sa have ἐν τῇ Ίουδαία; pc omit.
Were James and the elders reporting a complaint, or sharing in it?
Calvin 2.199: ‘Although they [the elders] do not condemn it openly,
yet because they distinguish themselves from the attitude of these
men, they are tacitly acknowledging them to be wrong.’ This is
probably the impression that Luke intended to convey. How far this
corresponds with the facts cannot be determined with confidence. In
the rest of the story James and his party show little concern for
Paul.

21. κατηχήθησαν (κατήχησαν, the active, in D* 104 gig, gives a

different picture—the Jewish Christians of v. 20 have been spreading
the report that Paul is a dissident; see below) is used in a different
sense from that of the same verb in 18.25: they have been informed.
No informers are named; Luke may be thinking of those mentioned
in 15.1, 5. Since Luke evidently believed the charge to be false (see
below on this) he must be thinking of trouble-makers who spread
false reports. But he does not name them; they do not appear on
stage. The general picture of a harmonious apostolic age is retained,
but it is clear that at most only the surface is smooth. Of the passive
verb Page (222) writes, ‘The word certainly describes Paul’s oppo-
nents as acting with deliberate purpose, and suggests that they were
in a position of authority and “teachers”.’ This impression is
strengthened if the active (see above) is read, but it is quite possible
that two letters, ηθ, dropped out by accident. ‘The reader knows this
[report] to be a canard, for Paul has from the very start himself shown
a commitment to the Jewish ethos by circumcising his coworker
[Timothy]’ (Johnson 290). This is a fair observation as far as Luke’s
own understanding and intention are concerned.
The report is that Paul teaches ἀποστασίαν ἀπό Μωϋσέως.
1008 COMMENTARY ON ACTS

αποστασία (the earlier form was άπόστασις) is revolt, defection. It

is often used in a political sense (e.g. Josephus, Life 43, τήν
αποστασίαν τήν από 'Ρωμαίων), but also of religious apostasy,
especially in the LXX (Josh. 22.22; 2 Chron. 29.19; 1 Macc. 2.15).
The revolt is από Μωϋσέως; cf. Plutarch, Galba 1 (1053), τήν από
Νέρωνος αποστασίαν. Paul is alleged to inculcate the abrogation of
loyalty to Moses, that is, to the law of Moses. The allegation is
illustrated by, and may even be alluded to in, a saying attributed to R.
Eleazar of Modiim (c. AD 120-140) (Aboth 3.12): If a man profanes
the Hallowed Things and despises the set feasts ... and makes void
the covenant of Abraham our father [i.e., undoes the effect of
circumcision], and discloses meanings in the Law which are not
according to the Halakah, even though a knowledge of the Law and
good works are his, he has no share in the world to come. On the
truth or falsehood of the report concerning Paul (περἰ σοῦ) see
below.
Those who are thus informed (according to the report) are τούς
κατὰ τά ἔθνη πάντας Ιουδαίους—not Gentiles, but Jews living
among (there is a similar though not identical use of κατά at 8.1;
11.1; 13.1; 25.3; 26.13; 27.5) Gentiles, Jews of the Dispersion. The
point here thus differs from that of 15.1, 5. No demand is made that
Gentiles converted to Christianity should be circumcised and
required to keep the Law; the conviction is implied that Jews who
become Christians should continue circumcision and other practices
of the Jewish Law, and that Paul is, wrongly, contradicting this
principle by teaching Christians not to retain their Jewishness. The
second πάντες is omitted by P74 A E 33 pc latt bo. Ropes (Begs.
3.205) thinks the word so awkwardly placed that it can hardly be
original; Metzger (484) reverses the argument, rightly. For ἔθνη, Dc
has έθη; if this is not an accidental error it must mean, ‘all the Jews
who live in accordance with the (Mosaic) customs’. This does not
make good sense. But έθνος, έθος invited scribal confusion. For
μηδὲ τοῖς ἔθεσιν, D* has μήτε ἐν τοῖς ἔθνεσιν αύτού.
As at 15.1, circumcision is mentioned explicitly as the crucial
example of legal observance, λέγων takes the accusative αυτούς and
infinitives περιτέμνειν and περιπατεῖv (BDR § 409.2, n. 3 mentions
without approving the—unnecessary—conjecture αῦτοῖς). For the
use of έθος cf. 15.1 and the note; for the dative τοις ἔθεσιν cf. 14.16
(ταῖς ὁδοῖς; contrast 8.39, τήν οδόν, accusative).
Did Paul in fact teach Jews to abandon circumcision and other
provisions of the Law? ‘It was in fact so’ (Calvin 2.200). Opinions
differ. Blass (229): ‘Falsum erat crimen, cf. 1 Cor. 7.18; tamen illi
non sine causa metuebant, quia haec αδιάφορα esse Christianis ille
docebat.’ Schmithals (197) argues that Paul would not have taught
Jews to abandon the Law because this would have made life
intolerable for Christian Jews in Palestine. In successive editions of
 53. PAUL AND THE CHURCH OF JERUSALEM. 21.15-26. 1009

NTD: Beyer (129), ‘... was ja durchaus der Wirklichkeit entspricht’;

Stählin 277, ‘Dieser Vorwurf ist ungerecht.’ Luke appears to assume
that Paul did not do what he was alleged to do; the charge was not
believed by the elders, it was false, and Paul will proceed by his
actions to demonstrate his innocence. It is certain that the epistles
contain no explicit instruction that the circumcision of Jewish
children should cease, and at 1 Cor. 7.18 Paul tells already circum-
cised Jews not to undo the marks of their circumcision. If he asserts
that circumcision is nothing he asserts equally that uncircumcision is
nothing (1 Cor. 7.19; Gal. 6.15). It is clear moreover that there is a
great difference between the practice in which devout Jewish parents
offer their children to God in circumcision, and the insistence that
Gentile converts must be circumcised if they are to be saved. It is
hard to believe that Paul forbade Jewish parents to circumcise their
children, but he probably classed those who insisted on and made
much of the rite as ‘weak’ (Rom. 14; 15; 1 Cor. 8; 10). Here however
circumcision is mentioned as the outstanding example of practising
the customs generally, and it is undoubtedly true that Paul taught his
fellow Jews to sit loose to legal regulations, since he evidently
believed (Gal. 2.12-14) that Jewish Christians might, and should, eat
with Gentile Christians without any compromise on the part of the
latter, and that they could buy and eat meat of any kind without
making scrupulous inquiry (1 Cor. 10.25, 27). Luke was probably
aware (cf. Begs. 4.271) of a body of Jewish Christians who continued
to practise Jewish customs without objection from their Gentile
brothers (cf. Justin, Trypho 47) and assumed that since this state of
things existed it must have been not merely tolerated but supported
by Paul. Delling (Studien 309), thus fails to touch the point when he
writes: ‘Paulus ist in Apg. 21.21 jedenfalls dann richtig verstanden
... wenn von ihm behauptet wird, nach seiner Unterweisung
(διδάσκεις) sei die Beschneidung der Abkömmlinge jüdischer Eltern
überflüssig—sofern dieser Satz zumal Kinder christlicher Eltern
meint—d.h. für Paulus: nicht heilsnotwendig.’ The allegation is not
that Paul taught that circumcision was superfluous but that he said
that it should not be done. Weiser’s statement (598), ‘Dadurch aber,
dass Lukas den Vorwurf mit dem Begriff “Bräuche” (ethē) wie-
dergibt, deutet er selbst an, dass es sich hier nicht um die Gefährdung
der religiösen Identität Israels handelt, sondern um den Bereich
kulturbedingter Normen’, seems to presuppose an underestimate of
the weight of ἔθος.

22. τί οὖν έστιν; cf. 1 Cor. 14.15, 16. What then is to he done?
πάντως ἀκούσονται, they will certainly hear. A widely supported
variant runs: πάντως δεῖ πλήθος συνελθεῖν, ἀκούσονται γάρ. This
is read not only by Western authorities but by many others also (P74
(*) A (C2) D E Ψ <a> latt), so that if it is in origin a Western
1010 COMMENTARY ON ACTS

expansion it must, as Ropes (Begs. 3.205) says, have been ‘adopted

by the Antiochian revisers’. For the πλήθος see v. 36. If the long text
is original it is not easy to see why B C*vid 36 453 614 1175 1739*
2495 pc syp should cut it down.
23. τοῦτο anticipates the relative clause.
BDR § 353.2b rightly point out that εἰσἰν ... ἔχοντες is not a
periphrastic tense; cf. 19.14, 24. The existence of the persons
concerned and their relation to the speakers (ἡμῖν) are first
mentioned, then their religious situation (εὐχήν). είναι with
the dative denotes possession: We have four men, who ... See BDR
§ 189.1, n. 1.
The vow (ευχή) is more like a Nazirite vow (see 18.18 and the
note; StrB 2.755-61) than any other, but difficulties arise as the story
proceeds; see below. For the use of ευχή cf. Philo, de Ebrietate 2, oἱ
τὴν μεγάλην ευχήν εὐξάμενοι; the reference in the context to
abstinence from strong drink shows that the Nazirite vow is
intended.
B pc bo Origen have ἀφ’ for ἐφ’. The variant is defended by
Ropes (Begs. 3.206) on the ground that it ‘yields good sense (“of
their own act,” in contrast to Paul’s intervention; ...). Especially in
view of the infrequency of agreement between B and in errors
peculiar to them, αφ is to be accepted against the testimony of other
witnesses to the reading εφ. The latter makes a weak phrase, which,
however it originated, would commend itself to the mind of tran-
scribers.’ Few agree. Metzger (484) describes ἀφ’ as an ‘Alexandrian
refinement’, Haenchen as a Greek substitute for the strange Septua-
gintal ἐφ’; similarly Conzelmann (122).
In Luke’s narrative James and his colleagues take the view that
only public action on Paul’s part will clear his name. ‘Als hätte nicht
ein Wort der Kirchenleitung selbst genügt!’ (Schille 414). Against
this, Bruce (2.405), ‘It would take more than a merely verbal
assurance ...’. It is impossible to pronounce definitely on one side or
other of this disagreement without a good deal more knowledge than
we possess of the structure of the Jerusalem church, the authority of
its leaders, and the extent to which their writ ran in the Diaspora.
24. The structure of the verse is complicated by a variant reading.
ξυρήσονται (future indicative) is read by P74 X B* Dc E 33 614 1175
1891 2495 al; ξυρήσωνται (aorist subjunctive) is read by A B2 C D*
(so NA26; D* seems in fact to read ξυρώνται—so e.g. Ropes’s text
of D in Begs. 3.207). All witnesses appear to have as the next verb
the future indicative, γνώσονται. If a subjunctive is read in immedi-
ate connection with ἵνα it seems natural to separate it from
γνώσονται: Pay their expenses in order that they may shave ...; and
(then) all will know ... If however both verbs are future it becomes
more natural (though by no means necessary) to coordinate them:...
 53. PAUL AND THE CHURCH OF JERUSALEM. 21.15-26. 1011

in order that they may shave ... and that ail may know ... The
difference in meaning is perhaps not very great. According to BDR §
442.2d the καί (before γνώσονται) may be ‘nach Konjunktiv das
Futur verbindend zur Bezeichnung des weiteren Ergebnisses’. They
add (η. 8) ‘Durch καί wird die Folge gewissermassen verselb-
ständigt, ohne dabei aber ein selbständiger Satz zu werden (die Folge
gehört in den ϊνα-Satz hinein ...)’. Here undoubtedly that all should
know is the intention of the proposed action.
The request to Paul is expressed by παραλαβών (see C. Burchard,
ZNW 69 (1978), 156f.; cf. v. 26; 16.33) and two imperatives,
ἁγνίσθητι and δαπάνησον. The meaning of the former imperative is
not clear. According to Conzelmann (123) it must mean, ‘Tritt mit
ihnen in das Gelübde ein!’ This however would involve the 30 day
period of the Nazirite vow, and it is clear that the proceedings did not
last so long (21.27). Marshall (345) points out clearly the possibil-
ities. (1) The four men had incurred some uncleanness during the
period of their (Nazirite) vow and thus needed to go through a
process of purification (see on v. 26). Luke assumes, incorrectly, that
Paul had to share the process with them, although he had not shared
the defilement. (2) Paul had been resident outside Palestine and
therefore needed purification before he could enter the Temple
(Haenchen 586). It was arranged that this should coincide with the
end of the four men’s Nazirite period. (3) An alternative view is that
Paul would not need cleansing in order to enter the Temple but
would do so in order to share in the Nazirite vow. He had not yet
terminated the vow of 18.18. None of those views is truly sat-
isfactory, and one must conclude that Luke was imperfectly informed
about the regulations for vows and uncleanness and the events that
were planned. Fortunately the details are less important than the
fundamental proposal that Paul should clear his reputation by taking
part in the Temple ritual relating to vows, and should do so as a
partner with a group of Jewish Christians.
δαπάνησον involves no such problems. Purification required
sacrifice and sacrifice required expense. Paul would pay. Was it
proposed that the collection he had made, or part of it, should be used
for this purpose? Let all see that Gentile funds were going into the
Temple? Luke does not say so, but his silence regarding the
collection is a major problem, and gives rise to such questions as
these.
ἵνα ξυρήσονται τήν κεφαλήν. It is probable that this refers not
only to the shaving of the head and offering of the hair but to the
whole process of release from the Nazirite vow. is used in this
sense at Nazir 2.5, 6: ‘This expression is used throughout for the
offering of the he-lamb, ewe-lamb, and ram and their associated
Meal-offering (Num 6.14, 15) which the Nazirite brings on the
completion of his vow and when he cuts off and burns the “hair of
1012 COMMENTARY ON ACTS

his separation” (Num 6.18)’ (Danby, Mishnah 282). Cf. Josephus,

Ant. 19.294, Ναζιραίων ξυρᾶσθαι διέταξε μάλα συχνούς, rightly
interpreted by L. H. Feldman (LCL Josephus, 9.353) to mean that
‘Agrippa had shouldered the expense for the offerings of poor
Nazirites.’
τήν κεφαλήν. Greek would normally (though not always or
necessarily) have the plural. The singular in such cases is possible in
Hebrew and normal in Aramaic (cf. Jer. 18.16, LXX (not A)), so that
some Semitic influence may be suspected here (BDR § 140, n. 3).
ὧν (the relative is attracted into the case of the antecedent τούτων,
which is to be supplied)... ούδέν έστιν: there is nothing in the things
which.
κατήχηνται, See v. 21.
αλλά στοιχεῖς. D* has ἀλλ' ὅτι πορεύου, D has πορεύη, without
significant difference in meaning. The imperative πορεύου (D*) is
surprising and yields an awkward sentence requiring a stop after
έστιν and leaving ὅτι unexplained. It can hardly be other than an
accidental error, probably caused by the two preceding imperatives
(ἁγν., δαπ.). στοιχεῖν in the sense of conform (to) is normally
followed by a dative (e.g. Gal. 6.16). LS 1648 quote an absolute use
from Dittenberger, SIG3 2.708.5, στοιχεῖν βουλόμενος καὶ τοῖς
ἐκείνων ἴχνεσιν ἐπιβαίνειν (cf. ΒΑ 1535, citing Dittenberger, OGIS
1.308.21; cf. 1.339.51), but in this inscription the context supplies the
equivalent of a dative, and this is so also in Acts, in φυλάσσων τόν
νόμον.
φυλάσσων τὸν νόμον. For an extra-biblical example of
φυλάσσειν used in this sense see ND 4.82. ‘Deutlich ist nur, dass
sich Paulus durch die vorgeschlagene Handlung als gesetzestreuen
Juden ausweisen wird’ (Haenchen 584).
Did Paul στοιχεῖν, φυλάσσων τόν νόμον? Would he have done so
even on one special occasion with the intention of proving to
Palestinian Jewish Christians that he like them was still a good Jew
as well as a Christian? According to Bornkamm (4.160f.) this was no
more than an application of the principle stated in 1 Cor. 9.19-24.
‘So zeigt Apg 21.10-26 wenigstens an einem historisch gesicherten
Beispiel, was Paulus mit seiner 1 Kor 9.19ff. erklärten Bereitschaft,
“frei von allem allen zum Knecht” zu werden, gemeint hat.
Bekanntlich isr gerade diese letzte Handlung des Paulus im Tempel
für ihn zum Verhängnis geworden. Mit anderen Worten: seine Treue
zu dem 1 Kor 9.20 formulierten Wort hat ihn in Gefangenschaft und
Tod geführt’ (161). Davies takes a similar view: ‘That the apostle no
longer recognized the authority of the Jewish Law did not signify
that every legal observance was closed to him when he was among
Jews: his very freedom from the Law enabled him to submit to it
when he so desired ... He was merely practising in Acts 21.17ff. his
policy, or rather strategy, as revealed in 1 Cor. 9.19ff. ... Acts does
 53. PAUL AND THE CHURCH OF JERUSALEM. 21.15-26. 1013

not contradict the Epistles on Paul’s attitude to the Temple’ (192).

(The epistles show no attitude to the Temple!). See also Begs. 4.273:
‘... in what way was he a Jew to the Jews if not by observing the
Law when he was with them?’ So also many others; but the real
question here is not whether on occasion Paul would do what Jews
did: 1 Corinthians 9 proves conclusively that he was prepared to do
this. The question is whether Paul was prepared to use a special
occasion such as the one described in order to suggest something that
was not true, namely that he too (καί αυτός, he just like the ardent
Jews who suspected his loyalty) was regularly observant of the Law
as understood within Judaism. Readiness to do this is not covered by
1 Corinthians 9. The issue is not only a moral one. Paul, one would
think, must have observed that a single action such as that suggested
to him could not prove the point, and that if his motives were
suspected this would enrage the Jews even more than simple
apostasy. Undoubtedly the plan, as described in Acts, misfired. That
is, the demonstration proposed by James' was ill adapted to its
purpose—unless indeed we are to suppose (cf. Brandon, Fall 135)
that James’s real but secret motive was to discredit Paul in the eyes
of the Gentile church. Occasional conformity’ is an arrangement
that does little credit to the parties on either side of the contract. For
Haenchen’s suggestion regarding what took place see above, p.
1000f.
Luke writes in a situation in which it is accepted that Jewish
Christians may and do observe the Law, and it is part of his
conviction that Paul was both a good Jew (this will be frequently
repeated in the ensuing chapters) and a good Christian. Paul was in
fact a Jewish Christian of a kind that could hardly continue to exist
after the first generation—a fact that was not clearly seen by Luke.
The story presupposes that Jewish Christians in Palestine, in
Luke’s day and before it, continued to observe the provisions of the
Law. This is brought out by Fitzmyer (Essays 280), ‘Jewish practices
were still admitted as part of the Christian way of life in Jerusalem as
late as c. AD 58, when Paul after a long apostolate among the
Gentiles went through the rite of the Nazirite at James’s request
(21.23-26).’ Cf. Black (Scrolls, 15, 82f.). The point is perhaps more
strongly made if Paul did not go through the rite of the Nazirite.

25. περί δῶ τών πεπιστευκότων. The concern of the Jewish

believers is not with Gentile believers; their behaviour has already
been regulated. It is presupposed here that the Decree of 15.29
applied not to a limited area only but to all Gentile Christians (see
Bruce 1.393). Jewish Christians are not asking that Gentiles should
be made Jews, only that Jews should not be made Gentiles, through
the abandonment of circumcision and other legal requirements. This
concern is clear, and it is easy to see (though not from Acts) how it
1014 COMMENTARY ON ACTS

would arise. When Paul expected Jews to eat with Gentiles he was
asking them to give up some of their Jewishness. This may account
sufficiently for the reference here to the Decree though perhaps not
for the citation of all the details. It has often been pointed out that, on
the surface, it seems that Paul is being informed of the Decree as if he
knew nothing about it, although, according to Acts, he was present
when it was formulated, approved of it, and was one of its sponsors.
The repetition may be introduced in order to inform Paul’s com-
panions, or to remind the reader (see Wilson, Law 81; cf. Gentiles
190); it may be that this is how the Decree was introduced in the
‘We’-source; it may be that we have here a trace of a divergent
tradition which did not represent Paul as having previously been
concerned with the Decree (see Maddox, 60). The last of these
possibilities (not incapable of being combined with others) may well
be true; but as suggested above Luke’s presentation of the matter is
not impossible: Here is a new point; the old one (conditions for the
admission of Gentiles) was settled long ago.
ἡμεῖς ἐπεστείλαμεν, We is emphatic, and if the reader were not
familiar with ch. 15 he would probably suppose that it referred to the
speaker and his companions—James and the Elders (v. 18). Taking
Acts as a whole it must mean ‘You Paul and I James’. It is however
natural to ask (see above) whether the wording reflects an earlier
stage of the tradition, at which James had to inform Paul of the
Decree, ἐπεστείλαμεν is we sent (or, wrote) a message, or injunction.
B C* (D) Ψ 049 614 2495 pc bo have ἀπεστείλαμεν. This is often
(e.g. Metzger 484f.) rejected as the more common word, likely
t erefore to be substituted for the other; but ἐπιστέλλειν is used at
h
15.20, and the word that differs from this parallel passage may well
be right, especially as B and D join forces in supporting it. Ropes
(Begs. 3.207) supports ἀπεστ., which is used absolutely in the LXX
and may have been changed into the more elegant word of 15.20.
κρίναντες, aorist, means after we had reached the decision. D* has
κρίνοντες (present), We sent, expressing the judgement that... This
is unlikely to be right; it may be due to the influence of d (iudicantes).
A longer variant occurs in D gig sa, which instead of ἡμεῖς have
ούδέν ἔχουσι λέγειν πρὸς σέ, ἡμεις γάρ. This reinforces the point
made above. They are the Jewish, law-observing Christians of
Jerusalem. They have exhausted their requirements with the Decree,
and wish only to make sure that they may continue to be Jews. γάρ is
added in order to use the rest of the sentence to make clear why the
Jewish Christians are no longer concerned about what may or may
not be done by Gentiles but wish only to reserve their own position.
φυλάσσεσθαι is used in the sense of to be on one's guard against.
At 15.20, 29 ἀπέχεσθαι is used; it is doubtful whether Luke
intended, or noticed, any differences in meaning. For the four evils
against which Gentile Christians must be on their guard see on 15.20,
 53. PAUL AND THE CHURCH OF JERUSALEM. 21.15-26. 1015

29.φυλάσσεσθαι is read by P74 A B 33 1175 pc vg syp co. Instead,

D <a> gig syh have μηδέν τοιοῦτον τηρεῖν εἰ μή φυλ.; and C Ε Ψ
453 945 1739 1891 2495 al have μηδέν τοιοῦτο τηρεῖν αυτούς εἰ μή
(ἀλλά 945 1739 1891 pc) φυλ. Ropes (Begs. 3.207) is probably right
in describing this as a Western explanatory expansion. τοιοῦτον is
the neuter form used in Attic, Epic, and regularly by Herodotus (see
LS 1802). The Gentile Christians are to observe no such thing (as e.g.
circumcision) but only (or, except) ... καί πνικτόν is omitted by D
gig; cf. 15.20, 29. The Golden Rule is not added.
Historical questions of great importance focus on this verse. 'One
may perhaps infer from Acts 21.25 that Luke knew that Paul himself
had not recognized the “apostolic decree”: thus James presents it as
something new and apparently unknown to him’ (Hengel 117). This
may be true about Paul but not about Luke. Schneider (2.311) agrees
with Conzelmann (123) that this verse is Lucan redaction; this is
presumably also the view of Wilson (see above) and of Weiser (599,
‘Lukas richtet sich an die Leser’). Those who give a more positive
estimate of the historical content of the verse take for the most part
the view that James and his party are saying, We are not going back
on our pledge to the Gentiles, you therefore may do what we ask so
as to deal with the residual problem of Jewish Christians in mixed
communities. See Stählin (278); also Knowling (450f.); Roloff
(315); Bruce (2.407).

26. Paul assents to the proposal and begins to put it into effect.
On the next day (for ἐχομένη, D has ἐπιούση, 24641 pc e syh have
ἐρχομένη, without differences of meaning) Paul took the men with
him (παραλαβών: cf. v. 24; the word suggests and the context
implies that Paul was taking charge of the financial arrangements).
Having been purified with them he went into (εἰσήει: D has
εἰσῆλθεν) the Temple. εἰσήει is imperfect; it is not clear that Luke
uses the tense with special intent (cf. the aorist in D, but also εἰσήει in
ν. 18) unless he is thinking of the four men as purified on different
days—but this does not seem probable.
ἁγνισθείς is presumably used in the same way as ἁγνίσθητι in v.
24. For the question whether Paul would need to be purified (possibly
on his return from Gentile territory?) see on v. 24; 24.18
(ἡγνισμένον) may simply mean that he was not in a state of impurity.
Cf. Num. 6.9; this suggests that the men have not reached the normal
completion of their vow (cf. 1 Macc. 3.49, ... τούς Ναζιραίους οἳ
ἐπλήρωσαν τὰς ημέρας) but are renewing if after incurring defile-
ment. σύν αὐτοῖς: Stählin (279) comments, ‘Richtiger wäre wohl:
für sie’.
διαγγέλλων ... ἁγνισμοῦ (the process of purification, not the
completed act—Begs. 4.274). Again this is presumably a reference
to Num. 6.9f.:... he shall shave his head on the day of his cleansing;
1016 COMMENTARY ON ACTS

on the seventh day he shall shave it. On the eighth day he shall bring
two turtle doves ... διαγγέλλειν suggests proclaim, announce, but
here it perhaps means notify (the priest in charge).
ἕως σὗ προσηνέχθη ..until the offering on behalf of each of
them had been brought. This would be consistent with the view (see
above—on εἰσήει) that the offerings for the four men were not all
due on the same day. 21.27, however, suggests that the seven day
period was the same for all. For έως συ D has ὅπως; this is regarded
by Ropes (Begs. 3.207) as a corruption, and so it may well be (d has
donec oblata est...). But a few MSS (323 945 1739 1891 pc) have
προσενεχθῇ, which is the verb form that ὅπως would require. It is
accordingly possible that the original form (or, an original form) of
the Western text had ὅπως προσενεχθῇ, a reading that no longer
appears in any extant witness.
 54. RIOT AT JERUSALEM 21.27-40

(27) When the seven days were almost completed, the Jews from Asia,
having seen Paul1 in the Temple, stirred up the whole crowd and laid their
hands upon him, (28) shouting, ‘Men of Israel, help! this is the man who
teaches all men everywhere against the People and the Law and this place.
Moreover, he has also brought Greeks into the Temple and has profaned this
holy place.’ (29) For they had previously seen Trophimus the Ephesian in the
city with him and supposed that Paul had brought him into the Temple. (30)
The whole city was excited and there was a tumultuous concourse of the
people. They laid hands on Paul and dragged him outside the Temple; and
immediately the gates were shut. (31) While they were seeking to kill him
word went up to the tribune of the cohort that all Jerusalem was in an uproar.
(32) He immediately took soldiers and centurions and ran down upon them.
When they saw the tribune and the soldiers they stopped beating Paul. (33)
Then the tribune came up, got hold of him, and commanded him to be bound
with two chains. He inquired who he might be and what he had done. (34)
Some in the crowd called out one thing, others another. Since he was not able
to get trustworthy information because of the tumult he ordered him to be
brought into the barracks. (35) When he got on the steps he was actually
carried by the soldiers because of the violence of the crowd, (36) for the
whole multitude of the people were following, shouting, ‘Away with him!’
(37) As he was about to enter the barracks Paul said to the tribune, ‘May I
say something to you?’ He said, ‘Do you know how to speak Greek? (38) 2So
are you not the Egyptian, who in the past3 raised and led out into the desert
four thousand men of the sicarii?’ (39) Paul said, ‘Truly I am a Jew, of
Tarsus in Cilicia, a citizen of no mean city. I pray you, permit me to speak to
the people.’ (40) He gave permission; Paul stood on the steps and gestured to
the people with his hand. There was dead silence, and Paul called out to them
in the Hebrew4 language:

Bibliography
J. Μ. Baumgarten, JJS 61 (1954), 215-25.
E. J. Bickermann, JQR 37 (1947), 387-405.
H. Braun, ThR 29 (1963), 174.
J. H. Iliffe, Quarterly of the Department ofAntiquities in Palestine 6 (1936),
1-3.
S. Légasse, RechScR 69 (1981), 249-56.
1Greek, hun.
2NEB, Then you are not; NJB, Aren’t you.
3Greek, before these days.
4Probably Aramaic is intended.
1017
1018 COMMENTARY ON ACTS

H. Sahlin, NovT 24 (1982), 188.

A. A. Trites, NovT 16 (1974), 278-84.
L. H. Vincent, RB 61 (1954), 87-107.

Commentary
Quickly read, this paragraph gives the impression of a straightfor-
ward account of violent events in the Temple, happening at a point
determined in relation to the process of purification in which,
according to 21.23-26, Paul had agreed to take part. Paul was
accused however not of anything amiss in the rite of purification, but
of bringing into the part of the Temple from which Gentiles were
excluded Trophimus the Ephesian. Trophimus had indeed been in the
city with Paul, but not, Luke clearly implies, in the Temple. Paul was
beaten, the city was in an uproar; the Roman tribune (Claudius
Lysias, we learn from 23.26) intervened at the head of his troops,
brought Paul into the barracks, and began to interrogate him,
believing him to be an Egyptian trouble-maker. Paul sought and
obtained permission to speak to the crowd in the Temple. This
appears to be, and is, a plain tale, and whatever its origin every part
of it is possibly historical.
There are however points of real or apparent difficulty, real
difficulty, for example, in v. 27 in the dating of the event. This
appears to be due to Luke’s lack of clarity with regard to the rules for
the Nazirite vow and other occasions of release from ceremonial
impurity. Inaccuracy here of two or three days one way or the other is
of no importance to the story. For this question see the note on v. 27
(also on 24.11). Haenchen (591) detects an inconsistency between
Paul’s being so weak after his beating that he had to be carried up the
stairs (v. 35) and his delivering shortly thereafter an energetic speech
(ch. 22). There is no problem. Paul was carried not because of
physical weakness but διά τήν βίαν τού ὄχλου. Paul was carried to
prevent his being tom limb from limb by the mob. We cannot infer
that he was gravely weakened; and, if he was, under the pressure of
great events sick men have been raised to make powerful speeches.
Paul was not without companions, and it is presumably to them
that Luke (who could himself have been one of them, though
probably he was not) owes this story and some of those that follow.
There is no use of the first person plural, nor could this be expected.
The conversation between Paul and the tribune must be based on
conjecture. There is little or nothing improbable in it. See the note.
27. αἱ ἑπτά ἡμέραι are most naturally understood in terms of
Num. 6.9 to mean that the four men who had undertaken (Nazirite)
vows had incurred some uncleanness, for removing which a seven-
day period was necessary. Alternatively (see e.g. Beyer 132; Pesch
 54. RIOT AT JERUSALEM. 21.27-40. 1019

2.224) Paul might have needed to remove uncleanness due to his

residence abroad. This would require seven days in view of Num.
19.11-13—a passage which is relevant because uncleanness caused
by life outside Palestine was caused ‘weil man befürchtete, dass dort
überall verunreinigende Gräber auch in verborgenen vorhanden sein
möchten’ (StrB 2.759). D gig (syp) have συντελουμένης δέ τής
έβδομης ημέρας, the seventh day was coming to its end, which is an
alternative expression for the same time, for ὡς δέ ἔμελλον means
when the seven days were about to be completed (though Bruce
1.394 thinks that they were about to begin). E. Schwartz, referred to
in Begs. 4.274, conjectured έβδομάδες in place of επτά ήμέραι: the
seven weeks (between Passover and Pentecost) were about to be
completed. This would introduce an otherwise surprisingly absent
second reference to the Pentecost of 20.16, but it has no other virtue,
and a reference to Pentecost in the midst of a passage dealing with
vows would be very odd. H. Sahlin (NovT 24 (1982), 188) suggested
that the number seven crept into the text from a marginal ζ', intended
as a contraction of ζήτει, meaning ‘This is doubtful; look it up’, and
misunderstood as a numeral.
oἱ (+ δέ, D) ἀπό τής ’Ασίας ’Ιουδαῖοι (+ ἐληλυθότες, D), with
whom Paul has already had trouble (20.19), saw him in the Temple,
where he had of course been for the purpose of purification, if not for
other reasons also. On the form of συνέχεον see Μ. 2.91, 265; also
BDR § 73.1, n. 5. These Jews were responsible for trouble in
Jerusalem as earlier they had been nearer home. It is they, not Jewish
Christians (who stand aloof), who attack Paul.
For ἐπιβάλλειν χεῖρας see 4.3. Instead of the aorist D has the
(historic) present ἐπιβάλλουσιν.
Schmithals (198) takes the view that vv. 27-29 are Luke’s
editorial work. There is little to be said for his opinion that ‘Die
Paulus-Quelle hat dagegen von einer Verhaftung des Paulus in
Jerusalem gar nichts gewusst.’ This is part of the theory (199) that
Paul in fact travelled to Rome as a free man, the story as we have it
being made up to imitate that of Jesus. Roloff (316) on the other hand
thinks that Luke had a narrative of Paul in Jerusalem and Caesarea:
21.27-36; 22.24-29; 23.12—24.23, 26f.; 25.1-12.

28. ἄνδρες Ίσραηλῖται: see 2.22 and the note.

βοηθεῖν in the NT usually means help; here it means come to the
rescue (see LS s.v. 2; p. 320; so e.g. Herodotus 7.158.2, ἥλθετε
βοηθήσοντες).
ό άνθρωπος ό ... διδάσκων. The emphasis lies on the participial
phrase; BDR § 270.1, n. 2. But the phrase is not consistent with the
charge brought in 21.21. There Paul is alleged by Jewish Christians
to teach Jews not to observe the Law; here he teaches all men, and
this must mean all Gentiles, since he teaches against the people
1020 COMMENTARY ON ACTS

(λαός) as well as against the Law and ‘this place’ (the Temple). Cf.
the charges against Stephen, which include a reference to ‘this place’
(6.13, 14). In the epistles Paul does not teach against the people (see
especially Romans 9-11, though on the other side 1 Thess. 2.14-16);
his teaching about the Law is notoriously complex, with positive and
negative elements; he does not refer to the Temple. Luke of course
does not mean to suggest that the accusations in this verse were
justified.
πάντας πανταχῇ. Such paronomasia is classical. BDR § 488.1a,
n. 2 does not refer to this passage in Acts but quotes what is a good
parallel: Plato, Menexenus 247a, διὰ παντός πάσαν πάντως
προθυμίαν πειρᾶσθε ἔχειν. For the spelling of πανταχῇ (with iota
subscript) see Μ. 2.84, and contrast πάντη at 24.3 (where NA26 has
the iota but Moulton does not).
Έλληνας, plural; in the next verse Trophimus alone is mentioned.
Greeks may be an inference, wilful exaggeration, or a pluralis
categoriae (cf. 19.38). See Μ. 3.26.
Non-Jews were forbidden to enter the Temple beyond the Court of
the Gentiles. For the inscription placed on the wall surrounding the
inner precincts see Dittenberger, OGIS 2.598:

ΜΗΘΕΝΑ ΑΛΛΟΓΕΝΗ EICΠO

PEYECΘAIENTOC TOY ΠΕ
PI TO ΙΕΡΟΝ ΤΡΥΦΑΚΤΟΥ KAI
ΠΕΡΙΒΟΛΟΥ OC ΔΑΝ AH
ΦΘΗ ΕΑΥΤΟΥ AITIOC EC
TAI ΔΙΑ TO ΕΞΑΚΟΛΟΥ
ΘΕΙΝ ΘΑΝΑΤΟΝ

A further fragment of an inscription was found in 1935; see J. H.

Iliffe in Quarterly of the Department of Antiquities in Palestine 6
(1936), 1-3. For the prohibition (and some references to the
inscription) see Josephus, Apion 2.103f.; War 5.193f. (an inscription
Έλληνικοῖς... 'Pωμαϊκοῖς γράμμασιν); 6.125; Ant. 12.145; 15.417
(on the last see the note by R. Marcus and A. Wikgren in LCL
Josephus 8.202f.); Philo, Legatio ad Gaium 212; Middoth 2.3; Kelim
1.8. Cf Eph. 2.14. On the whole see Sevenster (Do you know Greek?
116); NS 2.285f. Sherwin-White (38) observes that ‘the wording is
very curious and suggests lynching rather than execution’.
εἰσήγαγεν ... κεκοίνωκεν, aorist and perfect. ‘Die geschehene
Einführung hat die Befleckung zur dauernden Wirkung gehabt’
(BDR § 342.2, n. 5). In 1QpHab 12.6-9 the Wicked Priest defiles the
Temple; there is no reference here to Paul.
Tajra (123) emphasizes the importance of the charge of Temple
profanation. Only on this could Jews get a death sentence and have it
carried out.
 54. RIOT AT JERUSALEM. 21.27-40. 1021

29. ἦσαν προεωρακότες, periphrastic pluperfect. The pluperfect

forms were not always used in earlier Greek, and were tending to go
out of use in the NT period.
Trophimus, from Asia; 20.4. Had the others mentioned in that
verse not reached Jerusalem? Or was it by chance that Trophimus
had been seen and identified? It was a very common name; see ND
3.91-3. Hanson (212): They ‘would have known what Trophimus,
who came from the capital of their province, looked like.’ This seems
most unlikely; Ephesus was a large city, Asia a very large province.
ἐνόμιζον, an illogical but perhaps not unnatural supposition. D has
ἐνομίσαμεν; it is hardly possible to make sense of this first person
plural and it must be regarded as a simple error; Ropes (Begs. 3.207)
prints ἐνόμισαν as the (intended) text of D, with ένομίσαμεν at the
foot of the page. d has putaverunt.
ὅv is anticipated from the ὅτι clause = ἐνόμιζον ὅτι Π. εἰσήγαγεν
αυτόν. An accusative and infinitive construction would have been
ambiguous here; see BDR § 397.2, n. 5.

30. ἐκινήθη. Cf. Josephus, Ant. 3.13, τοῦ πλήθους ... πικρῶς
ἐπ’αὐτόν κεκινημένου. As the context indicates, violent excitement
is meant.
συνδρομή, tumultuous concourse (LS 1703). Cf. Judith 10.18,
with ἐγένετο, as here.
For the sequence of imperfect (εἷλκον) and aorist (ἐκλείσθησαν)
see Μ. 3.66 and especially BDR § 327.1, n. 1. BDR point out that
ευθέως implies ἐλκυσθέντων αυτών (but they must mean
ἐλκυσθέντος αυτού). When the process of dragging Paul out was
complete, the doors were shut. Or perhaps, the gate was shut, since
θύραι, plural, was used of the folding doors of such a gate as the
Temple may be supposed to have had. Presumably once the supposed
offender was outside the Temple authorities closed the gate(s) in
order to prevent further profanation—by the murder of Paul?
Marshall (348) follows Jeremias (Jerusalem 210) in thinking that
the gates of the inner courts, not the outer gates of the Temple, were
closed.
For ἐπιλαμβάνεσθαι see ND 4.104.

31. The subject of ζητούντων (which would be αὐτῶν, the

members of the crowd) is not expressed; such omission of the
pronoun in a genitive absolute is both classical and Hellenistic:
Zerwick § 50; BDR § 423. One would have supposed that a serious
attempt to kill Paul would have succeeded long before word could be
brought to the tribune, φάσις is not a formal report; nearer to rumour.
The φάσις άνέβη, that is, went up to the official’s headquarters in the
Antonia; see NS 1.361, 362, 366. For the watch over the Temple
from the Antonia towers see Josephus, War 5.242; there was a view
1022 COMMENTARY ON ACTS

of the whole area of the Temple. Troops were kept in readiness

during festivals: War 5.244; Ant. 20.106f.
σπείρα is the Latin cohors; see 10.1. χιλίαρχος is tribunus
(militum), the commander of a cohort.
The report was that the whole of Jerusalem συγχύννεται (so A
B D 33 (1175) pc vg; συγκέχυται is read by P74 Ε Ψ <a> d gig),
was in an uproar. This was the sort of event that could easily get out
of hand; it was the purpose of the Roman garrison to nip such
disturbances in the bud. Metzger (485) notes that syhmg* adds at the
end of the verse, See therefore that they do not make an uprising.
τε is read by A Β E 945 1739 1891 2464 pc. δέ by Dc Ψ <a> lat
syh; p74 p have neither. After ἀνέβη, P74 adds δέ.
32. δς ... κατέδραμεν. Continuation of the narrative by means of
a relative clause is characteristic of Acts. ἐξαυτῆς is an Acts word
(10.33; 11.11; 21.32; 23.30; Mk, once, Paul, once). These two
features however could prove at most some Lucan retouching.
BDR § 419.1: Luke uses the participle παραλαβών rather than an
adverbial clause, such as ἐπειδή παρέλαβεν, because here the
participle, like the classical λαβών, is virtually equivalent to with.
True, but there is in fact nothing unusual in the adverbial use of the
participle. λαβών is read by B (cf. d, sumptis militibus).
έκατοντάρχας. For centurions see 10.1. The plural means that at
least two were sent, presumably accompanied by two centuries, with
a paper, but not necessarily an actual, strength of 200 men. This
intervention had the intended effect: the crowd stopped beating
Paul.
33. That the tribune himself led his forces and began the inter-
rogation suggests that the disturbance was taken very seriously.
ἐπιλαμβάνεσθαι: see on v. 30.
Two chains: perhaps attached to two soldiers, one on each side.
ἐπυνθάνετο (imperfect—began to question) τίς εἴη καί τί έστιν
πεποιηκώς. The first dependent verb is optative, the second (peri-
phrastic perfect—cf. v. 29) indicative. If the two dependent ques-
tions are to be distinguished, the former indicates uncertainty, the
latter the conviction that the man (whoever he might be) had
certainly committed some crime. So Bruce 1.397; Μ. 1.199; and
BDR § 386.1, n. 2. But Page (225) thinks it possible that τίς εἴη καί τί
εἴη πεπ. may have been avoided as ugly.
From the outset, Paul is in Roman hands.
34. άλλοι δέ άλλο τι. For the construction cf. 19.32. There was a
popular move rather than a concerted plan against Paul.
μή δυναμένου δὲ αὐτού ... ἐκέλευσεν. The genitive of the
absolute construction becomes the subject of the finite verb, a serious
abuse of the construction. BDR § 473.3, n. 9.
 54. RIOT AT JERUSALEM. 21.27-40. 1023

τό ασφαλές, trustworthy information.

θoρυβόν: the sense of noisy tumult (cf. 20.1) is evident here.
For ἄγεσθαι, P74 has ἀνάγεσθαι, which corresponds with the
position of the Antonia, which is no doubt what is intended by τήν
παρεμβολήν. See Josephus, Ant. 15.403-409. παρεμβολή meant
originally a drawing up of soldiers in battle order, then a company of
soldiers, a camp; here, the soldiers’ quarters, or barracks. On the
origin of the word see Phrynichus (353): παρεμβολή δεινῶς
Μακεδονικόν καίτοι ἐνήν τῷ στρατοπέδῳ χρῆσθαι, πλείστῳ καὶ
δοκίμῳ ὄντι.
35. ἐγένετο here is little more than was: the sense of becoming
may be brought out by the colloquial, When he got on the steps.
συμβαίνειν (impersonal, it happened) was going out of use in
post-classical Greek; it was probably used here because the much
more common ἐγένετο had already been used in the sentence. It
came about that he was carried. Cf. Μ. 2.427 (comparing Tobit 3.7
BDR § 393.1, n. 2. There is more emphasis than in the simple
ἐβαστάσθη: he was actually carried.
τούς αναβαθμούς. ‘At two points, steps (καταβάσεις) led down
from the fortress to the Temple-court. This is precisely the situation
that emerges from the Acts. For when Paul was taken into custody by
the soldiers during the uproar in the courtyard of the temple, and was
led away to the barracks (παρεμβολή), he was carried up the steps
(αναβαθμοί) by the soldiers to protect him from the crowd, and from
there, with the permission of the “chiliarch”, addressed the people
once more (Acts 21.31-40)’ (NS 1.366). See Josephus, War 5.243:
καθὰ δέ συνήπτε ταῖς τού ιερού στοαῖς εις ἀμφοτέρας εἶχε
καταβάσεις, δι’ ὧν κατήεσαν οί φρουροί. See also Dalman (SSW
298).
For ἐπί, D has εις—less suitable, and therefore conceivably
original. For όχλου, D latt syp have λαού—possibly an attempt to
pin blame for the riot on the Jewish people. See however v. 36.
Haenchen (591) finds the beating and carrying inconsistent with
the speech that follows; on this see p. 1018 above.
36. ἠκολούθει γάρ explains διά τήν βίαν in ν. 35.
κράζοντες may be said to agree with πλήθος as a noun of
multitude.
αιρε αυτόν: Cf. Lk. 23.18 (αἶρε τούτον); Acts 22.22; Jn 19.15;
Martyrdom of Polycarp 3.2; 9.2 (αἶρε τούς ἀθέους). Luke probably
intends to recall the story of Jesus. Cf. also Isa. 53.8, as at 8.33. See
Wilcox (67).
τού λαού is omitted by D gig; in its place 614 pc syh have τού
όχλου. It cannot be said that the Western text here (cf. v. 30)
manifests an anti-semitic (anti-Jewish) tendency; for the Western
authorities it is just the crowd, not the crowd of the people (of God),
1024 COMMENTARY ON ACTS

that pressed upon Paul and sought his death. But it is doubtful
whether the variants have so much force. At the end of the verse D
has ἀναιρεισθαι instead of αἶρε αυτόν. The majority reading is
more forceful.
Roloff (318) comments that Jerusalem had now ‘seine heils-
geschichtliche Stellung verloren’. It is not clear what this means. The
Geschichte certainly is moving into a new phase.
37. Μέλλων ... εις τήν παρεμβολήν. Cf. ν. 35.
ὁ Παῦλος ... τῷ χιλιάρχω: D has τω χ. ἀποκριθείς εἶπεν. The
Old Uncial text makes explicit who is speaking and may well be
secondary; there is no evident reason why the name should have been
omitted. Equally there seems no reason why εἰπεῖν τε should be
replaced by λαλῆσαι (D gig), or vice versa. Some fairly free
rewriting has been done. εἰ introduces a direct question (πρός σἐ); cf.
1.6; 7.1.
The tribune is surprised (for a reason that appears in the next
verse) that Paul is able to address him in Greek. ‘Do you know how
to speak Greek?’ ‘Not “know” Greek, for γινώσκεις is an ellipse for
γινώσκεις λαλῆσαι Ελληνιστί’ (Begs. 4.276). On the general ques-
tion involved here see S. Lieberman, Greek in Jewish Palestine
(1965) and J. N. Sevenster, Do you know Greek? (1968). Sevenster
concludes (188): ‘Divers literary and archaeological data from
different centuries agree in their testimonial to a knowledge of Greek
in broad layers of the Jewish population in Palestine. And they
contain not a single manifest indication that this was lacking during
certain periods in the first three centuries A.D.’ See also ND 5.5-40,
on bilingualism. Paul is recognized as a ‘productive bilingual’.
Of vv. 37-40 Lüdemann (246) writes, ‘Der Abschnitt ist in toto
redaktionell.’ Marshall (350) (cf. the note on v. 33) emphasizes the
importance for Luke of Paul’s dealings with Romans and his journey
to Rome. These will occupy a quarter of the book.
38. οὐκ άρα σὺ εἶ ...; It is not easy to be certain of the precise
sense of the opening words of the sentence (for which D has οὐ σὐ
εἶ). ἅρα is probably used mainly to express surprise (BDR § 440.2,
n. 4, ‘Die Verwunderung bezeichnend’) but it probably includes also
an element of inference, missed in the Vulgate’s Nunc tu es ...?
Blass (233) cites as a parallel Aristophanes, Birds 280, ού σὺ μόνος
ἄρ’ ἦσθ’ ἔποψ ...; The two possibilities are, So you are not the
Egyptian? and So are you not the Egyptian? The former alternative is
taken by, among others, Haenchen (592): ‘Du bist also nicht ...?’
But the latter seems better, grammatically and because Egyptians
undoubtedly spoke Greek; there was nothing in Paul’s use of Greek
to show that he was not the Egyptian. It may be that Marshall (352) is
somewhat too positive with ‘Do you know Greek? Surely, then, you
are the Egyptian ...’, though in this rendering he is preceded by
 54. RIOT AT JERUSALEM. 21.27-10. 1025

Bengel (472): ‘Tribunus militum ita colligebat: Paulus Graece

loquitur; ergo est Aegyptius.’
ὁ Αιγύπτιος. Cf. 5.36f. For the Egyptian see Josephus, War
2.261-3; Ant. 20.169-72. See especially the note by Feldman on the
latter passage (LCL Josephus 9.480f.); also Klausner (Jesus 21);
Knox (Hell. El. 26) (‘It is at least possible that they [Theudas and the
Egyptian] claimed to be Jesus-Joshua returning to fulfil the pro-
phecies of the end (cf. Mk 13.6), hoping to enlist Christian support’)
Begs. 4.277. There is much to be said for the view that the Egyptian
is to be identified with the ben Stada of Jewish tradition (T.
Sanhedrin 10.11; Sanhedrin 67a; Shabbath 104b). According to
Josephus the Egyptian prophet or charlatan proposed to take Jerusa-
lem, either by a conventional attack (War) or by causing its walls to
fall down (Ant.). His forces were put down by Felix, but he himself
escaped; the tribune’s supposition that Paul was he, renewing his
efforts, was thus quite reasonable. The population of Jerusalem had
assisted the Romans in repulsing him (War 2.263) so that they might
well set on him if they saw him again.
ἀναστατώσας (Ε, ἐξανα.): unsettled, but here perhaps raised (a
mob—or perhaps an army, the 4000). According to Acts he led his
men into the desert. According to War 2.262 he led them by a
circuitous route (περιαγαγών) from the desert (ἐκ τής ἐρημίας) to
the Mount of Olives. According to Ant. 20.169 he counselled the
people to go with him to the Mount of Olives. For the significance in
this respect of the Mount of Olives see Zech. 14.1-5. For ‘messianic’
leaders who took the people into the desert see Bammel in E.
Bammel and C. F. D. Moule (eds.), Jesus and the Politics of his Day
(1984), 230. 1QS 8.1-9.11 and 1QM 11.9, though approved by
Braun (ThR 29 (1963), 174) as evidence for the ‘Wüstenzug als
Bestandteil der Heilszeit’ do not seem to be relevant here. Begs.
4.277 makes the suggestion that Luke has here misread Josephus,
who says (see above) that the Egyptian led his followers not into but
out of the desert (War 2.262).
τετρακισχιλίους. According to Josephus, War 2.261, 30,000. It is
a neat suggestion that the discrepancy may have arisen through
confusion of Δ (= 4) and A (= 30). Weiser (608) rejects this
suggestion because Josephus writes out the number, τρισμύριοι. But
how did the number appear in first century MSS of Josephus?
τῶν σικαρίων. Conzelmann (124) thinks that the Egyptian’s
movement is to be distinguished from the σικάριοι. According to
Josephus, Ant. 20.169, the Egyptian’s followers were simply τὁ
δημοτικόν πλήθος; in War 2.261 they are described as ἠπατημένοι,
but were intended to form a bodyguard, δορύφοροι (262). The
passage in Ant. goes on to speak of λησταί. Neither passage refers to
σικάριοι, but these are identified with λησταί in Ant. 20.186. See also
War 2.254-257. The sicarii carried concealed daggers (ξιφίδια), ‘in
1026 COMMENTARY ON ACTS

size resembling the scimitars of the Persians, but curved and more
like the weapon called by the Romans sicae' Ant. 20.186), and were
assassins, according to Josephus committing many secret murders in
the time of trouble leading up to the war with Rome. See StrB
2.762f.
If the tribune took Paul to be the leader of such a group, as he may
quite reasonably have done, he could hardly be expected to treat him
gently. He will learn the truth in two stages, (a) Paul is a Diaspora
Jew, (b) he is a Roman citizen. At this stage Paul appears as a
‘gebildeter und gesitteter Bürger einer angesehenen Stadt’ (Weiser
609).

39. The μέν and δέ in this verse are best taken as independent; the
μέν must be taken therefore to add weight to the affirmation (LS
1101, S.V.A I): Of a truth, I am ... Not so Bruce, 1.398 (similarly
Page 226): As regards your question to me, I am ... as regards my
question to you, I ask ... Paul is Ιουδαίος and not the Egyptian.
Ταρσεύς. MM 626 quote only the adjective ταρσικοϋφικός
(POxy 14.1705.6); se ND 4.173. For Tarsus and its significance in
Paul’s background see on 9.11; add now Μ. Hengel (The Pre-
Christian Paul (1991)). See Strabo 14.5.10-15; Dio Chrysostom 34,
referred to with reference to this verse by Sherwin-White (179f.; cf.
Lohse, Umwelt 88), who emphasizes the importance to Paul and
other Diaspora Jews of local citizenship (for Roman citizenship see
22.25). Whether Jews were citizens (in the strict sense of the term) of
Hellenistic cities is discussed at length by Tcherikover (309-22),
with reference to the views of other writers. His conclusion is: ‘Three
conclusions may be drawn from all this historical material. 1. The
civic status of Diaspora Jews was not uniform, and the extent of their
rights depended on when, how, and for what purpose the Jews came
to a given country outside Palestine. 2. The organized Jewish
community as a whole stood juridically outside the Greek city, and
the Jews who lived in it had no civic rights there. 3. Isolated Jews
could acquire civic rights individually’ (331). These points may be
accepted, especially perhaps the first. Most of the evidence available
refers to Egypt, and in particular to Alexandria; for Asia, see
Trebilco. It would be fallacious to assume that conditions obtaining
elsewhere applied also to Tarsus. We have little information about
Jews in this city: a Jew of Tarsus was buried at Joppa (CIJ 2.925,
cited by Safrai and Stem, 1.147, n. 6); an archisynagogue of the
province of Cilicia is mentioned by Epiphanius, Haer. 30.11. That
there was a partly Cilician synagogue in Jerusalem (6.9) suggests a
fairly considerable Jewish population, but how many resided in
Tarsus is unknown. It may be that Paul (Luke) uses πολίτης in a
loose sense: resident in, rather than enrolled citizen of, Tarsus. So
Tajra (78-80).
 54. RIOT AT JERUSALEM. 21.27-40. 1027

οὐκ ἀσήμου πόλεως, ‘a very characteristic Hellenistic addition’

which ‘touches the theme, with the help of an erudite quotation from
the classics, of half the municipal orations of Dio of Prusa’ (Sherwin-
White 179f.). Sherwin-White is presumably thinking of Euripides,
Ion 8, ἔστιν γὰρ οὐκ άσημος Ελλήνων πόλις (cf. Hercules Furens
849); but the litotes is not uncharacteristic of Luke (see on 12.18),
who could have made it up himself. Hemer (127) accepts the allusion
to Euripides. Conzelmann (124) quotes Dittenberger, OGIS
2.578.7f., Τάρσος, ή πρώτη κ[αί μεγίστη] καί καλλίστη
μ[ητρόπολις]. Paul when speaking to the crowd refers again (22.3) to
his Tarsiote origin, but then (see the note) depreciates it in compari-
son with his connection with Jerusalem. Paul does not refer to his
Roman citizenship till 22.25.
For the imperative ἐπίτρεψον, Ψ has the infinitive ἐπιτρέψαι, D
gig συγχωρῆσαι; there is no significant difference in meaning. For
Ταρσεὺς ... πολίτης, D (w syp) have ἐν Τάρσω δέ τῆς Κιλικίας
γεγεννημένος, missing the allusion to Euripides and assimilating to
22.3. δέ (after δέομαι) is omitted by P74 L 6 pc. Bauemfeind (249)
analyses Paul’s motivation: If he is not to die at the hands of his own
people let him at least speak to them; this may be his only chance.

40. The tribune gave permission; ἐπιτρέψαντος δέ αὐτού takes up

ἐπίτρεψον (ν. 39). The question must be asked whether, in these
circumstances, a responsible Roman officer would have done so. In
fact, though Paul was at first heard in silence the upshot was renewed
tumult and violence (22.22f.). What could the tribune have hoped
for? Would it not have been more sensible to hustle Paul imme-
diately into the Antonia—in the interests of security, supposing Paul
to be guilty of some kind of trouble-making, in Paul’s own interests,
supposing him innocent; in Rome’s interests, either way? Against
this, one might suggest that the tribune was impressed by Paul’s
personality; but in 22.24 he proposes to scourge him. Luke needs a
speech at this point in order to clarify his view of the situation.
ἐπἰ τῶν αναβαθμών. See ν. 35. ‘Er blickt in die Richtung des
Tempels’ (Schneider 2.319).
κατέσεισεν. Cf. 13.16 and other passages for the orator’s gesture,
several times referred to in Acts. D has καὶ σεῖσας, which could
easily have arisen as a mechanical error. This Western reading can be
reasonably connected, as one of a number of participles (ἐστώς ...
σεῖσας ... γενομένης) with προσεφώνησεν, but the Old Uncial text
yields a preferable sentence. For τω λαῷ, D syp have πρὸς αὐτούς.
For σιγής γενομένης, Β 945 1739 1891 pc have γεν. σιγ.; D has
ἡσυχίας γεν. Cf. Xenophon, Cyropaedia 7.1.25, πολλή σιγή.
τή Έβραῖδι (ιδία, P74 Α) διαλέκτῳ. Cf. 22.2; 26.14. Probably
Aramaic is intended. ‘That Aramaic and not Hebrew was really the
language of the people is proved by Aramaic proverbs and sentences
1028 COMMENTARY ON ACTS

which occur not only in the Midrashim, but also in the Mishna, and
first and foremost by rabbinical Hebrew itself, which is either an
aramaicized Hebrew or a hebraicized Aramaic, and so presupposes
Aramaic to have been the language of the country’ (Dalman, Jesus-
Jeshua (1929), 16). Cf. also Dalman (Words of Jesus (1909), 1-12,
especially 6f.). But see Μ. H. Segal (A Grammar of Mishnaic
Hebrew (1958), 1-20, especially 5-10); the Dead Sea Scrolls also
attest the currency of Hebrew. See also however NS 2.20-28: ‘The
principal language spoken by Jews in the various regions of Palestine
during the final centuries of the pre-Christian era was Aramaic’ (20,
with extensive bibliography). Also important is the discussion of
bilingualism in ND 5.5-40, especially 23 with the literature referred
to there.
 55. PAUL’S TEMPLE SPEECH AND THE SEQUEL 22.1-29

(1) ‘Brothers’ Fathers! Listen to the defence that I now bring to you.’ (2)
When they heard that he was addressing them in the Hebrew1 language they
gave even greater silence. He went on: (3) 'I am a man, a Jew, born in Tarsus
in Cilicia, brought up2 in this city, educated strictly in the ancestral Law at
the feet of Gamaliel, being zealous for God, as are all of you today. (4) And
I persecuted this Way to the death, binding and delivering to prison both men
and women, (5) as the high priest and all the company of elders bear me
witness. From them also I received letters to our [Jewish]3 brothers and
journeyed to Damascus, in order that I might bring those [Jewish Christians]4
who were there bound to Jerusalem, for punishment. (6) It happened to me as
I travelled and was drawing near to Damascus, about midday, that suddenly
there shone round about me from heaven a great light. (7) I fell to the ground
and heard a voice saying to me, “Saul, Saul, why are you persecuting me?’’
(8) I answered, “Who art thou, Lord?’’ And he said to me, “I am Jesus the
Nazoraean, whom you are persecuting.” (9) Those who were with me saw
the light, but did not hear the voice of him who was speaking to me. (10) I
said, “What am I to do, Lord?” The Lord said to me, “Get up and go into
Damascus; there you will be told about all the things which it has been
appointed for you to do.” (11) When I was unable to see because of the glory
of that light I was led by the hand by those who were with me and came into
Damascus. (12) One Ananias, a pious man according to the Law, who had a
good reputation with all the resident Jews, (13) came to me, stood by me and
said, “Brother Saul, regain your sight.” In that very hour I looked upon him.
(14) He said, “The God of our fathers has appointed you to know his will
and to see the Righteous One and to hear a voice from his mouth, (15) for
you shall be a witness for him to all men of the things you have seen and
heard. (16) And now5 what are you going to do? Get up,6 get yourself
baptized and wash away your sins by calling on his name.” (17) And it
happened to me when I had returned to Jerusalem and was praying in the
Temple, that I fell into an ecstasy (18) and saw him saying to me, “Make
haste and depart quickly from Jerusalem because they will not receive your
testimony about me.” (19) And I said, “Lord, they diemselves know that I
was engaged in imprisoning and beating from synagogue to synagogue those
who believe in thee, (20) and when the blood of Stephen thy witness was
shed I was myself standing by and approving and keeping the clothes of
those who were killing him.” (21) He said to me, “Go, for I shall send you
far from here to the Gentiles.” ’
1Probably Aramaic is intended.
2NJB, here in this city.
3Jewish is not m the Greek.
4Jewish Christians is not in the Greek.
5RSV, why do you wait? NEB, NJB, why delay?
6RSV, NEB, NJB, be baptized.
1029
1030 COMMENTARY ON ACTS

(22) They listened to him up to this word and [then]7 they lifted up their
voices saying, ‘Away with such a man from the earth, for it is not fit that he
should live.’ (23) As they were shouting and flinging off8 their clothes and
throwing dust in the air, (24) the tribune ordered that he should be brought
into the barracks, saying that he should be examined with the scourge in
order that he [the tribune]9 might know why they were so shouting him
down. (25) But when they10 stretched him out for the lash Paul said to the
centurion who was standing there, “Is it lawful for you to scourge a
Roman—and one not even tried at that?” (26) When the centurion heard this
he approached the tribune and reported the matter, saying, “What are you
going to do? This man is a Roman.” (27) The tribune approached him and
said, “Tell me, are you a Roman?” “Yes” he said. (28) The tribune
answered, “I acquired this citizenship at a high price.” Paul said, “But I was
born a citizen.” (29) Immediately those who were about to examine him
stood back, and the tribune was afraid when he recognized that [Paul]11 was a
Roman, and that he had bound him.

Bibliography
P. W. Barnett, NTS 27 (1981), 679-97.
O. Betz, FS Stählin, 113-23.
T. L. Budesheim, as in (51).
R. Bultmann, E & F 130-2.
C. Burchard, ZNW 61 (1970), 168f.
R. H. Connolly, JTS 37 (1936), 383f.
J. Μ. Grintz, JBL 79 (1960), 32-47.
K. Haacker, Das Institution Judaicum Tübingen 1971-1972 (1972), 106-
20.
P. W. van der Horst, NovT 16 (1974), 309.
J. Jeremias, FS Black (1969), 94.
S. Légasse, as in (54).
S. Lundgren, as in (23) [= StTh 25 (1971), 117-22].
S. Lyonnet, RechScR 69 (1981), 149-64.
H. R. Moehring, as in (23) [= NovT 3 (1959), 80-99].
Th. Mommsen, ZNW 2 (1901), 81-96.
J. Neyrey, in C. H. Talbert, Luke-Acts 210-24.
H. P. Rüger, ZNW 72 (1981), 257-63.
F. Schulz, JRS 32 (1942), 78-91; 33 (1943), 55-64.
Ο. H. Steck, ZNW 67 (1976), 20-28.
W. Stegemann, as in (43).

7Then is not in the Greek.

8RSV, waved; NEB, NJB, waving.
9the tribune is not in the Greek.
10RSV, tied him up with the thongs; NJB, strapping him down.
11Greek, he.
 55. PAUL’S TEMPLE SPEECH AND THE SEQUEL. 22.1-29. 1031

W. C. van Unnik, Tarsus of Jerusalem, 1952.

R. D. Witherup, JSNT 48 (1992), 67-86.

See also other works cited in (23) (I.438f.).

Commentary
The staple content of the speech which Paul, with the consent of the
tribune, makes to the crowd assembled in the Temple is an account of
his conversion. This event is described three times in Acts: in ch. 9,
here, and in ch. 26. The event itself, and the three forms in which it is
presented, are discussed in the introductory comment on Section 23
(9.1-19a), where the differences between the accounts are analysed.
‘The account in ch. 22 is adapted to the Jewish audience to which it is
addressed. The High Priest and the council of elders are invoked to
testify to Paul’s Jewish zeal (22.5); Jesus becomes Τησοῦς ὁ
Ναζωραίος (22.8); Ananias is described not as a disciple but as a
devout observant of the Law, respected by all the local Jews (22.12),
and he speaks not in the name of the Lord Jesus but in that of the God
of our fathers (22.14); it is in the Temple that Paul in a vision
receives his instructions (22.17); he takes the opportunity of giving
his Jewish credentials as a persecutor (22.19, 20); only when he
claims to have been sent by God to the Gentiles does he provoke
dissent (22.21, 22)’ (I.444f.). Further details are given below in the
notes on the verses.
This form of the conversion story is undoubtedly suited to a
Jewish audience (see pp. 1144f. for the adaptation of virtually the
same material to a different audience in ch. 26); whether it was suited
to the particular Jewish audience that Paul had before him is not so
clear. He is accused (21.28, 29) of teaching all men everywhere
against the people, the Law, and the (Holy) Place, and of having
brought Greeks into the Temple and thus of profaning it. In the
background lurks the report that he teaches all the Jews among the
Gentiles apostasy from Moses, saying that they should not circum-
cise their children or walk in accordance with the (Mosaic) customs
(21.21). None of these matters is touched upon, except in the
assertion that Paul is a Jew, that he is (or has been) zealous for God,
and was educated by the great Pharisee, Gamaliel, also that it was in
the Temple that he received the commission to act as a missionary to
the Gentiles. The immediate charge that he had brought Greeks into
the Temple receives no notice at all. To say this is not to say (as some
do) that the speech is irrelevant. On this occasion and throughout the
rest of the book Paul insists that he is a Jew, and to make this point as
Luke wishes to make it it was necessary to show that the conversion
was within and not from Judaism (so e.g. Schmithals 202, arguing
further that Luke was concerned to combat the Pauline Irrlehrer). In
1032 COMMENTARY ON ACTS

this sense the opening verses were relevant. Relevant also is the
conversion itself, told as an event within Judaism: supernatural
encounters must be taken seriously. It was no use denying that Paul
had lived in the Gentile world and in close relation with Gentile
communities, but this was not his choice; he himself had told God
that he would be an effective witness to Jews (vv. 19,20). It was God
who had overruled this and sent him to the Gentiles (vv. 18, 21).
At this point the speech breaks off. It is often said that this is an
artificial trick of Lucan style. The speech is in fact complete; Luke
uses interruption for effect, as he does in 4.1; 7.54, 57; 10.44; 17.32;
19.28; 23.7. This may be true—probably is true in the other cases. It
may not be true here. The speaker might well have continued: My
mission to the Gentiles was undertaken only on God’s command, not
at my desire. Obedience to God required contact with Gentiles
(whom I was seeking to bring to the Jewish God) which ordinarily a
Jew would not have undertaken. I was indeed ready to admit Gentiles
to God’s people without circumcision or other legal observance, but
Jews I never told not to circumcise their children or neglect the
divine precepts. Continued in some such way as this, the speech
could have been entirely relevant to the situation in which it is
placed. One hesitates however to affirm that so much was in Luke’s
mind, though he no doubt did see that if the course and circumstances
of Paul’s mission were to be justified at all the justification must
include reference to the event which was at once conversion and
commission, not because it was psychologically impressive but
because it included the presence of Jesus, alive after crucifixion.
The mob violence that follows the interruption of the speech
brings the Romans into the story and thus completes the main
elements of the dramatic structure of the rest of Acts. The story is
told simply; the points of difficulty are dealt with in the detailed
notes below. It was the tribune’s duty to preserve order, hence to
remove one who was evidently provoking disorder. It was conven-
tional to ‘examine’ him by the use of the lash, the sure method (it was
believed) of eliciting truth. It was natural that a Roman citizen should
wish to save himself from a possibly fatal flogging by drawing
attention to his status and his rights. Whether Paul in fact was a
Roman citizen (see 16.37), and how, if he was, he could in the
circumstances prove the fact to the tribune, are very difficult
questions; see the notes.
Weiser, in a clearly set out paragraph (607), gives nine reasons for
rejecting the historicity of the paragraph and seeing it as Lucan
redaction. The reasons are by no means of equal value, but to go
through them will cover important ground. (1) Paul had been
weakened by a severe beating; it is hardly conceivable that he should
wish to make a speech, and should make one in unfavourable
circumstances. But it is conceivable, and there are parallels; John
 55. PAUL’S TEMPLE SPEECH AND THE SEQUEL. 22.1-29. 1033

Wesley was violently mobbed at Wednesbury, escaped with a tom

waistcoat, and spoke with vigour. (2) The tribune would hardly
conclude from the fact that he spoke Greek that Paul was not the
‘Jew of Egypt’; many Egyptian Jews spoke Greek. This objection
may rest upon a misinterpretation of 21.38 (see the note); but it is
probably true that the conversation between Paul and the tribune was
written by Luke. (3) The silence of 21.40; 22.2 is improbable. A
matter of opinion; given a speaker of intense personality it is quite
likely. (4) The speech does not deal with the mob’s complaints. See
above. (5) The account of the conversion is Luke’s editing of the
story of 9.1-19; true. (6) The Temple vision and the reference to
Stephen are probably redactional; probably true. (7) The end of the
speech is artificial. The point is dealt with above. (8) The shout αἶρε
... τόν τοιούτον (ν. 22) recalls the redactional words of Lk. 23.18;
Acts 21.36. There is no force in this observation. (9) References to
Roman citizenship are all redactional. Perhaps; but how, without
citizenship, would Paul get himself transported to Rome?
These observations do no more than suggest that the material in
this chapter has passed through Luke’s hands; this is undoubtedly
true. He has made and emphasized the point that Paul the Christian
was a good Jew, as well as a Christian, a Tarsiote, and a Roman. He
was probably aware of the fact that there were those who identified
Christians with revolutionaries and trouble-makers, and made an
opportunity of showing (to his own satisfaction) that Paul was a quite
different kind of person. For the general picture of a tumult, as a
result of which the Romans took over Paul’s custody from the lynch-
minded Jews, there is much to be said. Luke’s own redactional
statement is made in 28.17, where Paul says that he was handed over
to the Romans. The Roman rescue of Paul from fanatical Jews, and
their, on the whole, fair-minded treatment of him, which sets in
motion the long legal process, is not far from the truth.

1. ’Άνδρες ἀδελφοὶ καὶ πατέρες. Cf. 7.2. The simple address

without πατέρες is more common; here Paul is showing proper
respect to a gathering which will include his seniors as well as
contemporaries and juniors. He asks their attention (ακούσατε).
μου stands first after ακούσατε, according to BDR § 473.1, n. 2 in
agreement with ‘die alte Regel, dass unbetonte (enklitische) Prono-
mina udgl dem Satzanfang möglichst nahe gerückt werden.’ In
§ 173.1, η. 1 μου is taken as one member of a double (object)
genitive, but it may be better to take it as a possessive (the difference
in meaning is in the end scarcely perceptible) and as bearing some
emphasis: Instead of listening to those who call for my death (21.36),
pay attention to what I have to say. Paul may (v. 2; 21.40) have been
speaking a Semitic language, but it is not easy to fit μου into one.
Delitzsch’s Hebrew is interesting: The
1034 COMMENTARY ON ACTS

Syriac diverges evem further from the Greek idiom. The Greek is
Luke’s own.
απολογία is a speech in defence. Haenchen (596) rightly observes
that this becomes the theme of the concluding chapters: 24.10; 25.8,
16; 26.1f., 24. So Schille (421): ‘Er zielt auf eine Verteidigung der
Kirche im Ganzen.’ ‘ “Verteidigung” ist das Stichwort für den
Schlussteil der Apostelgeschichte.’ Roloff (320), also correctly,
observes that the speech aims at ‘Solidisierung des Paulus mit seinen
Gegnern’. See further on v. 3.
2. For τή Έβραΐδι διαλέκτω see on 21.40. προσεφώνει is the
reading of A B <a>; P74 L 36 614 945 1739 1891 2495 al have
προσεφώνησεν; D Ε Ψ 1241 2464 pc vg have προσφωνεί.
Grammatically a case can be made for each of these verbal forms.
The use of a Semitic language presumably surprised the crowd, as
the use of Greek had surprised the tribune (21.37), though they had
not accused him of being a Greek invading the Temple but of
bringing Greeks into the Temple (21.28).
παρέσχον ήσυχίαν is a standard phrase. See Job 34.29; Josephus,
Ant. 5.235, τοῦ πλήθους ήσυχίαν αὐτοῦ παρασχόντος; Plutarch,
Coriolanus 18 (222), παρέσχεν αύτῷ σιωπήν καί ἠσυχίαν ὁ δήμος.
Silence naturally implies a measure of attention. D has ἡσύχασαν.
The meaning is unchanged; as frequently, the Western text displays
freedom with the wording of the text rather any difference in the
understanding of its meaning.
With καὶ φησίν the speech proper is introduced.
3. Paul introduces himself as a man, a Jew, ἀνὴρ ’Ιουδαῖος,
coordinating the noun with ἀνήρ (as at 10.28; cf. 21.39); see BDR
§ 242, n. 1.
γεγεννημένος ὲν Ταρσῷ τής Κιλικίας is much the same as
Ταρσεὺς ἐκ Κιλικίας in 21.39. For the partitive genitive of the land
in which a town is situated see BDR § 164.3 (with the impressive
term chorographic genitive!).
At this point the question of the punctuation and construction of
the sentence must be raised, together with the interpretation of ἐν τῆ
πόλει ταύτη; does this mean in Tarsus or in Jerusalem? It has been
customary to place a comma after Γαμαλιήλ and another after
νόμου. The meaning then is: I was born in Tarsus, brought up in this
city (?) at the feet of Gamaliel, instructed strictly in the ancestral law,
being zealous for God. This construction was called in question by
W. C. van Unnik, Tarsus of Jeruzalem (Mededelingen der Konin-
klijke Nederlandse Akademie van Wetenschappen, Afd. Letter-
kunde, 15:5, 1952: ET 1962), who concluded (169) that the verse
means, T am a Jewish man, born in Tarsus in Cilicia, but my parental
home, where I received my earliest upbringing, was in this city
(Jerusalem) and with Gamaliel, well known to you, I received a strict
 55. PAUL’S TEMPLE SPEECH AND THE SEQUEL. 22.1-29. 1035

education as a Pharisee, so that I was zealous for God, just as you all
are up to this day.’ This means punctuating after τῇ πόλει ταύτη
(though in the translation quoted van Unnik has no stop at this
point) and after νόμου. Two main reasons are given for van
Unnik’s punctuation. (1) The three participles, γεγεννημέος,
ἀνατεθραμμένος, πεπαιδευμένος, correspond to the three recog-
nized stages in which the career of a notable person would be
sketched (in Who ’s Who, as well as in Hellenistic literature—which
to some extent lessens the force of the argument): birth, childhood,
education. Haenchen (597) agrees with van Unnik. (2) To this
corresponds the fact that, as van Unnik shows, ἀνατρέφειν refers
rather to the upbringing of a child than to such training as would be
given by a notable rabbinic teacher such as Gamaliel. This argument
is referred to by Turner (Insights 83f.), who does not take sides over
the punctuation, but argues strongly that this city must be Tarsus.
‘Unless “this city” refers to Tarsus, St. Paul contradicts what he is
reported as saying to Claudius Lysias, when he emphasised his origin
in the Greek city of Tarsus and on the basis of so respectable an
association, political and cultural, he had begged leave to address the
crowd (Acts 21.39).’ This argument is baseless; the claim to Tarsiote
citizenship no more proves long residence in Tarsus than the claim to
Roman citizenship means that Paul had resided in Rome; in each
case the citizenship was inherited, C. Burchard (ZNW 61 (1970),
168f.) agrees with van Unnik, making two refinements. He places a
comma not after νόμου but after πεπαιδευμένος, and connects κατά
... νόμου with ζηλωτής ... θεού:
I am a Jew,
born in Tarsus in Cilicia,
brought up in this city,
educated at the feet of Gamaliel,
A zealot for God in accordance with the ancestral law,
As all of you are to this day.
For the translation ‘bisauf den heutigen Tag’ he cites Joseph and
Aseneth 4.7, ἔστι δέ οὗτος ... παρθένος ώς σὺ σήμερον (cf. ed.
Philonenko 4.9, ἔστιν ’Ιωσήφ ... παρθένος ώς σὺ σήμερον). There
is little point in Burchard’s translation; if you are a virgin on this day
you have been a virgin up to this day. It is probable that Luke meant
that Paul received not only his higher but also his elementary
education not in Tarsus but in Jerusalem. This if true would mean
that his roots were not in Diaspora but in Palestinian Judaism. This
seems to be what Luke is saying; whether or not he is correct is
another question, which requires further consideration. J. Jeremias
(FS Black (1979), 94) writes: ‘Die Apostelgeschichte behauptet,
Paulus sei Theologe und zwar Schüler des Hilleliten Rabban Gam-
li’el I gewesen (Ac 22.3). Die Nachprüfung ergibt: der Verfasser der
1036 COMMENTARY ON ACTS

Paulusbriefe verfugte in der Tat über die theologische Bildung seiner

Zeit und gehörte der hillelitischen Schule an.’ Cf. K. Haacker, ‘War
Paulus Hillelit?’ in Das Institution Judaicum Tübingen 1971-1972
(1972), 106-20. Bultmann (E & F 131) thinks that Acts 22.3 cannot
be correct since Gal. 1.22 shows that before his conversion Paul ‘has
not resided for any length of time in Jerusalem’. Gal. 1.22 does not in
fact show this; it claims only that Paul was not personally known to
the Christian churches in Judaea. And Paul’s claim that he advanced
in Judaism beyond his contemporaries (Gal. 1.14) almost demands
residence in Jerusalem, where Paul’s strongest rivals would be
found. The same claim refutes the assertion by Schmithals (202) that
there is nothing to show that Paul could read Hebrew or speak
Aramaic. Betz (107) thinks that this verse proves nothing regarding
the historicity of the connection with Gamaliel. It is characteristic of
θεῖοι ἄνδρες to provide accounts of their parentage and education.
Luke does this on Paul’s behalf.
For παρά τούς πόδας cf. Aboth 1.4, Be in the dust of their feet
(i.e. of the ).
κατά ακρίβειαν = ἀκριβῶς (Begs. 4.278). For ακρίβεια and
cognates in relation to Jewish religion cf. 26.5; Josephus, Ant. 17.41;
War 2.162; Conzelmann (125) adds Life 191; also Isocrates 7.40.
For ζηλωτής υπάρχων cf. Gal. 1.14. See also (for ζηλ. θεού)
Num. 25.13; Rom. 10.2; also Epictetus 2.14.13; Musonius Rufus
37.3 (many other references to Musonius are given by P. W. van der
Horst, NovT 16 (1974), 309).
Roloff (322) comments, ‘Das “ich bin Jude" ist gleichsam das
Motto, dem alles folgende untersteht.’ See on v. 1; and contrast Phil.
3.2-7.
There are in this verse a number of minor textual variants. After
ἐγῶ (P74vid A B D E 33 36 453 945 1739 1891 pc), Ψ <a> 60 add
μέν. B has Γαμαλιήλον, making the word declinable; see Μ. 2.144.
For πεπαιδευμένος D has παιδευόμενος, an unthinking variation,
certainly wrong. ὑπάρχων is omitted by D latt. For τοῦ θεού, 88 vg
have τού νόμου, syh** has τών πατρικών μου παραδόσεων (cf.
Gal. 1.14); Ψ 614 2495 pc vgmss have no genitive after υπάρχων. For
πάντες υμεῖς, P74 has πάντες, D has ὑμεῖς πάντες. In all these cases
the text of NA26 is correct.
4. Practically the whole of this verse appears in 9.2, 3. Cf. 26.10.
Christianity is referred to as ή οδός (9.2); Paul has Christians bound,
and that both men and women. Hengel (74) thinks that this verse
(ἄχρι θανάτου) exaggerates. At 22.19 (cf. 26.11) Paul inflicts only
synagogue punishments. A synagogue beating, however, could kill.
5. Paul in 9.1 makes contact with the High Priest and receives
official letters (9.2, to the synagogues). He goes to Damascus with a
view to finding Christians and bringing them bound to Jerusalem for
 55. PAUL’S TEMPLE SPEECH AND THE SEQUEL. 22.1-29. 1037

punishment. Here πᾶν τὸ πρεσβυτέριον, the whole body of elders, is

associated with the High Priest, as it naturally would be. For
πρεσβυτέριον see Lk. 22.66 (1 Tim. 4.14). The brothers are, as at v.
1, Jewish (not Christian) brothers.
ἄξων is a future participle, expressing purpose. ἐκεῖσε (thither) is
used where ἐκεῖ (there) would be expected (but see below). On this
(mis)use see Rutherford (Phrynichus 114): Tn late Greek the distinc-
tion between ποῖ ποῦ, οἷ οὗ, ὅπου ὅποι, εκεί and ἐκεῖσε, prac-
tically disappeared.’
On this verse Delebecque has detailed and important grammatical
notes. (1) παρ’ ὧν is ‘relative complexe’ (as at 11.6; 12.4; [15.29];
17.23; 19.25; [25.18]; 26.7, [12]; [27.17]; 28.8, 15—(passages in
square brackets though given by Delebecque seem doubtful or
mistaken). ‘La phrase équivaut à οἷς, παρ’αὐτών ... δεξάμενος ...
ἐπορευόμην’ (106). καί after ὧν is ‘explétif’. (2) πρός cannot mean
pour; it is contre. (3) ἐκεισε is used with a double meaning.
‘L’expression équivaut à ἄξων ἐκεῖσε (avec mouvement!) τούς ἐκεῖ
(sans mouvement) όντας. De là vient que ἐκεῖσε est pris ici, un peu
librement, dans le sens de δεῦρο’ (107). The resulting translation is:
‘Comme non seulement l’archiprêtre, m’en est témoin mais encore
tout le collège des anciens, du fait desquels, pour avoir reçu d’eux
des lettres contre des frères, j’étais en train de suivre le chemin de
Damas, chargé de mener attachés, ici à Jérusalem, ceux même qui
étaient là, afin qu’ils soient châtiés.’
Schneider (2.321) takes the sentence differently: ‘... mit
“Briefen” “an die (jüdischen) Brüder in (den Synagogen von)
Damaskus” ’. αδελφοί are Jews, οἱ ἐκεῖσε (= ἐκεῖ) ὄντες are
Christians. Cf. Haenchen (598).
δεδεμένοι, bound, is illustrated in ND 1.49.
After ἀρχιερεύς, 614 pc syh** add 'Ανανίας (from 23.2).

6. The construction differs from that of 9.3, but most of the sig-
nificant words are found in both accounts: πορευομένῳ
(πορεύεσθαι); ἐγγίζοντι (ἐγγίζειν); τῇ Δαμασκῳ (identical);
ἐξαίφνης (identical); ἐκ τού ουρανού (identical); περιαστράψαι
φῶς (περιήστραψεν φως). The present verse adds περὶ μεσημβρίαν,
about noon (for the meaning of μεσημβρία see on 8.26). The intention
may be to stress that this was no dream but an objective occurrence.
περί ἐμέ is added; this is obvious enough. The construction here is
ἐγένετο followed by a dative and an accusative and infinitive. Cf.
20.16, and see BDR § 409.4. One or two details recall Deut. 28.28f.,
but it is doubtful whether Luke had this passage in mind.
The reading of D is closer to 9.3, and provides a good example of
free Western re-writing: ἐγγίζοντι δ[έ μ]οι μεσημβρίας Δαμασκω
ἐξέφνης ἀ[πό] τού ουρανού περιήστραψέ μ[ε] φως ικανόν περὶ
έμέ.
1038 COMMENTARY ON ACTS

7. For έπεσα (weak aorist ending with strong aorist stem; see Μ.
2.208; BDR §§ 80, 81) τε, D has καί ἔπεσον. Again the construction
differs but the significant words are found in 9.5: έπεσα (πεσών);
ἤκουσα (ἤκουσεν); φωνής λεγούσης μοι (φωνήν—note the case—
λέγουσαν αὐτῶ); Σαούλ Σαούλ, τί με διώκεις; (identical). The main
difference in wording is that here we have εις τό έδαφος; in 9.4, ἐπί
τὴν γῆν. There is no difference in meaning. In construction the
difference is between a main verb and coordinate participle.
In 9.4 E 431 syp h**, and in 9.6, 629 it add σκληρόν σοι πρὸς
κέντρα λακτίζειν; the same words are added in 22.7 by E gig vgmss
syh. They are due to assimilation to 26.14; see the note on that
verse.
8. This verse substantially reproduces 9.5. The main differences
are that in ch. 9 Saul’s words are introduced by εἶπεν δέ, here by
ἐγὼ δὲ ἀπεκρίθην; Jesus’ words by ὁ δέ, here by εἶπέν τε πρὸς ἐμέ;
here the name Jesus is followed by ὁ Ναζωραίος (added in 9.5 by A
C E 104 pc h p t syp h**. For Ναζωραίος see on 2.22 and add Η. P.
Rüger, ZNW 72 (1981), 257-63.
9. At this point the narratives diverge, in order and to some extent
in substance. The parallel to this verse is found in 9.7.
Here Paul’s companions are described as οἱ δὲ σύν ἐμοὶ ὄντες; in
9.7 they are οἱ δέ ἄνδρες οί συνοδεύοντες αὐτῷ, and it is added that
εἱστήκεισαν ἐνεοί. Here it is said that το μὲν φῶς ἐθεάσαντο; there
is no parallel in ch. 9, but the companions are described as μηδένα δέ
θεωρούντες. The most interesting difference is that here in 22.9 the
companions τήν φωνήν οὐκ ἤκουσαν τού λαλοῦντός μοι, whereas
in 9.7 it is said ἀκούοντες μέν τής φωνής. Grammarians differ on
the possibility of explaining the apparent contradiction by means of
the different cases that follow ἀκούειν. Moulton (Μ. 1.66) seems to
have regarded the issue as clear: ‘The fact that the maintenance of an
old and well-known distinction between the acc. and the gen. with
ακούω saves the author of Ac 9.7 and 22.9 from a patent self-
contradiction, should by itself be enough to make us recognise it for
Luke, and for other writers until it is proved wrong.’ Turner (Μ.
3.233) is more cautious: ‘There may be something in the difference
between the gen. in Ac 9.7 (the men with Paul heard the sound) and
the accus. in Ac 22.9 (they did not understand the voice)', though
this does not seem to be quite what Zerwick (§ 69), to whom Turner
refers, says: ‘Solent quidem distinguere inter “vocem audire” et
“vocem intelligere”. Physice percipere haberet genitivum, intelli-
gere id quod dicitur, haberet accusativum. Sed haec distinctio est
arbitraria et potius facta “ad hoc’’. Fortasse licet explicare rem
potius ita: comites audiunt sonitum (quem Paulus et Lucas sciunt
esse Domini et ut talem exhibent: ή φωνή) sed sonitum non ut vocem
alicuius loquentis cognoscunt.’ This however does not explain 9.4,
 55. PAUL’S TEMPLE SPEECH AND THE SEQUEL. 22.1-29. 1039

ἤκουσεν φωνήν, and 22.7, ἤκουσα φωνής. It is wise to agree with

Moule (IB 36) ‘It seems to me (pace Proleg. 66 [= Μ. 1.66])
impossible to find a satisfactory distinction in meaning between the
Gen. and the Acc. in Acts 9.7 ... and 22.9 ...’. BDR § 173.2, n. 5
note the facts and make no attempt to explain them. BA 62 make an
interesting comparison with Maximus of Tyre 9.7.d.f., but do not
claim that this explains the variation in case. See also H. R.
Moehring, in NovT 3 (1959), 80-99: ‘To insist upon a difference of
meaning in Acts ix.7 and xxii.9 seems, to the present writer at least,
impossible’ (98). We must conclude that Luke was writing up a
familiar story freshly, and in each case included what seemed to him
to be impressive details in the most impressive way he could think
of.
After ἐθεάσαντο, D Ε Ψ <a> gig syh sa add καὶ ἔμφοβοι ἐγένοντο.
Cf. 9.7, είστήκεισαν ἐνεοί.
10. In ch. 9 there is no parallel to Paul’s question, τί ποιήσω,
κύριε; which admirably expresses his readiness to accept obediently
any charge laid upon him. The Lord’s words however are closely
parallel in the two accounts. For ἀναστάς πορεύου, the earlier
narrative has ἀνάστηθι καὶ εἴσελθε; for εις Δαμασκόν, it has εις τήν
πόλιν; for κάκεῖ σοι λαληθήσεται, it has λαληθήσεταί σοι; for περὶ
πάντων ὧν τέτακταί σοι ποιῆσαι, it has (the largest difference) ὅ τί
σε δεῖ ποιεῖν.
For the present imperative πορεύου (where an aorist might have
been expected) see BDR § 336.3, n. 5. πορεύεσθαι is one of a group
of verbs which constitute an exception to the rule: ‘Einige Imp. des
Präsens können für bestimmte einmalige Handlungen gebraucht
werden wie έγειρε, πορεύου ...’. See however 9.11; 18.26,
πορεύθητι.
The perfect τέτακται shows that the order is already in the thought
of Jesus, having been in the eternal thought of God (Delebecque
107).
11. There are close parallels with 9.8. οὐκ ἐνέβλεπον is parallel to
ούδέν έβλεπε; χειραγωγούμενος υπό τών συνόντων μοι puts into
the passive the active χειραγωγοῦντες δέ αυτόν, which continues
with εἰσήγαγον εις Δαμασκόν, for which the present verse has
ἦλθον εις Δαμασκόν.
There is no parallel in ch. 22 to ἠγέρθη δέ Σαῦλος από τής γῆς,
unless we follow d (ut autem surrexit non videbam) gig syhmg sa;
Clark thinks that the original Western text was, ὡς δἐ ἀνέστην, οὐκ
ἐνέβλεπον. Β, with ούδέν ἔβλεπον, agrees more closely with ch. 9,
and this reading is preferred by Moule (IB 89), apparently because
the role of ἐν in ἐνέβλεπον is far from clear. Probably it is a form of
emphasis, like the English a thing in T could not see a thing.’
Whereas ch. 9 has ἀνεωγμένων δὲ τών οφθαλμῶν αὐτοῦ, ch. 22 has
1040 COMMENTARY ON ACTS

ἀπὸ τής δόξης τοῦ φωτός έκείνου. Μ. 2.461 considers that this
causal use of ἀπό (cf. 11.19; 12.14) may be due to the causal use of
which ἀπό renders in the LXX, but observes that it is also
classical.
12. As in ch. 9, the story moves into a second stage with the
introduction of Ananias, though here there is nothing of his vision of
and conversation with the Lord (9.10-16). Some features of that
conversation, however, are taken up in the conversation between
Ananias and Saul (vv. 13-16). Ananias is differently described. In
9.10 he is a disciple; here he is ευλαβής κατά τόν νόμον,
μαρτυρούμενος υπό πάντων τών κατοικούντων Ιουδαίων, pre-
sumably though not quite certainly a Jew (Cornelius in ch. 10 might
have been similarly described). That he should have been a Jew does
not mean that he cannot have been also a Christian disciple. He was a
pious man, and his piety was determined by the Law. Many MSS
substitute ευσεβής for ευλαβή, without significant change of mean-
ing; the adjective is omitted altogether by P74 and Augustine. The
short text is very difficult; it might be necessary to connect κατά τόν
νόμον with μαρτυρούμενος: he was well known to live according to
the Law. This however is anything but satisfactory. For the use of
μαρτυρεῖσθαι, to have a good reputation with, see 6.3; 16.2.
κατοικούντων is used without any indication of where the persons
in question lived; no doubt for this reason ἐν (+ τή P41) Δαμασκῷ
was added by P41 Ψ <a> vgrass syh sa, εκεί by gig (syp).
It was important that Paul should be instructed not by a Gentile or
by an apostate from the Law (Calvin 2.215). ‘Dass Hananias damals
schon Glied der christlichen Gemeinde war, weiss ja nur der Leser!’
(Roloff 323).
13. The opening words give the sense of 9.17a but that verse adds
that Ananias laid his hands on Saul. In both passages Ananias begins
with the words Σαοὺλ αδελφέ, accepting Paul as already a Christian
brother. The simple command, ἀνάβλεψον, summarizes the longer
statement (9.17), The Lord has sent me, Jesus who appeared to you
on the way by which you were coming, that you may recover your
sight and be filled with the Holy Spirit. The cure of Saul’s blindness
is described with the use of the same word, ἀναβλέπειν, in both
passages. In 9.18 there is the narrative third person singular,
ἀνέβλεψεν, in 22.13 the first person singular, αυτή τή ώρα
ἀνέβλεψα εις αυτόν. For ἀνέβλεψα, 104 1739 1891 pc have
ἐνέβλεψα, Ρ74 A have ἐβλεψα. εἰς αυτόν is omitted by P41 pc d sa—
possibly (Metzger 487) to accommodate both meanings of
ἀναβλέπειν (to look up (at); to recover sight).
Saul’s recovery of sight took place αυτή τή ώρα. This is one of a
number of passages adduced by Moule (IB 93, 122) as exceptions to
the rule that if αυτός is to be used as an adjective meaning same it
 55. PAUL’S TEMPLE SPEECH AND THE SEQUEL. 22.1-29. 1041

must be placed in the position of an attributive adjective; for at the

same hour one would (according to the rule) expect τή αυτῇ ώρα.
The matter is discussed in Μ. 1.91 (Tn Luke particularly we feel that
the pronoun means little more than “that ’ ’); BDR § 288.2, n. 4
(with reference to the Aramaic Μ. 2.432 (using Acts
16.18; 22.13, where ‘Semitic sources are not in question’, to counter
the suggestion of Semitism); Black, A4 109-12 (concluding (111)
that ‘the temporal conjunction (ἐν) αὐτῇ τῇ ῶρα is, therefore, a
Lucan Aramaism’); differently Wilcox (128-30). If the expression is
an Aramaism it is so in the sense of a Septuagintalism (see Dan. 3.6;
5.5—LXX and Theodotion; 3.15—Theodotion). But it is probably
best to treat it as an extension of the usual meaning of αυτός ὁ and
translate that very hour.

14. Ananias here gives the substance of what is said to him by the
Lord in 9.15; the wording is in the main different. He refers first to
the God of our (thus showing—see v. 12—that he is a Jew and not
merely a Jewish sympathizer) fathers; he and Saul are servants of the
God of the OT; Christianity (it is implied) is the true version of
Judaism and Christians are heirs of the OT.
For προχειρίζειν see 3.20 and the note, also 26.16. Material in ND
1.28f.; 3.82 adds little. That God has appointed Paul to a special
function corresponds with the statement (9.15) that he is a chosen
vessel. Begs. 4.280 emphasizes that the word means appointed, not
fore-ordained, but appointment must logically precede the assump-
tion of the function in question, and in a theological context this
verges on fore-ordination. The appointment is in the first instance to
conversion and vocation. This is expressed in three clauses: to know
his will, to see the Righteous One, and to hear a voice from his
mouth.
It is not clear what sense Luke gives to the first clause: God’s will
in relation to what? This cannot be simply God’s will for Paul’s own
subsequent service for this is dealt with in v. 15 for the more remote
and in ν. 16 for the more immediate future. One may guess that
Luke’s meaning was that, beyond other men, Paul was to understand
God’s purpose in and his plan for saving mankind, the whole of
mankind, including the Gentiles. But this is quite uncertain; the
meaning may be simply that Paul will live on exceptionally intimate
terms with God.
The Righteous One, τόν δίκαιον, can hardly be understood
otherwise than as a title of Christ; a simple description of him as a
just person would not suffice here, though Stählin (285) may be right
in the view that as a Messianic title it also has the effect of
emphasizing the guilt of the Jews, who condemned one who was
righteous. See 3.14 and the note. Bengel (473) connects ό δίκαιος
with Paul’s emphasis on δικαιοσύνη. ‘Haec ejus justifia est summa
1042 COMMENTARY ON ACTS

evangelii, cujus testis fit Paulus. Hunc Justum vidit Paulus etiam
deinceps.’ Paul sees Christ and hears him speak; this, on the present
occasion, is the ground and content of his conversion. The vision and
audition constitute the ground of what is said in the next verse.

15. This verse corresponds to 9.15b: τοῦ βαστάσαι τό όνομά μου

ἐνώπιον ἐθνῶν τε καί βασιλέων υιών τε Ισραήλ. Bearing my name
is equivalent to being a witness to him, μάρτυς; for this word and
theme see 1.8 and the note. The stress here is on the fact that Paul is
to be a witness πρὸς πάντας ἀνθρώπους, that is, not only to Jews
but to Gentiles also (cf. 9.15, quoted above; the point is emphasized
by Bauemfeind 252). Presumably, in Luke’s narrative, the crowd did
not see this point, since it is only when the word Gentiles is explicitly
used in ν. 21 that their anger becomes uncontrollable. Paul as witness
will speak of the things which you have seen and heard. BDR
§ 342.1, n. 3 distinguish between the perfect ἑώρακας and the aorist
ἤκουσας. The former is more significant; cf. 1 Cor. 9.1. That Paul
has seen the Lord ‘gibt ihm dauernd die Weihe des Apostels’. This
overlooks the fact that Luke has defined the qualification of an
apostle to be that he has accompanied Jesus throughout his ministry
and has witnessed a (pre-ascension) resurrection appearance (1.21f.),
and very seldom (see 14.4, 14, and the notes) refers to Paul as an
apostle. It is significant that neither here, nor in ch. 9 or ch. 26, is it
said to Paul at his conversion, You are called to be an apostle. This
does not invalidate BDR’s grammatical observation (of the force of
the perfect tense); it would have been easy to write είδες καί
ἤκουσας. But inferences from it must be carefully expressed.

16. This verse is paralleled in narrative form in 9.18b: ἀνέβλεψέν

τε καί ἀναστάς ἐβαπτίσθη.
καί νῦν τί μέλλεις; to delay or linger (Haenchen 599) is a
relatively infrequent use of μέλλειν, but quite well attested: Aeschy-
lus, Prometheus Vinctus 36, τί μέλλεις καί κατοικτίζει μάτην;
Sophocles, Antigone 448, τί δ’οὐκ ἔμελλον; Conzelmann (126) (who
compares τί κωλύει, but for this see on 8.37) thinks that it may have
been connected with rites of initiation: Corpus Hermeticum 1.26,
λοιπόν, τί μέλλεις; Marshall (357) suggests, What are you going to
do?
βάπτισαι, middle, not baptize yourself, but get yourself baptized.
The middle voice is used here as (according to some MSS) it is at 1
Cor. 10.2 (of baptism into Moses). Cf. 1 Cor. 6.11, where the middle
ἀπελούσασθε is used, probably in the intransitive sense, ‘you
washed’, which of course means, ‘you washed yourselves’. Ananias
does not say (though he may mean), I will baptize you, or Baptize
yourself. But it is clear that baptism takes place as the result of a
decision made by the baptized person. Thus Zerwick (§ 232), ‘Fac ut
 55. PAUL’S TEMPLE SPEECH AND THE SEQUEL. 22.1-29. 1043

baptizaris et abluaris’; similarly BDR § 317, n. 1; also Dinkier,

Signum Crucis 211, 226.
ἀπόλουσαι accompanies βάπτισαι, reinforcing the interpretation
of the middle voice given above and also giving an interpretation of
baptism: it is for the washing away, that is, for the forgiveness of
sins. This is explicit at 2.38; cf. Bultmann (Theologie 139). The
thought is persistent through the NT; cf. also Barnabas 11.11. It is
possible in Paul’s case to relate it specifically to his resistance to the
Gospel and persecution of Christians, but the words attributed to
Ananias are probably formulaic rather than specific.
ἐπικαλούμενος τό ὄνομα αὐτοῦ helps to interpret the use of εις τὁ
ὄνομα, ἐν (ἐπί) τω ὀνόματι, in other baptismal sayings. See on 3.6.
The name is not a magical instrument effecting supernatural results;
the name is invoked, that is, it signifies faith and obedience directed
towards Christ. Bousset (Kyrios Christos (1913), 277), refers to
Barnabas 16.8 with the similar formula, ἐλπίοαντες ἐπί τό όνομα.
For καί νῦν cf. 3.17; 10.5, and see BDR § 442.8d, n. 26. Like the
Hebrew which it represents in the LXX, it often introduces
commands and questions, but it is not a Semitism.
There is no reference here to the Holy Spirit; contrast 9.17.

17. The construction of this verse is astounding; surprisingly it

does not seem to have provoked correction on the part of copyists.
Perhaps we are too fastidious. The speaker appears first in the dative
case, μοι ὑποστρέψαντι (after ἐγένετο), next in the genitive,
προσευχομένου μου (genitive absolute), and finally in the accusa-
tive, γενέσθαι με (accusative and infinitive, also dependent on
ἐγένετο). Zerwick (§ 49) describes this as ‘mira combinatio’. Cf.
15.22. The meaning is undoubtedly clear.
Paul here, omitting 9.19b-25, narrates part of the story of his
conversion and commissioning that does not appear at all in ch. 9 (or
in ch. 26). After the events near and in Damascus he returned to
Jerusalem. The aorist participle ὑποστρέψαντι means that the return
journey was completed; ὑποστρέφοντι (P74 33 pc) should mean that
Paul was on the way back. This does not agree with ἐν τῷ ἱερῷ; it
may be that the present participle as the harder reading should be
taken as original, but if Luke wrote it he must nevertheless have
meant that Paul had reached Jerusalem.
For ἐγένετο with the dative of the person to whom the event
happened, followed by an accusative and infinitive describing the
event, cf. 20.16; 22.6; and see BDR § 409.4, n. 6.
Having returned to the city Paul was praying in the Temple. Cf.
3.1. In the epistles Paul shows no interest in the Temple; that he
should, on the occasion of a visit to Jerusalem, have used it as a place
of prayer cannot be regarded as impossible, though Lüdemann (248)
thinks that to place the call and the vision in the Temple contradicts
1044 COMMENTARY ON ACTS

Gal. 1.15 and is therefore unhistorical. Haenchen (600-604) dis-

cusses at length the author’s intention in giving the Temple episode:
it shows that Paul had been a Jewish ζηλωτής, had wished to be a
witness to his own people, and had gone to the Gentiles only on
direct and repeated command from Jesus.
Prayer turned to ἔκστασις; see 10.10; 11.5 for Peter’s ἔκστασις,
with the notes. Paul fell into a trance, outside the ordinary limits of
consciousness.
18. ἰδειν continues the infinitives dependent on ἐγένετο: It
happened that I fell... and saw. ... 36 453 pc d co have εἶδον,
which makes equally good sense.
αυτόν is of course the Lord, who had appeared on the road. Why
Paul had returned to Jerusalem we do not know; he is now told to
leave the city quickly. This is emphasized by the use of both
σπεῦσον and ἐν τάχει. σπεύδειν is in the NT characteristic of Luke
(Lk., 3 times; Acts 20.16; 22.18; 2 Pet. 3.12) but it is common
enough in Greek of all periods. So is ἐν τάχει, equivalent to the
English at speed.
A prolonged stay in Jerusalem would serve no useful purpose. For
παραδέχεσθαι cf. 16.21: They (the—Jewish—inhabitants of Jerusa-
lem) will not accept your testimony. Paul was to be a μάρτυς (v. 15),
but his μαρτυρία would find no credence here.
This story of a visit to Jerusalem is not only wanting in Acts 9.29f.
and Gal. 1.17-19; it can hardly be fitted into them.
19. In this and the next verse Paul appears to argue with the Lord,
asserting that it was in fact probable that the Jews would believe
what he said. They would believe because they knew Paul’s past and
would conclude that his position had been changed only by over-
whelming considerations.
There is no reason why αυτοί should not be properly emphatic:
They themselves know. ἐπίστασθαι is a Lucan word: Acts 9 times (8
in chs. 15-28); rest of the NT, 5 times.
ἤμην φυλακίζων may be a periphrastic imperfect (I used to
imprison) or simply descriptive (I was (at work) imprisoning). There
is little difference. See BDR § 353.3, n. 8. See 8.3; 9.2 for Paul’s
activity as a persecutor.
20. See 7.58 (the witnesses laid down their clothes at the feet
of a young man called Saul) and 8.1 (Saul was in agreement,
συνευδοκῶν, with his killing, τῇ αναιρέσει αυτοῦ).
Stephen was a μάρτυς; Paul is to be a μάρτυς (v. 15). It seems
probable that Luke understood Paul to have taken Stephen’s place,
especially as the leader of a world-wide mission. This was, perhaps,
only superficially true. See Introduction, pp. xcviii, ciiif.
The part of this verse that has no explicit parallel in chs. 7, 8 is
 55. PAUL’S TEMPLE SPEECH AND THE SEQUEL. 22.1-29. 1045

ἤμην ἐφεστώς (with συνευδοκῶν and φυλάσσων another peri-

phrastic tense—see v. 19). This was an easy addition; Luke is
rewriting traditional material that he has used earlier.
‘When Paul in Acts 22.20 calls Stephen the μάρτυς of Jesus,
it sounds rather like a title’ (Maddox 86 (n. 37)—not quite
convincingly).
πρωτομάρτυρoς (L 614 945 1739v.1. 1891 2495 al syh) for
μάρτυρος introduces the language of later piety. The addition of τή
ἀναιρέσει αὐτού by Ψ <a> sy(p) assimilates to 8.1.
At this point the extant part of d ends with the words adsistens et
consentiens.
21. εἶπεν: the Lord, who had spoken in v. 18, renews his
command, knowing that Saul’s vocation is to be fulfilled in distant
parts.
πορεύου. Moule (IB 135) notes that this present imperative is
inconsistent with the aorists of v. 18. See however the note on
πορεύου in v. 10. If the tense is really significant it may be taken to
mean something like, Be on your way, possibly, Start your journey.
ἐξαποστελώ, future, is surprising; was not the Lord sending Paul
then, in the present? D improves to ἐξαποστέλλω, E to αποστέλλω.
B has the simple future, ἀποστελῶ. The tense is probably affected by
the thought that the fulfilment of the mission belongs to the future.
Whether ἔθνη means Gentiles or simply (far off, μακράν) nations,
the meaning is the same. Paul is sent to non-Jews. It is, according to
Luke, this that brings the anger of the crowd to a climax.
Knowledge of these words seems to be reflected in apocryphal
Acts.
Passio Sanctorum Apostolorum Petri et Pauli 38 (L.-B. 1.153):
Vade et ego ero in te Spiritus vitae omnibus credentibus in me; et
omnia quaecumque doceris et feceris ego iustificabo.
Acta Petri et Pauli 59 (L.-B. 1.204): Πορεύου, ὅτι ἐγὼ ἔσομαι
μετά σου, και πάντα ὅσα ἄν εἵπης ἢ ποιήσης ἐγὼ δικαιώσω.
Schneider (2.322) suggests that Luke could not put this commis-
sion in ch. 9 because he was making Peter in ch. 10 the first
missionary to the Gentiles. Schmithals (205) notes that the Gentile
mission is against Paul’s own will. But Luke does not say this, only
that Paul thought that the Jews would listen to him. Weiser (612)
makes the better point that the evangelizing of the Gentiles is the will
of the God of Israel.
22. ἤκουον, imperfect, they went on listening; followed by the
aorist ἐπῆραν: at that point, they lifted up ... In the NT ἐπαίρειν τήν
φωνήν is peculiar to Luke (Lk. 11.27; Acts 2.14; 14.11; 22.22); it is
used in the LXX (Judges 21.2 (A); Ruth 1.9,14; 2 Kdms 13.26; Ps.
92(93).3—rendering and might seem to be one of Luke’s
Septuagintalisms if it did not appear also in classical authors
1046 COMMENTARY ON ACTS

(Demosthenes 18.291 (323), ἐπάρας τήν φωνήν; Philostratus, Apol-

lonius 5.33).
αἶρε ἀπὸ τῆς γῆς recalls 21.36 (see the note), also Lk. 23.18; Jn
19.15; and Isa. 53.8, quoted at Acts 8.33. It is possible that Luke
wishes to draw a parallel between the suffering of Jesus and that of
Paul. Cf. Wilcox (67, 68). Davies (Land 282f.) thinks that γῆ should
be understood not as earth but as land; Paul is unclean and should
therefore be driven out of the land of Israel. This is unlikely in view
of the next clause, it is not fit that he should live. He should therefore
be killed, put out of the land of the living. That the mob repeat their
cry is ‘Zeichen unbelehrbarer Verhärtung’ (Weiser 612); perhaps
rather of Luke’s skill in heightening the dramatic feeling.
On the tense (imperfect though it refers to the present) of καθῆκεν
see BDR § 358.2, n. 3: ‘Das Impf. enthält dann eine Aufforderung:
Apg 22.22 οὐ γάρ καθῆκεν αυτόν ζήν, “denn er darf nicht am
Leben bleiben“ (sie fordern zur Tötung auf).’
‘Freilich ist dieser von den Hörem erzwungene Abbruch auch hier
wieder (vgl. 4.1; 17.32; 23.7) nur literarischer Stilmittel. In Wirklich-
keit ist alles, was zu sagen war, gesagt’ (Roloff 324f.). See however
p. 1032.

23. In addition to shouting, κραυγαζόντων τε (so A Β C; P74 D Ε

Ψ <a>have δέ), the crowd find other ways of expressing their feel-
ings.
ριπτούντων (from <b>ιπτέω; elsewhere in the NT ρίπτω) τά ίμάτια:
this is usually taken to mean throwing off their clothes; so e.g. BA
1474. Begs. 4.282; 5.275-7 suggests the meaning waving. It is
perhaps unnecessary to distinguish too nicely. Luke means to
describe a frantically excited crowd. Their action is the expression of
excitement, not a prophylactic against evil (as Job 2.12; Josephus,
War 2.322 might suggest). ‘Il s’agit de la jactatio togarum, geste
d’une foule excitée’ (Delebecque 108). Acta Isidori 35ff. might
provide a parallel but not as restored by H. A. Musurillo (The Acts of
the Pagan Martyrs (1954), 137).
κονιορτόν βαλλόντων εἰς τόν ἀέρα. Here there is no doubt what
is done, and the motive is the same. The dust is not itself a threat that
stones will follow, but it is a token of frenzy which will in due course
throw anything that comes to hand. Gregory Nazianzen, Oratio in
laudem Basilii 15 is not a close parallel, nor are those given by Betz
(72, 140).

24. The tribune (see 21.31) decided that it was time to put a stop to
the riot by removing and dealing with the man who had provoked it.
εἰσάγεσθαι passive after a verb of commanding, is not classical (Μ.
3.138; BDR § 392.4. n. 14), but Luke has already used this
construction (21.33) and will do so again (23.3); also in this verse,
 55. PAUL’S TEMPLE SPEECH AND THE SEQUEL. 22.1-29. 1047

εἴπας ἀνατάζεσθαι (but D* pc have ἀνετάζειν). It is Latin rather

than Greek.
Paul is to be brought into the παρεμβολή (see 21.34), the barracks
of the Roman troops quartered in Jerusalem. There he will be safe
from the mob, but not safe from every kind of injustice: he is to be
interrogated with scourges. This is not punishment but the recog-
nized way of ‘interviewing’ a slave or other lower class and possibly
reluctant witness and finding out the truth, ἐξετάζειν (but not
ἀνετάζειν) was a technical term for this process. It is the tribune’s
intention to find out, ἵνα ἐπιγνῷ (γνῷ, P74 A 6 33 36 pc), the reason
why they had so shouted at him. In this indirect question the relative
ἥν is used where the interrogative pronoun would have been correct.
It has been suggested that the tribune may not have understood the
(Hebrew or Aramaic) speech.
It is not easy to know why Begs. 4.282 translate ἐπεφώνουν in the
text ‘they shouted against him’ but write the note ‘The ἐπι- is not
therefore “against” so much as “after”, referring to the applause or
other demonstration which follows what is said.’ It is of course clear
that on this occasion the. demonstration was not one of applause but
of violent rejection. It may be best to render, ‘they were shouting him
down’. This would be possible with the simple dative, αὐτῷ; there
are variants: αυτού (Ψ 614 pc; this goes with the reading
κατεφώνουν, contained also in D, which would mean ‘to shout
down’); περὶ αὐτού (D gig).
For the form εἴπας see BDR § 81.1, n. 1. A more important
grammatical question is whether we have here an example of the
aorist participle expressing action subsequent to that of the main verb:
He ordered him to be brought in ... and [then] he said ... This is
denied by Moulton (Μ. 1.133, ‘Lysias presumably said in one sen-
tence, ‘ ‘Bring him in and examine him’’ ’—aorist participle of coinci-
dent action). Turner (Μ. 3.79f.) seems to allow a few exceptions to the
rule (of past action): Mt. 10.4; Jn 11.2; Acts 16.6; 25.13. He does not
mention the present case. Nor do BDR § 339.1, n. 4; at § 420.2, n. 3
coordination with a participle of saying (as here) is treated as a special
case. See on 16.6 and 25.13; Mt. 10.4 and Jn 11.2 are irrelevant
μάστιξ represents the Latin flagrum.

25. For temporal ώς see 16.4 and the note.

προτείνειν is to stretch forward and is used in a large number of
ways. Here the soldiers stretched Paul forward τοῖς ἱμᾶσιν. ἱμάς is a
leather thong, used for example for a shoe lace (Lk. 3.16; the word is
not used at Acts 13.25); but it is also used in the plural for the leather
strips that together make a whip (ιμάντες is used as equivalent to
lora; Homer, Iliad 23.363; POxy 1186.2, τήν διά τών Ιμάντων
αἰκείαν). The prisoner about to be given a lashing was tied to some
sort of frame; it is possible to think of his being stretched out by the
1048 COMMENTARY ON ACTS

leather thongs with which he was fastened, or stretched out for the
lash. The latter is probably correct.
There is a centurion in charge and Paul addresses him. For the Lex
Porcia and Lex Julia, which (with certain exceptions) forbade the
lashing of a Roman citizen, see 16.37, with the notes. Sherwin-White
(71) points out with reference to the present passage that ‘the
narrative of Acts agrees with the lex Julia except that it adds the
qualification “uncondemned”’. He proceeds to discuss this addition,
which seems to imply that ‘the provincial authority might administer
a flogging after sentence, presumably in a case in which a Roman
citizen had not exercised his right of appeal, or alternately (sic) in a
special category of cases at present unknown in which the lex Julia
did not apply’ (71f.). The discussion is probably unnecessary. To
ἄνθρωπον 'Ρωμαῖον Paul adds καί ἀκατάκριτον, and the καί seems
to justify the translation ‘a Roman—and uncondemned (perhaps
untried—re incognita) at that’. Cf. Tajra (83f.).
The claim to be a Roman citizen (in circumstances very different
from the aftermath of a horrifying earthquake, in ch. 16) must have
been ineffectual unless Paul was able to prove it true. A false claim to
citizenship might be punishable with death (Suetonius, Claudius 25;
cf. Epictetus 3.24.41). It is unlikely that Paul was wearing a toga
(which only citizens might wear—it was often dispensed with in the
East—Sherwin-White 149f.), but he may have been carrying his
diploma, a small wooden diptych which would attest his registration
(and birth) as a citizen. For these diplomata see Suetonius, Nero 12
(given, conferring citizenship); Sherwin-White (148f.), and espe-
cially the articles referred to by him: F. Schulz, JRS 32 (1942),
78-91; 33 (1943), 55-64. For Paul’s citizenship see further Tajra
(81-9); also 27, 74 for ἀκατάκριτος.
Citizenship and lack of trial are separated more sharply in Begs.
4.283: ‘He claimed immunity from scourging because he was a
citizen, if it was an adjunct to examination, and because he was
uncondemned, if it was intended as a punishment.’
Why did Paul not disclose his citizenship earlier? ‘Adversus
vincula v. 29, jus civitatis non allegavit Paulus, nam haec praedicta
erant; allegavit adversus verbera, ut corpus vitaque tueretur, evan-
gelium posthac praedicaturus’ (Bengel 474). This is not very con-
vincing, a fact that adds weight to Schmithals’ (205) opinion that vv.
25-29 are Luke’s literary composition, but not to his judgement that
it is ‘kaum vorstellbar’ that a Jew in Tarsus should be born a citizen.
See Lüdemann (249f.) for a special note defending Paul’s citizen-
ship. Lüdemann thinks that vv. 24-29 show Lucan tendency but are
also traditional (217f.).

26. This verse, which carries the narrative forward in a simple

way, is marked by a number of Western variants which do not add to
 55. PAUL’S TEMPLE SPEECH AND THE SEQUEL. 22.1-29. 1049

or change its meaning. One is an unnecessary expansion; one livens

the style a little. They reflect, as many Western variants do, a time at
which the wording (not necessarily the sense) of the text could be
treated without the respect due to what was regarded as an inspired
document.
ἀκούσας δέ ] τούτο ἀκούσας, D lat. After v. 25 it was scarcely
necessary to point out what the centurion heard.
ἑκατοντάρχης ] + ὅτι 'Ρωμαῖον εαυτόν λέγει, D gig vgmss—
further unnecessary expansion.
λέγων, P74 A B C Ε Ψ 33 36 614 945* 1739 2495 al vg sy bo] +
ὅρα D <a> gig p sa—a more numerously attested variant, and one that
gives an intelligible, and livelier, sentence: Look out! What are you
up to? That the centurion’s question is in any case a warning is clear
from the context and the γάρ that follows: You had better be careful,
for the man is a Roman.
27. The centurion had approached the tribune (προσελθών, v. 26);
now the tribune approaches Paul. Again D (joined by (sa)) has a
somewhat pointless variant: τότε προσελθὼν ό χιλίαρχος
ἐπηρώτησεν αυτόν. He wishes to verify the claim that Paul has
made. There is some emphasis on σύ: You, whom at first I took to be
the Egyptian rebel (21.38), whom I have heard speaking to the Jews
in their own Aramaic and claiming to be one of their race, you who
have just escaped with your life from a violent mob—are you a
Roman?
For ἔφη. vai, P74 has εἶπεν αύτω· ναἱ, and D gig have εἶπεν,
ε’ιμί.
28. The tribune had grounds for taking Paul’s claim with scepti-
cism. He himself had not been born a citizen but had acquired
(ἐκτησάμην) the citizenship (πολιτείαν); and it had cost him a large
sum, πολλοῦ κεφαλαίου (genitive of price). κεφάλαιον (reflecting
its derivation from κεφαλή, head) had a variety of meanings; it
could mean principal, in various senses, including that which the
English word also may bear, of capital sum (over against interest or
income; so Demosthenes, 27.64 (834)); hence it came to mean sum in
general, e.g. in the same oration, 27.10 (816): κεφάλαιον τέτταρα
τάλαντα καί πεντακισχίλιοι. This meaning is well attested in the
papyri; see MM 342 (further evidence in ND 3.70). Again D(*) has a
slightly livelier version, which according to Begs. 4.284 makes better
sense: καί ἀποκριθείς ὁ χιλ. εἶπεν ἐγώ οἶδα πόσου κεφαλαίου ...
The Old Uncial text, however, may be defended by means of an
observation by Hemer (170), ‘The sale of citizenship was certainly a
feature of life under Claudius (Dio Cassius 17.5-7). Dio’s statement
is placed annalistically in AD 43 (60.17.1) and he says that the
privilege, first sold at great cost, became cheapened later under
Claudius, an interesting parallel with our present passage. This man,
1050 COMMENTARY ON ACTS

whose name Claudius Lysias (23.26) sufficiently confirms his

enfranchisement by Claudius, had presumably gained his rights early
in the reign, and had seen his pride reduced by Claudius’ later
practice, and his remark reflects this.’ Bede (in Ropes, Begs. 3.215)
has the text, Dixit tribunus, tam facile dicis civem romanum esse?
ego enim scio quanto pretio civilitatem istam possedi.
Paul had not had to acquire the citizenship; he had it by birth, ἐγώ
δέ καὶ γεγέννημαι. Cf. Cicero, Ad Familiares 10.32.3, Civis Roma-
nus natus sum. δέ καί here as elsewhere (e.g. 1 Cor. 15.15)
emphasizes the following word: I was born a citizen. ‘Zum höheren
Ansehen des Altbürgers vgl. Ovid, Tristitia 4.10.78f.’ (Schneider
2.327). There has been much speculation on the means by which
Paul’s family, or at least his father, acquired citizenship. If it is true
(Jerome, De Viris Illustribus 5) that the family came originally from
Gischala, they may on account of some services been granted
citizenship by Antony. Ramsay (Cities 161-3, 198) finds evidence
that some Jews became citizens of Tarsus in 171 BC and that some
became Roman citizens under Pompey. All this is guesswork, at least
as far as Paul is concerned; and it must be remembered that there is in
the epistles no hint the writer was a citizen. There is greater
probability in the inference (see above) that the tribune received the
citizenship under Claudius (41-54).
‘Lukas zeichnet Paulus nicht nur als den überzeugenderen Israeli-
ten, sondern auch als den besseren Römer’ (Weiser 612).

29. ἀπέστησαν. Cf. 5.38; literally, they stood away from; in

English, perhaps, they stood back. The word does not in itself mean
that they marched away to another place. For ἀνετάζειν cf. v. 24.
καί ό χιλίαρχος δέ. The position of δέ is unusual. Cf. Mt. 16.12;
Jn 8.16; Acts 3.24. BDR § 447.1(d) suggest ‘und auch’. Not only the
soldiers directly involved but the tribune also was apprehensive. Cf.
16.38.
The aorist participle ἐπιγνούς is in an unusual position, but its
meaning is that which is usual for the aorist participle: he perceived
the truth (or what might be the truth), and having done so was afraid.
The perception preceded the fear, if momentarily.
ἦν δεδεκώς: The periphrastic pluperfect active is uncommon; cf.
8.16; 21.29. It was not illegal for the tribune to arrest a citizen if he
thought him a threat to public order. It was not however permitted
(Lex Porcia; Lex Julia, quoted at length by Haenchen (606f.); Cicero,
Pro Rabirio 4.12; Livy, 10.9.4; Paulus, Sententiae 5.26.1; see above
on ν. 25) to inflict on him the disgrace of being led in public bonds
(in publica vincula).
At the beginning of the verse D, instead of ευθέως σὖν, has τότε.
At the end of the verse 614 syh** sa add καί παραχρῆμα ἔλυσεν
αυτόν which at first seems obvious enough, but causes difficulty
 55. PAUL’S TEMPLE SPEECH AND THE SEQUEL. 22.1-29. 1051

when taken with a second ἔλυσεν in 22.30. Ropes’s note (Begs.

3.215) is, ‘The “Western” addition in v. 29 ... makes ελυσεν αυτoν
και in vs. 30 otiose, and that phrase is omitted by sah. The insertion
before ελυσεν, vs. 30, of πεμψας 614 1611 minn, “misit” hcl*,
suggests that the “Western” text here substituted επεμψεν for
ελυσεν.’ Cf. Clark (379): ‘According to Γ [Clark’s symbol for the
Western text], which has ἔλυσεν αυτόν καί in ν. 30 before
ἐκέλευσεν, Lysias in spite of his fear kept Paul in prison until next
day.’ There is an excellent note in Begs. 4.285: ‘... confusion has
arisen because a superficial reading of v. 30 and an imperfect
visualisation of the facts has suggested that in spite of his fear the
tribune kept Paul “bound”, in an illegal sense, until the next day.
But in vs. 30 ἔλυσε surely refers to release, at least temporarily, from
custody. Had Paul satisfied the Sanhedrin he would have been free.’
It is unlikely that the Sahidic alone contains the original Western
text.
For proof of citizenship see on v. 25. Tajra (86-8) argues at
length, with reference to Lüdemann and Stegemann, that Paul was in
truth a Roman citizen. 'It is quite likely that Paul produced a copy of
his birth registration in order to corroborate his claim to Roman
citizenship’ (85).
Codex Bezae (D), as at present extant, ends with the words τότε
ἀπέστησαν απ’ αὐτοῦ. For the end of d see on v. 20.
 XIV
PAUL AND THE JEWS
(22.30-23.35)

56. PAUL BEFORE THE COUNCIL 22.30-23.11

(30) The next day, wishing to know the truth, that is, what accusation was
being brought by the Jews, [the tribune]1 took off his bonds and ordered the
chief priests and all the Sanhedrin to assemble. He brought Paul down and
caused him to stand before them.
(1) Paul fixed his eyes on the Sanhedrin and said, ‘Brothers, up to this day
I have conducted myself before God with an entirely good conscience.' (2)
The high priest Ananias ordered2 those who were standing beside him to
strike him on the mouth. (3) Then Paul said to him, ‘God will strike you, you
whitewashed wall. You—do you sit there to judge me according to the Law,
and do you contrary to the Law command me to be struck?’ (4) Those who
were standing there said,3 ‘Do you insult God’s high priest?’ (5) Paul said, ‘I
did not know, brothers, that it was the high priest, for it is written, You shall
not speak evil of a ruler4 of your people.’ (6) Paul5 noticed that one part of
them was made up of Sadducees, the other of Pharisees, and cried out in the
Sanhedrin, ‘Brothers, I am a Pharisee, the son of Pharisees; it is for the hope
of6 the resurrection of the dead that I am on trial.’ (7) When he had said this
there arose a conflict of Pharisees and Sadducees, and the company was
divided. (8) For Sadducees say that there is no resurrection, nor angel nor
spirit, while the Pharisees confess belief in both.7 (9) There arose a loud
outcry, and some of the scribes on the side of the Pharisees rose up and
contended, saying, ‘We find no evil in this man; and8 what if an angel spoke
to him, or a spirit?’ (10) There was a great conflict, and the tribune became
afraid lest Paul should be tom in pieces by them. He ordered the detachment
to come down, seize him out of their midst, and take him into the barracks.
(11) In the following night the Lord stood by [Paul]9 and said, ‘Be of good
courage, for as you have testified in Jerusalem to the things concerning me so
must you bear witness in Rome too.’
1Greek, he.
2NEB, his attendants.
3RSV, Would you revile; NEB, Would you insult.
4NEB, the ruler.
5NEB, NJB, was well aware.
6RSV, hope and.
7RSV, in them all; NEB, them; NJB, all three.
8NEB, perhaps an angel; NJB, suppose a spirit.
9Greek, him.
1052
 56. PAUL BEFORE THE COUNCIL. 22.30-23.11. 1053

Bibliography
B. J. Bamberger, JBL 82 (1963), 433-5.
C. K. Barrett, FS Furnish 161-72.
C. H. Cosgrove, NovT 26 (1984), 168-90.
D. Daube, JBL 109 (1990), 493-7.
L. Finkelstein, HThR 22 (1929), 239.
K. Haacker, NTS 31 (1985), 437-51.
J. Jeremias, as in (55).
S. T. Lachs, Gratz College Annual of Jewish Studies 6 (1977), 35-42.
S. Legasse, as in (54).
J. Lührmann, JSNT 36 (1989), 75-94.
H. Strathmann, TWNT 6.525-8.
G. G. Stroumsa, RB 88 (1981), 42-61.
A. A. Trites, as in (54).
B. T. Viviano and J. Taylor, JBL 111 (1992), 496-8.
J. A. Ziesler, NTS 25 (1979), 146-57.

Commentary
This short paragraph contains a number of historical difficulties,
fully, but with some exaggeration, discussed by Haenchen
(611-16).
In v. 30 it is said that the tribune loosed, ἔλυσεν, Paul. If already
on the previous day (22.29) the thought that he had bound a Roman
citizen gave the tribune ground for fear, why did he wait for 24 hours
before rectifying his error? There is no serious problem here, nor any
reason for elaborate discussion of the question whether the binding
referred to (αὐτὸν ἦν δεδεκώς, 22.29) was that of the original arrest
or that preparatory to scourging and whether heavy or light bonds
were in mind. Luke is not writing a police report; he wishes to
indicate that Paul appeared before the Council as a free man; the
precise point at which his bonds were removed was of no interest or
importance. We note however that Luke is writing his own story; he
is not copying out a written source.
The next question invites a similar reply. The tribune ordered
(ἐκέλευσεν) the Sanhedrin to meet. Did he, did even the Governor,
have the authority to do this? Schneider (2.330) thinks that he had
not; evidence one way or the other is hard to come by. It is also very
unlikely (Conzelmann 127) that a Roman military officer would
simply hand over a Roman citizen to a local, non-Roman, tribunal. It
is however by no means unlikely that a Roman officer would require
a local body to serve as a fact-finding (and fact-understanding)
commission, allowing it to make inquiries and present a report to
him. This leads to a further double problem . Would a Roman leave a
1054 COMMENTARY ON ACTS

fellow Roman to stand unprotected before an alien body, obviously

moved by strong passions? Would a Roman—undoubtedly
unclean—be allowed to be present at a meeting of the Sanhedrin?
Verses 30 and 1 certainly suggest a formal meeting of the Sanhedrin,
and v. 10 suggests (but does not prove) that Claudius Lysias was
present. If however the meeting was an informal one the difficulties
are eased. Again: Luke was not a historian of a formal kind; he was
introducing a running three-cornered battle, in which the parties were
Jewish officialdom, Roman officialdom, and Paul (who was both
Jewish and Roman). He knew that this triangular debate took place
and put its beginning on a high level, at least as far as Judaism was
concerned. In the ensuing verses he will bring out a number of major
points which will reappear from time to time in the course of the
story. These are more important to him than the technical details. We
see what Luke wishes us to see: Paul, Jerusalem, and Rome, standing
face to face. And in fact they did so stand.
It is highly inappropriate that, in a trial, the accused should open
the proceedings, as Paul does in v. 1. This is a correct observation,
but as a complaint against Luke’s story it is specious. We have good
reason to doubt whether this was a formal trial, or indeed an informal
one; and we have no reason to conclude that Paul’s words in this
verse were the first that were uttered. It is not inconceivable that he
was invited to make a statement, and that Luke, who did not have
unlimited space, decided to omit what would be no more than a stage
direction. He is not writing as a professional historian; but we may at
least say that he is not writing like a fool. In this verse he puts on
Paul’s lips the proposition that he will maintain up to the end of the
book: Paul has acted in good faith and good conscience. He is not
only a good Christian but a good Jew and a good Roman; of course
he has his own understanding (which is in Luke’s mind correct) of
what being a good Jew and a good Roman means.
The prisoner ought not to have begun the proceedings (but Luke
knows that in the whole affair the initiative was his) and the Chief
Justice ought not to have initiated a physical attack upon one as yet
untried, much less proved guilty of a fault (but Luke knows, or
believes, that the official Jewish party was constantly guilty of
injustice and violence). Again Luke is describing a scene which will
enable the reader to see in advance how the whole story will proceed.
This continues in the next verse. Paul is reproved by commentators:
Being reviled, we bless, did he say? What Luke wishes us to see (and
doubtless he was right) is the courage with which Paul faced official
opposition, injustice, and violence. When many a man would cringe,
Paul answers back, and points out (this is more important than the
curse—if curse is the right word) that the Jewish judge is himself not
observing the Law that he is appointed to administer. This is a point
that Luke emphasizes throughout his book; his complaint against
 56. PAUL BEFORE THE COUNCIL. 22.30-23.11. 1055

Jews is not that they are Jews but that they are not good Jews. The
better they are the greater the sympathy they will show to the
Christians. This point will shortly reappear in another dress.
The next verse returns to the theme of the relation of Paul to
Judaism, viewed from a different angle. Paul’s ‘I did not know that it
was the High Priest’ has been variously evaluated. Did he speak
ironically? Did he nervously withdraw the aggression he had shown
in v. 3? The point seems to be that he will not knowingly violate a
rule of Scripture, however strong the provocation. If Exod. 22.27
forbids him to speak evil of the High Priest, however badly the High
Priest behaves (and Paul says nothing to moderate his judgement of
the High Priest’s action), he will abstain from doing so. He is a good
Jew, a Christian Jew.
Paul’s next contribution to the debate has also evoked condemna-
tion. The historical Paul of course knew (and did not need to observe)
that there was in the Sanhedrin a mix of Pharisees and Sadducees.
Setting the two parties against each other was a good way of
escaping from both. He is a Pharisee, he asserts, and it is the key
issue of the resurrection of the dead that is at stake. Historically it
was not a very noble piece of self-defence, nor was it one whose
effect was likely to last; it would not take the Pharisees long to notice
that the resurrection that Paul was proclaiming was not exactly what
they had in mind and were prepared to defend. But again Luke has
achieved the goal, or goals, that he seeks. He has made clear (a) that
Paul was a faithful Jew; (b) that he was a Jew of the Pharisaic kind,
so that Pharisees are nearer than Sadducees to Christians and should
be encouraged; (c) that the central issue for Christians is resurrec-
tion—the resurrection of Jesus; (d) that the Jews are too divided to
take effective action. As the story proceeds he shows in v. 10 that
Rome is Christians’ defence against official Judaism, and in v. 11
that the Lord approves of what Paul has done and guarantees his
ultimate arrival in Rome.
The whole passage is certainly Luke’s work, but he was probably
using a foundation of traditional material as he constructed a
narrative that would foreshadow the themes that control the rest of
his book.

30. τῆ ἐπαύριον occurs 10 times in Acts (Mt., once; Mk, once; Jn,
5 times); this is no doubt due mainly to the narrative character of the
book. Here τῇ ἐπιούση (1241 2495 pc; cf. τή ἐπιούση ἐπαύριον,
614) may be correct—the less usual word could have been assimi-
lated to that which is more common.
The tribune, it seems, wishes to use the Sanhedrin in an advisory
capacity; he wishes γνῶναι τό ασφαλές. The same words occur in
21.34. ασφάλεια and cognates constitute a Lucan group (Luke-Acts
7 times; rest of the NT, 4 times). The indirect question (τί
1056 COMMENTARY ON ACTS

κατηγορεῖται... ), introduced by the article τό (see on 4.21), stands

in parallel with τό ασφαλές. He wished to know the facts, namely,
what accusation was being brought... Page (229) is probably right in
taking τί to be nominative and the subject of κατηγορεῖται cf.
Thucydides 1.95.2, αδικία πολλή κατηγορεῖτο; Sophocles, Oedipus
Tyrannus 529. Alternatively, Why, or of what, he was being
accused.
Before ἔλυσεν, 614 pc syh** have πέμψας, possibly a relic of a
Western substitution of ἔπεμψεν for ἔλυσεν, prompted by ἔλυσεν in
22.29; see on that verse, and see Ropes (Begs. 3.215) and Dele-
becque (109). After spending the night in custody, in part, perhaps,
for his own safety, Paul was a free man. Conzelmann regards it as
improbable that the tribune, already alarmed (ἐφοβήθη, 22.29) when
he discovers that he has arrested a Roman citizen, should leave him
in bonds. Schweizer (Beiträge 78) notes the view of Wendt (Die
Apostelgeschichte (1899), 359) that the whole narrative of
22.30-23.10 may have grown out of ἀκριβέστερον in 23.15.
ἐκέλευσεν can only mean that the tribune ordered the Jewish
Council to meet. It is doubtful (see above) whether he would have the
authority to do this (Haenchen 612; Schneider, 2.330), but the
Romans may well have used the official body as a means of
obtaining information about Jewish matters, and in this case the
Council would be glad to be called out; it would give them an
opportunity to formulate and present a case against an adversary.
Wilson (Law 67) thinks that the tribune intended to be and was
present at the meeting; Luke does not say this, and, though it would
be consistent with v. 10, it cannot be regarded as certain. He may
have been content to receive a report. The report that reached him in
ν. 10 was not what he hoped for.
αρχιερείς: see on 4.5. συνέδριον: see on 4.6. καταγαγών, he
brought Paul down—from the Antonia (see 21.31). For the Sanhe-
drin’s place of meeting see on 4.5, and note especially StrB
1.997-1001; NS 2.223-5.
ἔστησεν εις αυτούς is a pregnant expression: the tribune brought
Paul into their gathering and caused him to stand before them.

1. ἀτενίζειν (Lk., twice; Acts, 6 times with εἰς, 4 times with

dative; 2 Corinthians, twice) is a Lucan word. If Knox’s observation
(Hell. El. 16) about the choice of vowels and consonants in Paul’s
words in this verse and the following two verses (vv. 1-3) is correct
the paragraph has been carefully composed, presumably by Luke.
Paul’s ‘opening words have 18 long vowels and the harsh consonants
are avoided (κ, 0; π, 3; τ, 5; ξ, 0). But when the High Priest orders the
bystanders to strike him on the mouth St. Paul loses his temper and
we have only 15 long vowels out of 40 (allowing τύπτειν to be
naturally short), and 17 harsh consonants (κ, 7; π, 3; τ, 7).’ Without
 56. PAUL BEFORE THE COUNCIL 22.30-23.11. 1057

using this observation Schmithals (206) reaches a similar result:

‘Sprache, Stil und Intention sind durchaus lukanisch, so dass
22.30-23.11 ganz der schriftstellerischen Komposition des Lukas
entsprungen sein muss.’ Schneider (2.329) describes the passage as
an Einzelanekdote.
ἀτενίσας, aorist participle: Paul fixed his eyes on the Sanhedrin
and said ... ἄνδρες αδελφοί. Cf. 1.16. If the address is not purely
formal it may mean that Paul regards the members of the Sanhedrin
as equals, not as judges or superiors.
πάση συνειδήσει αγαθή. For συνείδησις; (= τό συνειδός) see C.
Maurer in TWNT 7.847-918—for background rather than for Acts
itself. Paul claims an entirely good conscience; that is, his conscience
is clear in regard to all his behaviour. For Paul’s conscience Stählin
(288) compares 24.16; 2 Cor. 1.12; 2.17; 4.2. 1 Cor. 4.4 should be
added. ‘Even his persecution of the church had been carried out with
good conscience; it was, as he thought, his bounden duty (cf. 26.9)’
(Bruce 2.424). See also Haenchen (609); ‘Diese Theologie des
“guten Gewissens’’ wird in der nachapostoloschen Literatur
beliebt.’ He cites 1 Tim. 1.5,19; 3.9; 2 Tim. 1.3; 1 Pet. 3.16,21; Heb.
9.14; 13.18. This is correct as far as it goes, but it is doubtful whether
Haenchen does full justice to the present passage. ‘Es geht um die
bürgerliche Wohlanständigkeit ... nicht um eine ‘ ‘Theologie des
‘guten Gewissens’ ’’ ’ (Schille 425). Dr Thrall’s article on conscience
(NTS 14 (1968), 118-25) deals only with the Paul of the letters. Paul
(Luke) here means simply, T have always done what (at the time) I
thought was right.’ But see further FS Furnish, 165f.
πολιτεύεσθαι meant originally to live as a free citizen of a πόλις,
or to take part in the government of a πόλις; it came however to have
the more general meaning, to conduct oneself (in private as well as in
public affairs). This use seems to have been predominantly Jewish
and Christian; see H. Strathmann, TWNT 6.525-8. This may be
illustrated by Phil. 1.27; 2 Macc. 6.1; 11.25; 3 Macc. 3.4; 4 Macc.
2.8, 23; 4.23; 5.16; Philo, De Virtutibus 161; Josephus, Life 12;
Aristeas 31; 1 Clement 6.1; 21.1; also by the synagogue inscription
from Stobi (most conveniently in E. L. Sukenik, Ancient Synagogues
in Palestine and Greece (1934), 79-81), where the father of the
synagogue (ὁ πατήρ τής ἐν Στόβοις συναγωγής) claims
πολειτευσάμενος πάσαν πολειτείαν κατά τόν ’Ιουδαϊσμόν. The
background of this Judaeo-Christian development is well given by
Page (229): ‘The ideal Jewish state was a state under the direct
government of God—a theocracy. Paul says “You accuse me of
speaking against the Jews, the Law, etc.; I answer that in the sight of
God, the ruler and lawgiver of the Jewish nation, I have acted as a
good citizen.” ’ Cf. Delebecque (109f.), ‘... que ma vie, à moi, est
celle d’un citoyen de Dieu’. ‘Le parfait grec donne toute sa force à
cet état de citoyen à la conscience directe en face, non de la Loi, mais
1058 COMMENTARY ON ACTS

de Dieu.’ But Delebecque recognizes that πολιτεύεσθαι normally

means simply ‘to comport oneself’. In the present passage the
distinctive feature is the dative τῷ θεῷ. There is a partial parallel in 2
Macc. 6.1, τοῖς τοῦ θεοῦ νόμοις μὴ πολιτεύεσθαι, and a letter in
UPZ 144.13f. (cf. 110.78), if this is rightly restored by LS (1434; s.v.
VI), πρὸς [τούς θεούς] ὁσίως καί δικαίως (LS do not indicate the
place of the verb). This papyrus will also, presumably, show a non-
Jewish use of the verb in a religious sense; for this see also
Demosthenes 18.4 (226), 8 (227). Paul means that his life has been
lived in the sight of God, and in obedience to God. The intention may
be to supplement the subjective appeal to conscience with an appeal
to the testimony of God himself. Cf. Rom. 9.1.
ἄχρι ταύτης τής ημέρας implies ‘up to and including my becom-
ing and living as a Christian’. Paul was still as a Christian an honest
Jew, living as all Jews ought to live.
At the beginning of the verse there are three different forms of the
text which do not differ from one another in meaning:
ὁ Παύλος τῷ συνεδρίῳ Ρ74Ψ <a>
τω συνεδρίῳ ό Παύλος A C Ε 33 945 1739 1891 pc latt
Παύλος τω συνεδρίφ B 36 453 614 2495 pc
It is not easy to see why NA26 choose the reading which lacks the
earliest attestation.

2. 'Ανανίας, Cf. 24.1; appointed by Herod of Chalcis. ‘Ananias

son of Nedebaeus (c. AD 47-59), Ant. 20.103. Cf. 20.131; War
2.243 ... As a result of his wealth, Ananias remained an influential
man even after his deposition, but was unpopular on account of his
greed, Ant. 20.206-13. He was murdered by the revolutionaries at
the beginning of the war, War 2.429, 441-2’ (NS 2.231). See also
Jeremias (Jerusalem 378, and the references there); Hemer (170f.
with a possible explanation of Paul’s failure to recognize him—see
below on v. 5). It may be that this account of Ananias is less than fair.
According to Josephus, Ant. 20.205 so far from being unpopular he
enjoyed the goodwill and esteem of the citizens, ἧν γάρ χρημάτων
ποριστικός. As Feldman (LCL Josephus 9.498f.) points out, all
depends on whether ποριστικός refers to the getting or giving of
money; the popularity referred to suggests that Feldman is right with
‘able to supply them with money’. It is true however that the charge
of greed is supported by b. Pesahim 57a. Again, Ant. 20.206 supports
a charge of violence only in Ananias’ unwillingness, or inability, to
control his subordinates, and we are scarcely prepared for the
violence used on Paul, except so far as a turbulent society breeds this
sort of perversion of justice. This context suggests (but its historical
value remains to be assessed) that the High Priest was indignant at
Paul’s claim to be a conscientious Jew; and so he might well be, for
 56. PAUL BEFORE THE COUNCIL. 22 30-23.11. 1059

Paul’s conscientiousness would have destroyed Judaism as currently

understood. This however is not the impression of the High Priest
that Luke wishes to give. An alternative explanation of the High
Priest’s action is given by Blass (241): ‘quia locutus erat [Paulus]
non interrogatus’.
τοῖς παρεστῶσιν—probably in a general, non-technical, sense;
not ‘his assessors’. For the blow and response cf. Jn 18.22f.
For ἐπέταξεν, P74 has παρήγγειλεν; C 36 453 945 1739 1891 pc
have ἐκέλευσεν. This seems to be mere pointless Western verbal
variation.
Cf. Acta Andreae et Matthiae 26 (L.-B. 2.1.104), ἔλεγεν (the
devil) τοῖς ὄχλοις. Τύπτετε αὐτοῦ (Andrew) τό στόμα ἵνα μή λαλῇ.
If there is any dependence here it is clearly on the part of Acta A. et
Μ.

3. Paul ‘répond par une explosion de gutturals’ (Delebecque 110;

cf. Knox on v. 1). He answers the assault made upon him: τύπτειν σε
μέλλει ό θεός. The form of the sentence (with μέλλειν) suggests a
statement of (future) fact: At some future time God will smite you. It
is apparently so taken by Stauffer (Theologie 188); God judges the
world in righteousness, and such flagrant unrighteousness is sure to
meet with its due reward. Luke may however have in mind the actual
fate of Ananias (Josephus, War 2.441: ὅ τε ἀρχιερεύς Άνανίας περί
τὸν τῆς βασιλικῆς αὐλῆς εὔριπον διαλανθάνων ἁλίσκεται καί
πρὸς τῶν λῆστών αναιρείται). But the intention may be to express
an imprecation (= τύπτοι). If this is so, cf. Shebuoth 4.13: [If a man
said], ‘God smite thee’ or ‘Thus may God smite thee’
this is the ‘adjuration’ that is written in the Law. The
Torah reference is given by Danby as Lev. 5.1, by StrB (2.766) as
Deut. 28.22. The two interpretations of Paul’s words are not
incompatible with each other. Beginning from confidence in divine
vengeance the speaker may rejoice in and seek to hasten its
fulfilment; a curse may be uttered with the conviction that it
coincides with God’s judgement and is sure to take effect. Opinions
differ. ‘It is not a curse, as the Greek context makes plain enough, but
rather a reproof coupled with the announcement of a punishment’
(Calvin 2.227). ‘Eine geläufige jüdische Verwünschungsformel'
(Schmithals 207). Not unimportant is the observation of Blass (241),
‘inde μέλλει, qui fut. erat non τύψω sed πατάξω’. LS 1835f. confirm
that Tn Att. and LXX the fut. and aor. are supplied by πατάσσω’, but
also quote an Attic future τυπτήσω from Aristophanes, Plato, and
Demosthenes.
τοῖχε κεκονιαμένε. The precise point of this image is not clear. It is
safe to reject Bengel’s neat ‘Fortasse etiam canos capillos aut albam
vestem habuit’ (474). It may depend on the image of the white-washed
or plastered wall in Ezek. 13, especially 13.14f.: an insecure wall,
1060 COMMENTARY ON ACTS

though covered with whitewash and therefore apparently sound, will

easily and quickly fall. Or the whitewashing may suggest hypocrisy,
as at Mt. 23.27 (but note that this saying appears in a different form in
Lk. 11.44). Bauemfeind (257) refers to the ‘Scheingerechtigkeit’ of
the High Priest; but according to Preuschen (132) the image is ‘durch
Hinweis auf Mt. 23.27 nicht erklärt’. The Ezekiel image is picked up
in CD 8.12: those who build the wall and
smear it with plaster. The LXX appears to have taken to be
derived from (πεσεῖται).
καὶ σύ. καί in a question expresses indignation or astonishment.
κάθη: this second person singular is derived from κάθομαι; the
LXX still uses the regular κάθησαι (from κάθημαι). So Μ. 2.206.
The question runs through to the end of the verse: And do you sit in
judgement on me ..., and do you command ...?
κατά τόν νόμον. The court had been convened by the Roman
tribune, but evidently (at least in Luke’s mind) considered itself to
have met in accordance with the Law to try an Israelite accused of
apostasy.
παρανoμῶν. See Lev. 19.15 which may be regarded as requiring
fair play for the person accused. Sanhedrin and Makkoth go out of
their way to emphasize this requirement, though there is no special
prohibition of striking the accused. Paul had not at this stage been
formally accused of crime, much less found guilty. Demosthenes
uses παρανομεῖν, συνείδησις, πολιτεύεσθαι, κονιᾶν: ‘ ... est-ce un
hasard?’ asks Delebecque (110). Not exactly chance; Demosthenes,
and, at this place, Luke are both concerned with the operation of
courts. It is not surprising that their vocabularies overlap.
κελεύεις με τύπτεσθαι. The passive infinitive is used where
classical usage would have preferred the active; see Μ. 3.138. It is
unlikely that this is in any direct sense a Latinism, though the
construction is Latin rather than Greek. Cf. 22.24.4

4. οι παρεστῶτες. See v. 2. These words can be punctuated with

either a full stop or a question mark. It makes little or no difference to
the sense.
According to Begs. 4.288, ἀρχιερεύς is used in the LXX only at
Lev. 4.3. To this must be added Josh. 22.13, and, according to some
MSS, Josh. 24.33; 3 Kdms 1.25; 1 Chron. 15.14. It is frequent in 1
Esdras, 1 and 2 Maccabees, and seven times each in 3 and 4
Maccabees. Begs. says also that ‘the addition “of God” (Lord) is
occasionally found with “priest” (e.g. 1 Kings (=3 Kdms) 2.27) to
denote the high priest’. In the passage cited, τού κυρίου is a
rendering of Cf. 1 Kdms 14.3, ἱερεὐς τού θεού, See
also Gen. 14.18, Melchizedek ἱερεύς τού θεού τού ὑψίστου,
Probably here in Acts we should recognize ἀρχιερεύς as in
the main a post-canonical development and the addition of τού θεού
 56. PAUL BEFORE THE COUNCIL. 22.30-23.11. 1061

as intended to heighten the offence: it is no common man whom you

insult, but the high priest who has been appointed by and thus
represents God himself. Schille (426) writes, ‘Die Hörer mis-
verstehen die Prophetie als Fluch.’ Perhaps they understood it as a
curse, and did so correctly. In any case, to tell the High Priest, God
will in due course send you to hell, or even, God will have you
murdered by terrorists, could hardly be regarded as complimentary.
Cf. the reaction to Jesus’ reply to the High Priest in Jn 18.22.
Cyprian (Epistles 3.2; 59.4; 66.3) quotes the verse in the form, Sic
insilis in sacerdotem dei maledicendo? This may represent either a
Western, or perhaps Cyprian’s own, addition, made in preparation
for the quotation that follows in v. 5, Principem plebis non mal-
edices. It was hardly possible to do this with the Greek οὐκ ἐρεῖς
κακώς. Ropes (Begs. 3.215) speaks of ‘some kind of intensifying
expansion’.
Conzelmann (128) says of the incident, ‘Zugrunde liegt eine vage
Nachricht, nicht ein geschichtstreuer Bericht.’ It is however in the
next verse that the difficulty becomes acute.

5. οὐκ ἤδειν ... ὅτι έστίν άρχιερεύς—one of the most puzzling

sentences in Acts. Apparently Paul withdraws his disrespectful
remark; he did not know that the person who had ordered him to be
struck was the High Priest. Is this conceivable? In the past (9.1f.)
Paul had been a trusted member and agent of the highpriestly party;
true, this was some years back and there had been a change of high
priest, but Ananias can hardly have been unknown to Paul—though
Hemer (171) makes the interesting suggestion that ‘Paul’s previous,
briefly noted visit to Palestine (18.22) was ostensibly in summer 52,
likely to have been the very time when Ananias had been sent in
chains to Rome.’ Who other than the High Priest should have
presided over the judicial gathering? Begs. 4.288 points out that the
Mishnah regulation that the high priest should always preside may be
an idealized constitution drawn up after the Sanhedrin and High
Priest had ceased to exist. The same passage continues, ‘It is also
possible that Paul merely meant that he did not know who had given
the order’; cf. Blass (241): ‘Errat Paulus, putans aliquem ex assessor-
ibus illa locutum esse.’ Cf. Chrysostom, Homily 48.2, and Knowling
(466): T wist not that it was the high priest (who spoke).’ Those who
think that Paul suffered from defective sight have sometimes found
evidence for it here: not a convincing explanation. A more widely
accepted explanation is that Paul was speaking ironically. So e.g.
Calvin (2.229): T myself agree with Augustine, and have no doubt
that this excuse is ironical.’ ‘Paul manie l’ironie’ (Delebecque 110);
also Marshall (364). But Haenchen (610) is probably right in arguing
that the address αδελφοί (v. 1) and the quotation of Exod. 22.27
show that this is not irony. Bauernfeind (257) finding that irony is
1062 COMMENTARY ON ACTS

improbable concludes that we know too little to be able to explain the

text. This may be the right conclusion to draw, but there are other
possibilities. Conzelmann (129) observes that Paul withdraws; this is
not the moment for final conflict. It would be just possible to
translate, I did not know that there was a high priest; that is, that
under Roman occupation and restriction it was possible for a high
priest to function effectively, that is, to be an effective high priest.
This takes no account of the OT quotation. Alternatively again, Paul
shows his Gesetzestreue (Pesch 2.243); he shows himself to be a
better Jew than the members of the Council (Schmithals 207). This
may be held as historically true, or as Luke’s non-historical repre-
sentation of Paul and his insistence on the continuity between
Christianity and Judaism (Schweizer, Beiträge 75). Finally, it is ‘not
inconceivable that he remembered his own writing λοιδορούμενοι
εὐλογουμεν, διωκόμενοι ἀνεχόμεθα (1 Cor 4.12)’ (Begs. 4.288).
And there is Bengel (474): Nesciebam, modeste dictum, interpretere,
non veniebat mihi in mentem.
The multiplicity of explanations shows that the historical question
is not an easy one. Paul (in Luke’s view—and in his own?) was or is
represented as being a good Jew; he believed that his (Christian)
version of Judaism was the one that God intended. But there was
much more to say, and he would have been careful to present a
balanced view on so important an occasion. There is historical
material behind this paragraph, but it is deep and remote, and it is
wise to recognize with Bauemfeind that there are limits to our power
to explain.
That it would be wrong to address the high priest in the manner of
v. 3 is proved by the OT. The quotation of Exod. 22.27 agrees
exactly with the MS A of the LXX; this however has probably been
conformed to the NT, and the better LXX text has άρχοντες (plural),
and κακώς έρεις. The Hebrew has the singular noun Paul
could have appealed to an interpretation of the passage that appears
in Mekhilta 103a. The Hebrew has and this is explained as
follows: ‘Was wollen die Worte besagen, “In deinem Volk”? Wenn
sie [die Richter und Obersten] die Sitte deines Volkes innehalten
[andernfalls geniessen sie den Schutz von Exod 22.27 nicht]’ (StrB
2.766f.). Paul could then have argued that a high priest who
commanded illegal assault forfeited the protection of the Law. Luke
was probably unaware of this interpretation—or perhaps preferred to
represent Paul as unquestioningly obedient to the OT.
For κακῶς λέγειν cf. Aristophanes, Acharnions 503: ξένων
παρόντων τὴν πόλιν κακώς λέγω.
Robertson (422) finds in άρχοντα ... κακώς an iambic trimeter,
but, as BDR § 487.2, n. 6 point out, three of the feet are spondees.

6. γνούς: the aorist participle is a further indication of Lucan

 56. PAUL BEFORE THE COUNCIL. 22.30-23.11. 1063

composition; Paul would not have to notice but would know in

advance the mixed membership of the court.
If τό εν ... τὸ δέ ετερον is a Semitism (Black, AA 108) it is a very
mild one. Cf. Μ. 2.438. τό εν ... τό εν would correspond more
closely to Semitic idiom ( ... ). Σαδδουκαίων, Φαρισαίων,
are predicative partitive genitives (BDR § 164.1c, n. 1): one part
consisted ofS. the other of Ph. On Sadducees see 1.219, on Pharisees
1.292. There was always (before AD 70) a potential division in the
Sanhedrin, which could be exploited by one who was prepared to
align himself with one party and to enlist its support.
ἄνδρες αδελφοί repeats the address of v. 1.
Paul proclaims himself to be a Pharisee, and a son of Pharisees
(Φαρισαίων); E <a> syh have Φαρισαίου, an easier reading, referring
simply to Paul’s father. υἱός Φαρισαίων will refer either to a line of
ancestors all belonging to the Pharisaic party, or will represent the
Semitic use of ‘son of’, describing essential character: Paul claims to
be the quintessential Pharisee, and thus that it was essentially
Pharisaism that he was contending for. For Paul as a Hillelite
Pharisee see J. Jeremias (‘Paulus als Hillelit’), FS Black (1969),
88-94; K. Haacker (‘War Paulus Hillelit?’) (reference on 22.3).
Büchsel (Theologie 13, 177) notes the consequence that Pharisees
were to be found not only in Palestine but also in the Diaspora (see
however Rapske, Book of Acts 3.95-7). In the Diaspora they would
constitute a group determined theologically rather than sociolog-
ically (for the sociological setting of Palestinian Pharisaism see the
fundamental work of L. Finkelstein, The Pharisees (1962); also NS
2.381-403, with the excellent bibliography on 381f.). Paul claims
that it is as a Pharisee, in the interests of Pharisaic doctrine, that he is
standing trial—κρίνομαι. Strictly speaking, Luke’s narrative is not
an account of a trial; the tribune is trying to ascertain facts (22.30).
One can understand however that such an event might seem like a
trial; and it suited Luke’s purpose to make it seem like one. The
charge in the seeming trial is expressed in two terms: περί ἐλπίδος
καὶ ἀναστάσεως νεκρών. Already Bengel (475), followed by many
interpreters, saw this as a hendiadys: the two propositions were in
fact one. Zerwick (§ 460) renders ‘Propter spem in resurrectionem’ ;
cf. BDR § 442.9b. Conzelmann (129) notes that by expressing
himself in this way Paul (Luke) avoids a double genitive. Preuschen
(133) points out that καί is missing in the Philoxenian Syriac; rightly
he thinks. For Paul’s assertion see further below. It is certainly true
that the resurrection of Jesus was at the heart of his faith (e.g. 1 Cor.
15.13-19), and that if the Sadducees’ belief (v. 8) that resurrection
did not happen were accepted Paul’s Christian belief would be
denied; but the Pharisees’ belief in resurrection was in general terms,
and did not carry with it the specific, and unique, assertion that Paul
made about Jesus. The line of argument is not greatly altered if hope
1064 COMMENTARY ON ACTS

and resurrection are distinguished; thus Haenchen (615): ' ... ich
stehe vor Gericht wegen der (messianischen) Hoffnung and der
Auferstehung der Toten’.
Of Paul’s argument Calvin (2.230) writes, ‘ ... that was not far
from lying’; and has much difficulty in dealing with the problem.
Paul’s speech ‘wirkt wie eine Taktik, die eines Apostels unwürdig
ist’ (Stählin 288, who goes on to point out (289) that Paul’s
understanding of resurrection is different from the Jewish). Some
have seen in 24.21 a confession of error or fault; this is denied by
Bruce 1.411; it is squeamish to blame Paul for his approach. More
fully Bengel (475): ‘Hic in bonam partem valuit illud: Divide et
impera. Non usus est Paulus calliditate rationis aut strategemate
dialettico; sed ad sui defensionem simpliciter eos invitat, qui propius
aberant a veritate.’ See also the discussion of ‘Christians and
Pharisees’ in Maddox (40-42). Maddox distinguishes between
Luke’s rejection of the Pharisees’ position with regard to the Law
and his acceptance of their doctrine of the resurrection. ‘Jervell has
rightly seen that the “bond of sympathy”-theme has to be not with
the relation of Christians to the Roman state but with relations
between Christians and Jews. It is a doctrinal matter: but the doctrine
in question is not the Law, as Jervell supposes, but the resurrection’
(41). There is a further pointer in Weiser (617): ‘Damit ist nach luk
Verständnis die Hoffnung Israels auf die endzeitliche Totenaufer-
stehung (24.15, 21), von der Paulus sagt, dass sie sich als Erfüllung
göttlicher Verheissungen (13.32ff.; 26.6) in der Auferstehung Jesu zu
verwirklichen begonnen habe.’ If the resurrection of Jesus is defined
as an anticipatory first stage of the general resurrection at the last
day, there is enough agreement between the Pharisees and the
Christians to satisfy Luke, and to justify what Schille (426) describes
as neither history nor a naive anecdote but a piece written with Lk.
10.3 in mind. (How much better if one could refer to Mt. 10.16!).
Whether Paul himself would have been satisfied, would have used
the words put in his mouth, is a different matter. Paul refers to his
Pharisaic background in a different way in Phil. 3.5f.
In this verse the second ἐγώ is omitted by B gig sa.

7. λαλήσαντος (C Ψ <a>) and εἰπόντος (P74 A E 33 323 945

1739 1891 2464 pc; and the text of NA26) are both improvements on
λαλοῦντος (B pc), and are to be rejected; the dissension would
naturally arise, as the copyists perceived, after Paul had completed
his brief remark rather than while he was making it. B however is
surely wrong in supporting ἐπέπεσεν (B*) or ἔπεσεν (B 2138 pc sy)
against ἐγένετο. A στάσις can hardly be said to fall, even metaphor-
; the copyist may have subconsciously remembered 10.44.
στάσις is a strong word; see v. 10. τό πλήθος is the whole
company present. The division was of course what Paul intended.
 56. PAUL BEFORE THE COUNCIL. 22.30-23.11. 1065

There is truth in both Conzelmann (129), ‘... die Hoffnungslos-

igkeit des Judentums ... es ist in sich gespalten’ and Haenchen (615),
the truth ‘dass zwischen Juden und Christen die Brücken nicht
abgehoben sind’. Pesch (2.245), quoting Roloff (327), goes too far in
the claim that Judaism ‘seine innere Identität und äussere Glaubwür-
digkeit verloren hat’. Paul is claiming to be a Pharisee and to believe
what Pharisees believe. Pesch also (2.241f.) notes the repetition of
στάσις in v. 10 and argues that vv. 6-9 are a Lucan insertion in a
source. Lüdemann (252) argues with greater probability that vv. 7-9
are redactional.

8. For the Pharisees see on v. 6; for the Sadducees see R.

Leszynski (Die Sadduzäer (1912)) and NS 404-14, with the biblio-
graphy on 381f. Here Luke concentrates on the matter in hand. The
verse contains two notable problems.
The statement that the Sadducees say that there is no resurrection
is not a problem. It is a widely recognized fact about the Sadducees;
see Mt. 12.18 (and parallels); Josephus, War 2.165 (ψυχής τε τήν
διαμονὴν καί τὰς καθ’ ᾃδου τιμωρίας καί τιμάς ἀναιροῦσιν), and
other passages. Cf. Sanhedrin 10.1.
But the statement that the Sadducees do not believe in angels or
spirits, if taken in its most obvious sense, has no parallel, and indeed
can have none, for the Sadducees accepted the authority of the
written Torah, and the Pentateuch contains many references to
angelic and spiritual beings, in whose existence the Sadducees must
have believed. ‘ ... wirkliche radikal Bestreitung aller Engel- und
Geistermetaphysik liess sich mit dem AT aber doch schlechthin nicht
vereinigen! Es gibt auch kein einziger Zeugnis sonst, das diese
Angabe der AG bestätigte’ (Bauemfeind 255). Bauemfeind, assum-
ing that the statement is correct, finds the origin of this scepticism in
Hellenistic free-thinking; cf. StrB (2.767), ‘Die Leugnung von
Engeln und Geistern seitens der Sadduzäer lässt sich aus jüdischen
Quellen nicht belegen, entspricht aber ganz ihrer Diesseitigkeitsreli-
gion.’ Starting from a different point Conzelmann (129), followed by
Schille (427), puts the matter thus: Luke knows that the Sadducees
do not believe in resurrection but does not know why; he makes them
therefore rationalists, who must in consequence deny also angels and
spirits. Cf. also Roloff (328). A new explanation was given by D.
Daube (JBL 109 (1990), 493-7), as follows. The theme that is of
substance to Paul is resurrection, but this is viewed under two
aspects, (1) the final resurrection, and (2) the span between death and
resurrection, ‘which, in widespread belief, a good person spends in
the realm or mode of angel or spirit’ (493). It is not claimed that the
Sadducees denied outright the existence of spiritual beings—they
could not have done so—only that they denied the existence of an
interim state, in which those who had died existed as angels or
1066 COMMENTARY ON ACTS

spirits, these being more or less synonymous terms (see 1 Enoch

22.3, 7; 45.4-5; Mt. 22.30: Mk 12.25; Lk. 20.36). Thus in v. 9 the
spirit or angel that may have spoken to Paul is Jesus between death
and (final) resurrection. Cf. Weiser (617). Daube’s article was taken
up by B. T. Viviano and J. Taylor (JBL 111 (1992), 496-8), who
‘take the two nouns άγγελον and πνεύμα as standing in apposition to
ἀνάστασιν, and ... translate “the Sadducees say that there is no
resurrection either as an angel (i.e. in the form of an angel) or as a
spirit (i.e. in the form of a spirit) but the Pharisees acknowledge them
both” ’ (498). Viviano and Taylor refer to an essay by S. T. Lachs in
Gratz College Annual ofJewish Studies 6 (1977), 35-42 as the origin
of their view; but cf. already Bengel (475): 'Spiritus angelo opposi-
tus, dicit hic spiritual hominis defuncti.’
The second problem lies in the words Φαρισαίοι δέ ὁμολογοῦσιν
τὰ ἀμφότερα. But ἀμφότερα means both (of two); and on the usual
interpretation three things have been mentioned, resurrection, angel,
spirit. The problem was already seen by Chrysostom, Homilies 49.1:
καὶ μήν τρία ἐστίν πῶς οὖν λέγει ἀμφότερα; ἢ ὅτι πνεύμα καἰ
άγγελος ἕν ἐστιν, ἢ δτι ού μόνον ή λέξις περί δύο, ἀλλὰ καί περί
τριών λαμβάνεται, καταχρηστικῶς οὖν ούτως εἶπεν, καί ού
κυριολογῶν. This points out the problem, and also, on its own terms,
the necessary solution: we must simply say that Luke used the word
incorrectly—as he appears to have done at 19.16. Parallels (but very
few) can be found, e.g. Diodorus Siculus 1.75.1 (so BA 93; but
ἀμφότερα here could mean both). Alternatively we may say with
Blass (242), ‘άγγελος et πνεύμα paene pro una notione sunt... hinc
τὰ ἀμφότερα’. Cf. Knowling (468). Blass (242f.) adds, ‘Μή ... μήτε
... μήτε ibi locum habet, ubi post primam negationem partitio fit, ut
Mt 5.34sq’. If the recent suggestions summarized above are accepted
there is of course no problem in άμφότερα.
In this verse μέν is omitted by B pc latt; this removes the μέν ... δέ
construction and may, as stylistically the harder reading, be correct.
For the first μήτε, P74 L 104 323 1241 pm have μηδέ. This may have
been an attempt to mitigate the difficulty of άμφότερα.

9. κραυγή occurs here only in Acts; cf. the verb κραυγάζειν in

22.23; the court behaves like a rioting mob. διαμάχεσθαι, like
κραυγή and στάσις, is a strong word—stronger than διαλέγεσθαι
(Begs. 4.290). The scribes fight for their opinion. For defence of
Christians by Pharisaic scribes cf. 5.34, 39 (Gamaliel), μέρος here
must look back to and be explained by v. 6, which describes parts of
the company assembled in the Sanhedrin, but the word was occasion-
ally used for a party; see BA 1025, with reference to Josephus, War
1.143 (τό ’Αριστοβούλου μέρος) and a number of papyri (e.g. POxy
1278.24). For the Pharisee party’s view, ούδέν κακόν εὑρίσκομεν ἐν
τῷ ἀνθρώπω τούτῳ, cf. 23.29; 25.25; also Mt. 27.23, and especially
 56. PAUL BEFORE THE COUNCIL. 22.30-23.11. 1067

Lk. 23.4, 14, 22, where there is close verbal similarity. It seems that
Luke wished to draw attention to the similarity between the innocent
suffering of Paul and that of Jesus.
The verse concludes with a conditional sentence that lacks an
apodosis (supplied by <a> sa: μή θεομαχώμεν, which is as effective a
supplement as ‘was können wir dagegen machen?’ (BDR § 482.2, n.
3)). It is best perhaps to supply ‘What’ and treat the whole as a
question (cf. Jn 6.62). Delebecque (111), however, says that this is
not aposiopesis; the scribes dare not continue.
The scribes assume that it is at least possible that a spiritual being
(πνεῦμα ... ή άγγελος) has spoken to Paul, making a revelation to
him or ordering him to behave in a certain way; better to be safe than
sorry. For the words used cf. v. 8 and the note. The story of Paul’s
conversion and claim to have seen Jesus could easily suggest a
manifestation of one between death and resurrection at the last day.
This would be against Calvin’s view: ‘This [the use of πνεῦμα]
certainly ought to be taken as applying to the Holy Spirit’ (2.235).
Cf. Begs. 4.290: ‘πνεῦμα and ἄγγελος are here tautological. Possibly
this is why the Western text found in the African Latin reads sanctus
Spiritus.’ The middle of the verse shows an apparently pointless
textual complication.
τινές τών γραμματέων τού μέρους τών Φαρισαίων: Β C (Ψ)
36 (945) 1175 1739 1891 2495 al gig (h) sy sa
οἱ (om. L al) γραμματείς τ. μ. τ. Φ.: <a>
τινὲς τών Φ.: P74 A Ε 33 104 pc vg bo
10. The present participle γινομένης ( B C 1175 pc) gives a
better sense than the aorist γενομένης (P74 A Ε Ψ <a>); it might be
regarded as a scribal improvement but is probably too ancient a
reading to be dismissed on this ground. While, and because, the
uproar was going on the tribune became afraid (φοβηθείς, aorist).
διασπασθῇ, torn apart, is probably used literally, as it is at
Herodotus 3.13.2 (τούς ἄνδρας κρεοργηδὸν διασπάσαντες): some-
thing more than a debate was in progress. Cf. Demosthenes 5.5 (58),
μόνον οὐ διεσπάσθην. στράτευμα is not here an army but a military
detachment—the force available and thought suitable. It is clear that
if the tribune was not present in the council (see v. 30) he was near at
hand and available to take any necessary steps. For εις τήν
παρεμβολήν cf. 22.24.
τε is omitted by B 69. It is needed as a connecting particle and may
therefore be a secondary improvement.
Beyer (138) writes, ‘Nur aus dem Munde des Paulus selbst ist die
offenkundig einseitige Beleuchtung und Verspottung des Bildes
jener Vorgänge voll verständlich und erträglich.’ He adds that this is
confirmed by v. 11, which can have come only from Paul. Neither
argument is convincing; this, of course, does not prove the opposite.
1068 COMMENTARY ON ACTS

11. τῇ δέ ἐπιούση νυκτί, as in the more common ἐπιούση ήμερα,

the following night. Paul is comforted by a vision (for presumably he
sees the Lord standing by him) and a heavenly voice. Cf. 18.9f.;
27.23f. The Lord himself bids Paul take courage (θαρσεῖ; C3<a> h
vgmss add Παῦλε, from 27.24). ἐφιστάναι, used in a variety of senses,
often of aggressive presence, is a Lucan word (Lk., 7 times; Acts, 11
times; rest of the NT, 3 times). There is a similar use (of a divine
presence—Sarapis) in IG 10.2.255.3 (ND 1.29-32) and IG
42 1.123.12 (ND 2.21-3). Cf. παριστάναι (of Artemis) in Strabo
4.1.4.
The encouragement constitutes implicit commendation of the past
(Paul’s bearing witness in Jerusalem is evidently approved) and
prediction of the future (he need not fear that he will die in Jerusalem
since he will survive to bear witness in Rome). ‘Quod Paulus in
spiritu sibi proposuerat, 19.21, id nunc maturo tempore confirmat
Dominus’ (Bengel 476). For διαμαρτύρεσθαι, with a direct object,
cf. 20.21, 24 (without direct object, 20.23). Luke’s comment on this
incident is thus not that Paul has used a clever trick to get out of
trouble but that he has borne the witness he was intended to bear and
that the Lord has protected him and will continue to do so.
εἰς is twice used for ἐν, as frequently in Acts.
‘Sein Weg steht unter dem göttlichen dei' (Lüdemann 252). For
δεῖ cf. 19.21; also Josephus, Life 208f. (... 'Ρωμαίοις δεῖ σε
πολεμήσαι).
 57. THE PLOT; PAUL REMOVED TO CAESAREA 23.12-35

(12) When day broke the Jews made a plot and bound themselves by oath,
saying that they would neither eat nor drink till they had killed Paul. (13)
Those who made this conspiracy were more than forty in number. (14) They
approached the chief priests and elders and said, ‘We have bound ourselves
with an oath to eat nothing until we have killed Paul. (15) So do you make a
representation to the tribune, with the Sanhedrin, that he may bring him
down to you, as though you wanted to make a further inquiry into his case
with a view to reaching a decision; but we, before he comes near, will be
ready to kill him.’ (16) But the son of Paul’s sister heard of the ambush,
came and entered the barracks, and reported the matter to Paul. (17) Paul
called one of the centurions and said, ‘Take this young man to the tribune,
for he has something to tell him.’ (18) So he took him, brought him to the
tribune, and said, ‘The prisoner Paul called me and asked me to bring this
young man to you since he has something to say to you.’ (19) The tribune
took him by the hand, withdrew privately, and asked, ‘What is it that you
have to tell me?’ (20) He said, ‘The Jews have planned to ask you to bring
Paul down to the Sanhedrin tomorrow, on the pretext of making some more
accurate inquiry about him. (21) Now do not you be persuaded by them; for
above forty men of them are lying in wait for him, men who have bound
themselves with an oath not to eat or drink until they have killed him, and
they are ready now, awaiting your promise.’ (22) So the tribune dismissed
the young man, with the command not to tell anyone ‘that you have given me
this information’.
(23) He summoned two of the centurions and said, ‘Get ready two
hundred soldiers’ to go to Caesarea, with seventy cavalry and two hundred
lancers.2 [They are to be ready]3 from the third hour of the night, (24) and to
provide beasts in order to mount Paul and bring him safely to Felix the
governor.’ (25) He wrote a letter, as follows: (26) Claudius Lysias to His
Excellency the governor Felix: greeting. (27) I came up with the troops and
rescued this man, who had been taken by the Jews and was about to be killed
by them, when4 I learned that he was a Roman. (28) Wishing to discover the
ground on which they accused him, I brought him down to their Council.
(29) I found that he was accused over disputes arising out of their Law but
under no charge calling for death or imprisonment. (30) When it was made
known to me that there was to be a plot against the man I immediately sent
him to you, commanding his accusers also to state the case against him
before you.
(31) So the soldiers, in accordance with their instructions, took Paul and
brought him by night to Antipatris. (32) On the next day, having let the
1NEB, infantry.
2RSV, spearmen; NEB, light-armed troops; NJB, auxiliaries.
3They are to be ready is not in the Greek.
4NEB, because.
1069
1070 COMMENTARY ON ACTS

cavalry go on5 with Paul6 the infantry7 returned to the barracks. (33) The
cavalry entered Caesarea, delivered the letter to the governor, and also
presented Paul to him. (34) He read the letter,8 inquired of what province
Paul9 was, and having found out that he was from Cilicia (35) said, T will
hear you when your accusers also are present.’ And he ordered him to be
guarded in Herod’s praetorium.

Bibliography
F. F. Bruce, JSNT 1 (1978), 33-6.
S. Dav and S. Applebaum, PEQ 105 (1973), 91-9.
C. Hemer, JSNT 31 (1987), 45-9.
G. D. Kilpatrick, JTS 14 (1963), 393f.
S. Légasse, as in (54).
W. Stegemann, as in (43).

Commentary
In chs. 24,25 and 26 Paul’s story is based upon Caesarea, where he is
in Roman custody, custody designed not only to prevent him from
stirring up trouble but also to protect him from the Jews, whose
motives, and the legality of whose actions, were under suspicion. It
was thus necessary for Luke, in telling his story, to have Paul
removed from Jerusalem to Caesarea, and taken under Roman
protection as a Roman citizen who, though he might himself be in the
end found guilty of crimes, was entitled meanwhile to protection
from illegal attacks. The present paragraph effects these two objects;
the question that arises is whether Luke made it up in order to serve
these purposes or drew it from some traditional source. If the latter
alternative is accepted, the historical value of the tradition must be
considered. It is hard to see any theological motive in the paragraph
beyond the conviction that Paul is a better Jew—both in theology
and in mores—than the official representatives of Judaism, and that
the Lord protects his own. Weiser (621) is undoubtedly correct in the
view that ‘die jetzige Fassung des Textes [vv. 12-35] ist weitgehend
von Lukas gestaltet’; he gives many details. Similarly Lüdemann
(252f.): the passage is ‘sprachlich und inhaltlich von Lukas geprägt’;
also Schmithals (209). But opinions on its origin are divided.
Probably a majority consider that the story is based on good tradition.
Preuschen (133) thinks that since the plot has no outcome and does
5Greek, go off.
6Greek, with him.
7Greek, they.
8The letter is not in the Greek.
9Greek, he.
 57 THE PLOT, PAUL REMOVED TO CAESAREA. 23.12-35. 1071

not presuppose Paul’s appearance before the Council (23.1-10) the

story goes back ‘auf einen guten Bericht’. Roloff (330) similarly
believes that vv. 12-35 ‘geht sicher auf Tradition zurück’, and
further that it is part of the Prozessbericht and links with 22.29.
Haenchen (621) observes that in this story the Sanhedrin appears to
be united whereas in 23.1-10 it is divided: the two scenes are
independent and each is self-contained. Schneider (2.336) thinks
similarly; Pesch (2.247) disagrees. Marshall (367) comments on
Haenchen’s observation that in 23.1-10 the Pharisees took Paul’s
side that in fact only some of them did so (23.9).
The following points must be borne in mind. Luke’s strength as a
writer lies in the presentation of a single event, not in the logical
linking of a sequence of events. The story of the intervention of
Paul’s nephew and the story of the march to Caesarea are both told
with clarity, force, and liveliness. The question whether there was
any inconsistency between them and the story of 22.30-23.11, which
as we have seen had its own motivation and served its own purposes,
is not one that would have troubled him, if indeed he noticed it. In
contrast with that paragraph the present one has no directing purpose.
Luke’s main story would have been adequately served if he had
written: The tribune, hearing of a plot against Paul’s life, sent him by
night and under guard to Caesarea. That Luke did not effect the
transference in this way leads to the probable inference that he found
in the tradition the story of the nephew and some account of the
surprisingly powerful force that escorted Paul to Caesarea. Further, it
must be remembered that Paul had friends in Jerusalem, even if only
those whom he brought with him (and these could have included the
person responsible for the We-passages, who could not now use the
first person plural since Paul was alone and the only person who
mattered). Again, if it is correct that Paul spent his youth in
Jerusalem (22.3) it is not impossible that a married sister lived there.
One must not make too much of bare possibilities, but it is at least
conceivable that family memories persisted and that someone
watched the cavalcade set out (vv. 28f.).
The letter (vv. 26-30) is another matter. It is hard to imagine how
Luke could have obtained access to Roman archives whether in
Jerusalem or in Caesarea. It makes (with a little variation on the
events as described in chs. 21, 22) the points that concern Luke: the
trouble was caused not by Paul but by Jews; Paul was a Roman
citizen and entitled to protection, especially against plots to assassi-
nate him; the only matter at issue was interpretation of the Jewish
Law, and that therefore Paul had committed no offence against the
state. The tribune must have written an accompanying letter; must it
not have been cast in these terms? So Luke would think.

12. συστροφή in itself need mean no more than at 19.40, where

1072 COMMENTARY ON ACTS

the ‘gathering together’, though directed against Christians and not,

in the opinion of the town clerk, a properly constituted ἐκκλησία,
was not for the purpose of secret plotting. The word however is
capable of suggesting such a purpose (LS 1736 quote Polybius
4.34.6), and the context shows that this is (Begs. 4.290 ‘may be’) the
meaning here. Zerwick (§ 227) notes that the active verb ποιήσαντες
(contrast v. 13) is used: ‘non est simpliciter “convenerunt” sed
“institueront concursum”.’
οἱ Ιουδαίοι can hardly mean more than some Jews (see the next
verse). But they were Jews, and no doubt in Luke’s mind represented
the Jewish people as the chief enemies of the new faith. It was the
impossibility of believing that ‘the Jews’—all of them—undertook
the vow described that led to the variant τινες τῶν ’Ιουδαίων in P48
(but for the full reading of this MS see below) <a> lat syp samss.
The sense of ἀνεθεμάτισαν εαυτούς is given by v. 14 where
ἀναθέματι is added. Those who thus agreed undertook to accept the
ban of the synagogue if they failed to accomplish that which
they pledged themselves to do. The abstinence from eating and
drinking (which if adhered to must in this case have resulted in
death) is augmented by the religious sanction. Derrett (Law 347)
draws attention to Philo, De Specialibus Legibus 2.12, 13, where
Philo refers to those who excuse their wrong-doing on the ground
that they are fulfilling vows. ‘There are those who swear at random
(ἐὰν τύχη, perhaps, as opportunity offers) to commit acts of theft or
sacrilege or rape and adultery or assaults and murders or other
similar crimes and carry them out without hesitation on the pretext
that they must be faithful to their oaths.’ Philo properly points out the
overriding authority of the prior (implied) oath to observe the law,
which forbids such acts. Josephus (Ant. 20.163f.; cf. War 2.254-6),
however, describes the use of hired assassins in Jerusalem for
revenge on private and public enemies, though it exaggerates
somewhat to describe this (W. Foerster, Palestinian Judaism in NT
Times (1964, 1967), 105, n. 8) as the assassination of opponents of
the Law. Cf. Sanhedrin 9.6. It was possible to be released from
certain vows; see Nedarim 3.1-3 (’Four kinds of vows the Sages
have declared to be not binding: vows of incitement, vows of
exaggeration, vows made in error, and vows [that cannot be fulfilled
by reason] of constraint...’).
For ἀποκτείνωσιν, A pc Chrysostom have ἀνέλωσιν, which is
probably due to preference for the less usual and obvious word,
though it could be original, ἀποκτείνωσιν having been introduced by
assimilation to v. 14. The story of a plot going to such lengths is not
necessarily to be dismissed as unhistorical because it accords with
the picture of the Jews that Luke wishes to paint; we know that Paul
sometimes went in danger of his life (2 Cor. 11.24-26).
At this point the fragmentary P48 (see I.3) begins, and offers a
 57. THE PLOT; PAUL REMOVED TO CAESAREA. 23.12-35. 1073

series of striking and sometimes unique variants. The present verse

includes, instead of ποιήσαντες ... ’Ιουδαῖοι, βοήθειαν (a force of
auxiliaries, as in Xenophon and Thucydides; see LS 320, s.v. II)
συστραφέντες τινες τῶν ’Ιουδαίων (cf. the reading of <a> lat syp
samss, above).
For δέ, Β Ψ 614 2464 2495 pc h have τε. For the first μήτε, P48 pc
have μή.
13. For ποιησάμενοι (middle) cf. v. 12 (where the verb is active)
and see Zerwick (§ 227). They were comurati, they had made their
conspiracy among themselves. For συνωμοσία cf. Plutarch, Caesar 7
(710) (Catiline); 64 (738) (Brutus and Cassius). MM 616 cite no
papyrus evidence; ND 2.99 supplies P.Mil.Vogl. 6.287.9.
14. For chief priests and elders see 1.223. Official Jewish collab-
oration was necessary if Paul was to be placed in a situation in which
he could not be defended by Roman swords.
For ἀναθέματι ἀνεθεματίσαμεν cf, v. 12. BDR § 198.6, n. 9
regard this as an imitation of the Hebrew infinitive absolute. Μ.
2.443f. recognizes the parallel, but in view of classical parallels,
adds, with Radermacher, that it is only in the extension of such uses
in the NT that we should describe the use as a Semitism. Any Semitic
influence behind the examples in Acts is likely to have come through
the LXX.
εαυτούς is used for the first person plural of the reflexive pronoun
(ἡμᾶς αυτούς). This is very common in the NT and in Hellenistic
Greek generally. See Zerwick (§ 209); BDR § 64.1.
γεύσασθαι presumably covers both φαγεῖν and πιεῖν (ν. 12).
After γεύσασθαι, P48 gig h Lucifer add τὸ σύνολον—a characteristic
Western intensification. For ου, P48 has ὅτου; Ψ 104 1175 2464 pc
have έως αν.
StrB (2.767) refer to b. Sanhedrin 82a. The members of the
Sanhedrin are not to give counsel to a Zealot in such a case.
15. The conspirators disclose their plot and the part to be played in
it by the authorities.
The meaning of ἐμφανίζειν is not clear. At Jn 14.21, 22, with a
reflexive pronoun, it means to manifest; At Heb. 11.14, introducing a
ὅτι clause, it has the same meaning. The passive means to appear
(Mt. 27.53; Heb. 9.24). At Acts 23.22 the verb means to disclose, but
at 24.1; 25.2, 15 it appears to have a special legal meaning. At 24.1
(ἐνεφ. τῷ ήγεμόνι κατά τού II.) it differs little from to accuse, or
perhaps better, to lay an information against; 25.2 has the same
construction; at 25.15 the construction is similar (περί οὗ ἐνεφ.) and
the meaning is probably the same. The only reference given by LS
549 is to the Argument to Aristophanes, Lysistrata, but it appears to
have developed in Hellenistic usage; so e.g. 2 Macc. 3.7 (other
1074 COMMENTARY ON ACTS

examples in BA 520; also ND 2.104; see Josephus, Ant. 10.166;

14.213, 226). It is not easy however to fit this meaning into 23.15,
which is closely parallel to v. 20 and may mean little more than ask.
Preuschen (134), quoting Esther 2.22, prefers mitteilen; Page (233) is
nearer with ‘ “make a statement” or “declaration to the tribune to
induce him (ὅπως) to bring Paul down to you, or the place that you
propose Make a representation to may perhaps suggest the
sense of asking with the addition of a legal flavour supplied by a
word sometimes used in legal contexts.
σὺν τῷ συνεδρίῳ (on the text see below). With whom is the
συνέδριον connected, ὑμεῖς or ὁ χιλίαρχος? Neither alternative is
easy. The Sanhedrin was not asked to do anything; only the tribune
would bring Paul down—and that to the Sanhedrin. Moreover, you
in effect were the Sanhedrin (ἀρχιερείς καὶ πρεσβύτεροι, ν. 14).
This must probably be regarded as a careless piece of writing;
perhaps Luke wished to make it clear that the Sanhedrin would be
involved in any mischief that was afoot.
On καταγάγη see below. <a> adds αὔριον, a circumstantial touch
from ν. 20. For εις, P74 C Ψ <a> have πρός, a scribal improvement.
ὡς μέλλοντας, on the pretext that; ὡς suggests deceitfulness that
would not be implied by the participle alone.
διαγινώσκειν is a technical term in legal use: to decide or
determine a suit. Cf. 24.22; (and 25.21, διάγνωσις). Begs. 4.291 note
that the word means something more than inquiry, but find it difficult
to get a rendering for its double meaning. One must be prepared to
fill out: to make a further inquiry into his case (τά περί αυτού) with a
view to reaching a decision.
ἀκριβέστερον may be an elative comparative, as accurately as
possible, see BDR § 244.2, n. 3. There is some support for this in
papyri cited in MM 19. But there is no reason why the word should
not be a true comparative: to find out more accurately than you have
so far been able to do.
Such would be the pretext, and your share in the proceeding; but
we (ἡμεῖς δέ, emphatic) before he comes near (to you, ὑμιν, is added
by P48 vgmss sy) ἕτοιμοί ἐσμεν, present for future, giving greater
vividness: there we are, all ready ... τού ἀνελεῖν, genitive of
purpose, or of the thing aimed at (Page 233). At the end of the
sentence 614 h syhmg add ἐὰν δέη καί ἀποθανεῖν, even if we have to
die for it—a characteristic Western sharpening of the narrative.
The principle of the plot is clear; it could hardly be expected that
the details should be. καταγάγη implies that Paul was to be brought
down from the Antonia (see on 21.31) to wherever it was that the
Sanhedrin met (see I.223f.); but where the conspirators intended to
set upon him, and how they intended to circumvent the Roman guard
that would certainly be present, is not explained.
P48, which must be regarded as an important witness to the
 57. THE PLOT; PAUL REMOVED TO CAESAREA. 23.12-35. 1075

Western text, together with gig h syhmg sa Lucifer, rewrites the

beginning of the verse as follows: παρακαλοῦμεν ὑμᾶς ποιήσατε
ἡμῖν τούτο, συναγαγόντες τὸ συνέδριον, ἐμφανίσατε τῷ χιλιάρχω
ὅπως ... This looks like a free improvement, which, among other
things, avoids the problem, noted above, of σὺν τῷ συνεδρίῳ. Clark
(379), to whom P48 was unknown, gives the Latin, Syriac, and
Coptic material, reconstructing a Greek text on 149f. See also Ropes,
Begs. 3.228f.

16. ό υιός τής αδελφής. We have no other information about

members of Paul’s family. 22.3, however, suggests, but does not
prove, that some were, or had been, resident in Jerusalem.
τὴν ἀνέδραν. Cf. 25.3; also ἐνεδρεύειν in Lk. 11.54; Acts 23.21.
The word could be used for trickery or treachery of any kind, but its
primary sense was ambush, and it is clear from v. 15 that something
of this kind was included in the plot.
παραγενόμενος καὶ εἰσελθών appears to mean, having arrived
and having entered the barracks. This is clumsy, and Begs. 4.291
suggests that παραγενόμενος should be connected with what pre-
cedes: Having heard ... having been present. Paul’s nephew heard
about the plot because he was present, inadvertently included in the
group. As Begs, adds, 40 is too many for successful assassination.
Paul’s nephew has access to him; his residence in the barracks is
protective rather than penal, though, if the authorities were favour-
able, prisoners under charge or even undergoing punishment could
be visited (cf. 27.3).
Schmithals (210): this and the following verses show that the
Christians ‘das Vertrauen der römischen Obrigkeit verdienen und zur
Zeit des Paulus auch besassen’. In Luke’s time the persecutors did
not understand this. 212: ‘Das Christentum ist religiös eine Erschei-
nung des Judentums, politisch ein integrierender Teil der römischen
Ökumene.’

17. Again Paul appears to be in a favoured position; he is able to

summon a centurion and give him a message for the tribune.
άπαγε ( B 81 pc) is probably original; ἀπάγαγε, the aorist
imperative (P74 A C Ε Ψ <a>), is the more natural to use and probably
an editorial ‘improvement’.

18. ό δέσμιος. Paul is a prisoner, though favourably treated; that

is, the tribune considers that it might be dangerous to let him go.
ἠρώτησεν. As at 10.48, the use of the aorist corresponds to the fact
that the request was now fulfilled. See BDR § 328.2, n. 3.
νεανίσκον, P74 A E 33 81 323 453 945 1175 1739 1891 al;
νεανίαν, Β Ψ <a>: a textual problem impossible to solve. Has an
original νεανίσκον been assimilated to νεανίαν in v. 17, or an
1076 COMMENTARY ON ACTS

original νεανίαν to νεανίσκον in v. 22? Luke evidently regards each

word as equally suitable, and uses each several times in Acts.
ἔχοντα, causal: since he has something to say to you.
19. ἐπιλαβόμενος, used correctly with the genitive of the part
taken; BDR § 170.2, n. 2.
ἀναχωρήσας κατ’ ιδίαν. The young man will no doubt speak more
freely in private.
20. συνέθεντο corresponds to συστροφήν (v. 12) and
συνωμοσίαν (v. 13), ἐρωτῆσαι to εμφανίσατε (v. 15), without
suggesting the semi-technical legal atmosphere of these words. The
τοῦ is the genitive of purpose or aim, as in v. 15. μέλλον is probably
the reading we should accept, but there is great textual variation.
(a) μέλλον 33 1891 pc
(b) μέλλων P74 A B E 81 453 pc
(c) μέλλοντα <a>
(d) μέλλοντες 630 al lat sy
(e) μέλλοντας 2127 pc
(f) μελλόντων Ψ 36 614 945 1175 1739 2495 al
(a) is the reading that gives the required sense, since, in the terms of
the plot, it is the Sanhedrin that wishes to carry out further
investigation. Begs. 4.292 is prepared to accept this, though as an
emendation—the occurrence of the reading in * is accidental. But
since * is joined by 33 this need not be so. Ropes (Begs. 3.219)
thinks (b) an orthographical variation of (a). This is possible, since
the two words would be pronounced identically, but some scribes at
least will have understood μέλλων to be masculine singular nom-
inative, and to refer to the tribune. It is difficult to make sense of (c),
though it is widely supported; Ropes describes it as ‘particularly
unfortunate’. Probably one must refer it back to σε and give it the
same meaning as (b) (understanding this as written, masculine
nominative), (d), (e), (f ) will all, in rather odd ways, give the same
sense as (a): They (= the Sanhedrin) with to make a further inquiry.
ἀκριβέστερον. Cf. v. 15.
21. This verse relates matters that we already know, with only
small variation in wording.
ἐνεδρεύουσιν. Cf. ἐνέδρα, v. 16.
οἵτινες, men who ...; but in Lucan usage οἵτινες is scarcely
distinguishable from oἵ.
ἀνεθεμάτισαν. Cf. v. 14.
Begs. 4.292: ‘επαγγελία means most frequently a promise; but its
original sense is “a favourable message” and so “consent” or
“assent”, which is clearly the meaning demanded by this context.’
BA 567 similarly gives the meaning Zusage, supporting it with
 57. THE PLOT; PAUL REMOVED TO CAESAREA. 23.12-35. 1077

references to the Protevangelium of James 7.1, where the meaning

seems to be promise, and Ignatius, Ephesians 14.2, where promise
and profession are equally suitable—it is not promising or professing
to live as a Christian but actually doing so that counts, There is no
valid objection to the meaning promise here; indeed, consent implies
promise. If the tribune replies, ‘Very well, I will bring Paul down
tomorrow’ he is promising to do so.
22. In the course of the verse the construction changes from
indirect to direct speech: μηδενὶ ἐκλαλῆσαι ... ἐνεφάνισας πρὸς
ἐμέ. Cf. 1.4; and see BDR § 470.2, n. 3 (’auch bei Klassikern ...
nicht ungebräuchlich’). It is interesting that in vv. 23f. the reverse
change takes place.
In ν. 15 ἐμφανίζειν takes the dative; here πρός and the accusative.
This may suggest a small difference in meaning—in v. 15 the legal
sense, lay an information before, in v. 22 the non-technical, disclose
to. See on v. 15.
23. δύο τινας (τινας δύο in Β 33 81 pc; δύο in P74 pc lat syp) is
the reading of A Ε Ψ <a>; copyists felt and exercised considerable
freedom over such trifles. The meaning can hardly be about two
(which is the usual classical sense of τις with a numeral); Μ. 3.195
and BDR § 301.1, n. 3 suggest a certain two or a pair. Cf.
Thucydides 8.100.5, τινὲς δύο νῆες. The tribune acts through his
centurions; normal military practice.
στρατιώτας. The context shows that infantry are in mind. These
would naturally determine the speed at which the convoy would
proceed. See below. Two hundred men would (on paper) correspond
to two centurions.
Καισαρείας, where the governor of the province normally resided.
See 10.1 and the note. Jerusalem was no place for the trial of an
unpopular Jew who was also a Roman citizen; and no doubt Claudius
Lysias would be glad to hand on the explosive package that had been
thrust upon him.
ἱππείς ἑβδομήκοντα. The garrison at Jerusalem was provided
with cavalry (Josephus, Ant. 20.171). The Western text (614 1241
2495 pc h syhmg sa) has εκατoν as the numeral. It is not easy to see
how this variant could have arisen accidentally; at the same time the
difference is scarcely great enough to satisfy anyone who wished to
heighten the effect of the narrative. The Latin and Syriac texts at this
point are unusually free; see Ropes (Begs. 3.220f.; Metzger 488f.).
δεξιολάβους (δεξιοβόλους, in A 33, is not attested elsewhere and
may be simply a lapsus calami) διακοσίους. The meaning of
δεξιολάβος is a notorious problem. The Vulgate here reads lancear-
ios, lancers, spearmen; the Syriac has shooters or slingers
with the right hand—a wooden imitation of the Greek, though
perhaps of δεξιόβολοι rather than of δεξιολάβοι. Until further
1078 COMMENTARY ON ACTS

evidence is found it will probably be wise to follow the Vulgate

here—even if Jerome was guessing, his guess is at least as likely to
be right as those of modern times. The suggestion of Begs. 4.293
should however be noted: led horses. Between Jerusalem and
Antipatris (v. 31) the cavalry would need a change of mount; a led
horse would be taken by the right hand. See also Μ. 2.272f.:
δεξιολάβοι ‘... is supposed to mean taking (a spear) in the right
hand (instr. or loc. dependence). In military phraseology the spear
was always connected with the right, as the shield with the left. It
was certainly not a coined word, but as it does not reappear till vii/
AD we must suppose it a technical term of limited range.’ Cf.
Schneider (2.339): ‘militärischer Fachausdruck’. It may be added
that when the word does reappear (see BA 349) it seems to mean
light-armed troops (though this may be an inference from a connec-
tion with spears). See further below.
The escort comprised 470 men (500 according to the Western
text). If the garrison in Jerusalem (which seems to have had only one
tribune) was a single cohort, the escort amounted to half the troops in
the city, that is, if we assume a cohors miliaria; if the tribune
commanded only a cohors quingenaria he sent virtually his whole
force! Using these figures G. D. Kilpatrick (JTS 14 (1963), 393f.)
argued that the δεξιολάβοι, whatever they were, did not belong to the
cohort; only 270 regular soldiers were sent. Who then were the
δεξιολάβοι? The later Greek evidence interprets δεξιολάβος as
παραφύλαξ, and for this word LS give gendarme or police-officer
(in the Addenda, 2097; in the main text, 1330, watcher, guard; in the
Supplement (1968), 115, ‘name of an official, perh. chief ofpolice');
so also Revised Supplement (1996), 241. Kilpatrick concludes: ‘The
200 δεξιολάβοι will be spearmen from the local police and it would
be natural for police to be used to help escort a threatened prisoner to
a safer area.’ Delebecque (112) approves the suggestion. But have we
any evidence for local police in Jerusalem in addition to the Temple
police (see I.218f.), who would have been most unsuitable as a guard
for Paul, and to the Roman military force? The numbers (which
prompted Kilpatrick’s suggestion) certainly constitute a problem, but
the explanation of them is probably that Luke exaggerated them—or
guessed on the high side. Hengel (174) however quotes Josephus,
War 2.540-53 to illustrate dangers on the Palestinian roads, which
justify a large military accompaniment. The story as a whole (though
of course there is no means of confirming it) need not be fictitious
because Luke, who would not have access to official records of troop
movements, overestimated the figures. It may be that the infantry did
not go all the way to Antipatris.
A similar problem is raised by the distances involved, but this will
be considered at v. 31, since though Caesarea (the governor’s
residence) is mentioned in this verse as the goal of the journey it is
 57. THE PLOT; PAUL REMOVED TO CAESAREA. 23.12-35. 1079

not said here when the party is expected to arrive there. Caesarea was
about 50 miles from Jerusalem as the crow flies, probably at least 70
by road. See Josephus, War 1.79 (600 stadia = 69 miles).
ἀπό τρίτης ώρας must mean that the escort was to be ready from
the third hour onwards, that it must be ready to start at any time after
that hour. The time would be three hours after sunset (or more
precisely 3/12 of the time between sunset and sunrise; according to
Begs. 4.293, about 9.30 p.m.).
For the Western text at the end of this verse see on v. 24.
Bauemfeind (260), on vv. 23-30; ‘Lk hält den Leser bei diesen
Bildern fest, nicht aus Freude am Erzählen, auch nicht deshalb, weil
Paulus es wert ist, dass man seinem Erleben im grossen und im
einzelnen folgt, sondern weil man wissen muss, dass die Korre-
spondenzen der obrigkeitlichen Dienststellen sowie die Soldaten
sämtlicher Waffengattungen alle dem Lauf des Ev dienen müssen,
wenn es ein wirklicher Streiter Jesu ist, dem ihr Aufwand gilt.’ But
was Luke so subtle? And did he not have Freude am Erzählen? See
Introduction, pp. xlix, cxi.

24. The text of vv. 23, 24, (25) is in considerable confusion. The
text printed in NA26 (which for convenience if for no other reason
will form the basis of the commentary on these verses) is sub-
stantially the Old Uncial text. The reconstruction of an alternative
(Western?) test is itself a matter of some difficulty, though most of
the variations are of no great significance. The Western test as
reconstructed by Clark (151, 379f.) is as follows.
Ετοιμάσατε στρατιώτας, ὅπως πορευθῶσιν έως Καισαρείας,
ιππείς εκατόν καί δεξιολάβους διακοσὶους. καὶ από τρίτης
ὥρας τῆς νυκτὸς κελεύει ετοίμους εἶναι πορεύεσθαι. καί τοῖς
ἑκατοντάρχοις παρήγγειλεν κτήνη παραστῆσαι, ἵνα
ἑπιβιβάσαντες τὸν Παῦλον διὰ νυκτὸς διασώσωσιν εἰς
Καισάρείαν πρὸς Φήλικα τόν ἡγεμόνα. ἐφοβήθη γάρ μήποτε
ἁρπάσαντες αὐτὸν οἱ ’Ιουδαῖοι ἀποκτείνωσιν καί αυτός
μεταξύ ἔγκλησιν ἔχη ὡς ἀργύριον εἰληφώς, ἔγραψεν δέ ...
The one substantial addition here supplies an additional motive for
the tribune’s provision of so large a force to convey a prisoner safely
from Jerusalem to Caesarea: he was afraid that if Paul were killed on
the way he would be accused of having accepted a bribe to allow this
to happen. There is nothing improbable in the suggestion; equally
there is nothing in the narrative outside v. 24 to prompt it. The
tribune was in any case under obligation to secure the safety of a
Roman citizen, as yet not merely uncondemned but untried, and
therefore presumed innocent. This is one of the most extensive
variations introduced by the Western text; it is noteworthy that it
occurs in a passage where there is (unless we follow Bauemfeind—
1080 COMMENTARY ON ACTS

see on v. 23) no theological interest. This is (apart from one person

involved) a secular story, and editors felt no obligation to maintain its
wording intact.
κτήνη τε παραστῆσαι. The construction changes from direct
speech (imperative, ετοιμάσατε, v. 23) to indirect (the infinitive
depends on εἶπεν). Contrast v. 22, where the opposite change takes
place. It does not seem likely that this should be an example of the
use of the infinitive as imperative (see Moule, IB 126; also BDR §
470.3). κτήνος usually means an ox or sheep, often in the plural,
flocks and herds. The only examples given by LS 1002 for the
meaning horse or mule for riding are the present passage and Lk.
10.34. It is possible however to add Josephus, Ant. 8.241 (λέων ...
κατασπάσας αυτόν από τού κτήνους ἀπέκτεινεν; the context
shows that the animal was an ass, ὄνος) and POxy 17.21253.16, 17,
20.
ἐπιβιβάσαντες is not to be taken as literally as it is by Delebecque
(113), who concludes that Paul was perhaps ‘un cavalier médiocre’.
Paul was mounted on the animal but not necessarily lifted up and
seated on it. He was being well treated; not for the first or last time
Luke illustrates the favour shown to him by Roman authorities. The
tribune, for whatever motive, means to bring him safely through
(διασώσωσιν; a Lucan word—5 times in Acts, once in Lk., twice in
the rest of the NT) to Felix, κτήνη, plural, may go with the
suggestion that the δεξιολάβοι are led horses.
Two Western additions, νυκτός and εις Καισάρειαν, are scarcely
necessary after v. 23, but perhaps make the story more vivid.
Felix: see Begs. 5.286, 465-72; also NS 1.459-66. According to
Tacitus, Histories 5.9, his full name was Antonius Felix, according to
Josephus, Ant. 20.137, Claudius Felix. This however may be an
error. The MSS have πέμπει δὲ καὶ Κλαύδιον Φήλικα, but it is
probably right, with the Epitome, to read Κλαύδιος ..., (the
Emperor) Claudius sent Felix. He was a brother of the influential
freedman Pallas, and was himself a freedman, probably, if we may
follow Tacitus in regard to his name, of Antonia, the mother of
Claudius. Such an appointment was unusual; provinces such as
Judaea were usually entrusted to Roman knights. According to
Suetonius, Claudius 28, he was married three times, trium reginarum
maritum. Tacitus, loc. cit, says that he married Drusilla the grand-
daughter of Cleopatra and Antony (Drusilla Cleopatrae et Antonii
nepte in matrimonium accepta). Josephus, Ant. 20.141-4, describes
how he took Drusilla, the sister of Herod Agrippa Π, from her
husband Azizus, king of Emesa, and married her. It seems that there
may be some confusion here; see Acts 24.24. The dates of Felix’s
governorship are disputed. NS loc. cit. gives as his dates AD 52-60?;
Begs. 5.465-72 prefers 52 or 53-55. The time of his recall will be
discussed at 24.27. Before coming to the appointment of Felix
 57. THE PLOT; PAUL REMOVED TO CAESAREA. 23.12-35. 1081

Josephus narrates the misgovernment of Cumanus and the appear-

ance before Claudius in Rome of rival groups of Samaritans and
Jews. Claudius decided against the Samaritans, whom he put to
death, and Cumanus, whom he exiled, and, at the suggestion of one
of the Jewish party, Jonathan the High Priest (Ant. 20.162), appoin-
ted Felix. He adds (20.138), ‘When he had completed the twelfth
year of his reign, he granted to Agrippa the tetrarchy of Philip
together with Batanaea ...’. The twelfth year of Claudius ended on
24 January 53. Thus if Josephus is maintaining a strict chronological
sequence of events, Felix will have been sent to Judaea in 52
(possibly in January 53). Tacitus’s account (in Annals 12.54) is
different. Felix had shared in the misdeeds of Cumanus. These were
ended by Ummidius Quadratus, governor of Syria, who, however,
put Cumanus on trial and appointed Felix one of his judges. Felix had
thus been active in Palestine before his appointment as procurator.
This may account for Paul’s words in 24.10 (see the note). Felix’s
character, and the character of his rule, are described by Tacitus in
the words (Hist. 5.9) per omnem saevitiam ac libidinem ius regium
servili ingenio exercuit. Most writers accept this judgement; but cf.
Begs. 5.286: ‘The appointment of Felix was undoubtedly a bad one,
if we judge it from the state of near anarchy into which he allowed
Judea to sink. Nevertheless in passing judgement upon Claudius’s
responsibility for sending him to that country, one should not forget
that the appointment was made at the request of the high priest
Jonathan, and that the procurator’s wrongs were acts of omission
rather than commission. One wonders what procurator could have
proved successful in a country where the Zealots had been teaching a
whole generation that relentless opposition to the Roman govern-
ment was willed by God. The appearance of organized guerrilla
warfare forced Felix to resort to acts which appeared unduly cruel to
the excited populace, and, in turn, increased the fury and influence of
the Zealots.’
See F. F. Bruce, ‘The full name of the Procurator Felix’, JSNT 1
(1978), 33-6; C. J. Hemer, ‘The name of Felix again’, JSNT 31
(1987), 45-9.
For έγκλημα see Dittenberger, OGIS 1.229.41 (ND 3.66).
25. The text of this apparently straightforward verse is compli-
cated by many variants, which fortunately make little difference to its
meaning.
For γράψας
P48 has ἔγραψε δὲ αῦτοῖς. The participle is replaced by the finite
verb and a connecting particle (as in the other variants). Unless
αὐτοῖς means for them—for the guard so that they may deliver it
to the governor—this is a lapse or a misconception; the tribune
was writing to the governor, not the guard.
1082 COMMENTARY ON ACTS

614 has έγραψε δέ

2147 vg have ἔγραψε δὲ καί
syp has ἔγραψε δέ καί ἔδωκεν αυτοῖς—a more circumstantial
account of the delivery of the message.
For ἔχουσαν τόν τύπον τούτον
A have περιέχουσαν τόν τύπον τούτον
614 (2147) (vg) have περιέχουσαν τάδε
Ρ48 has ἐν ᾗ ἐγέγραπτο
syp hmg have ᾗ ταῦτα
περιέχειν (for the transitive and intransitive uses see BDR § 308,
n. 5) was probably introduced as a more literary way of referring to
the contents of a document. See 1 Macc. 15.2, ᾖσαν περιέχουσαιτόν
τρόπον τούτον; Philo, De Decalogo 168, with τύπον. τύπος (like
exemplum) of the purport or content of a letter: 3 Macc. 3.30; but see
E. A. Judge in ND 1.77 f.(see also 2.75), arguing that the word
suggests an exact copy, not merely the drift of a document. ‘Such a
letter, containing a summary of the facts, when a charge was referred
to a superior magistrate, was technically termed elogium' (Page 234).
Bengel (476), hopefully: ‘Hoc sine dubio Latine scriptum et in
archivis Romanis servatum Romanos postea de veritate historiae
apostolicae, hanc cum legerent, convicit.’ Roloff (332), however:
‘von Lukas formuliert'. But it is fair to say that the letter contains on
the whole the sort of report that the tribune is bound to have made to
his superior.

26. For the first time we learn the name of the tribune—Luke’s
own invention, according to Roloff (332); possibly, but not neces-
sarily so. We know from 22.28 that he had obtained Roman
citizenship, and the citizen’s nomen, Claudius, has been taken to
indicate that he received the citizenship, as many did, under the
Emperor Claudius; again, possible but not certain. His letter is
addressed in conventional fashion, τω κρατίστῳ ἡγεμόνι Φήλικι.
κράτιστος corresponds to egregius and was an appropriate address.
Conzelmann (131) quotes Dittenberger, OGIS 2.667.3f. (from the
time of Nero): ἐπ<ε>ὶ ’Ιουλίου Οὐηστίνου τού κρατίστου
ἡγεμόνος. Cf. I Eph 1.24.Α.6 (quoted in ND 4.82). There are several
unimportant variants.
Φήλικι τῷ κρατίστῳ ἡγεμόνι Ρ48 gig
Κρατίστῳ ἡγεμόνι Φήλικι 614 2495 pc
τφ κρατίστφ ἡγεμόνι 1838
χαίρειν is the usual greeting in a Greek letter, illustrated many times
in the papyri. It appears in the NT at Acts 15.23; Jas 1.1.
‘Der griechische Name Lysias macht seine Herkunft aus der
 57. THE PLOT; PAUL REMOVED TO CAESAREA. 23.12-35. 1083

griechisch sprechenden Bevölkerung des Küstengebiets oder Samar-

iens wahrscheinlich, aus der die meisten der in der Provinz Judäa
stationisierten Soldaten stammten’ (Stählin 292).

27. Claudius Lysias proceeds to give an account of the events

leading up to his letter. It does not correspond exactly with the
narrative in chs. 21, 22. This may be simply careless writing on
Luke’s part (cf. his different accounts of Paul’s conversion) or an
attempt by the tribune to set the story (especially so far as it relates to
Paul’s citizenship) in a more favourable light than the facts war-
ranted. It would be unwise to exclude either as a contributing factor,
though it is hard to see how Luke could have obtained a copy of the
official letter. The style however shows a somewhat convoluted
officialese (but Luke could write in different styles when he
wished—witness Lk. 1; 2), and there are several technical legal
terms.
συλληφθέντα, a technical term, to arrest, but here rather of
seizure by a mob.
σὺν τῷ στρατεύματι: cf. 23.10. Not with the army, but with the
military, with a detachment of troops.
ἐξειλάμην] έρυσάμην, P48.
μαθών δτι 'Ρωμαίος έστιν. It would be difficult to make this
mean, T rescued him, and subsequently discovered that he was a
Roman.’ The aorist participle almost certainly refers to action before
that of the main verb. ‘I rescued him when I discovered, or because I
had discovered, that he was a Roman.’ If the dialogue of 22.25-29 is
derived from Paul’s own recollection and is correct one can only
suppose that the tribune is improving on the facts for the benefit of
his superior (or that Luke wished to attribute this motive to him). If
the wording of the letter is correct it is possible that the tribune was
speaking the truth and that it was information that Paul was a Roman
that led him to take his troops out of the Antonia; why otherwise
should he have taken such provocative steps to prevent one insignif-
icant Jew from being beaten by his compatriots? This would do no
harm to anyone but the man himself. In this case the narrative of
22.25-29 is in error. It must be remembered that events such as those
described in these chapters are seldom clear cut. The situation was no
doubt very confused and different people will have retained different
impressions of what took place. Schille (430) speaks of Luke’s love
of variety; Rackham (440) returns to his view of Luke’s use of
participles; they are to be taken in order, so that we have, T rescued
... I learned ...’. Johnson (405) repunctuates, putting a stop after
ἐξειλάμην and starting a sentence with μαθών: having learned that
... and wishing ... I brought him down. But surely this would require
a particle after μαθών.
P48vid gig give a more natural picture, replacing μαθών ... ἐστιν by
1084 COMMENTARY ON ACTS

κράζον[τα καί λέγοντα εαυτόν] είναι 'Pωμαῖον. Neither does this

correspond with ch. 22.
The letter as a whole shows: (1) Paul is a Roman citizen and under
the Procurator’s jurisdiction. (2) His rescue is due to Roman power.
(3) No Roman offence was committed. (4) The charge is due to
internal Jewish party feeling. So Roloff (332f.). Schneider (2.340)
agrees with Haenchen that the variant account in the letter is intended
to show Rome’s respect for the citizen.
‘Perhaps the letter merely meant to say that the tribune intervened
in the riot, learned of Paul’s citizenship, and examined the case’
(Begs. 4.294).

28. This verse recalls 22.30. αιτίαν, charge, and ἐνεκάλουν,

accuse are technical terms.
For τε, <a> e gig syh have δέ.
κατήγαγον ... αυτών is omitted by B* 81. Ropes (Begs. 3.221)
followed by Metzger (489) explains this as due to homoeoteleuton;
this could be so (αὐτῷ ... αυτῶν) if the latter pronoun was written
(as frequently) ΑΥΤΩ and the horizontal stroke was missed.
κατήγαγον is left without an object in Ρ74 A 33 614 945 1739
1891 2464 pc; αυτόν is added by Ρ48vid Β2ΕΨ <a> lat sy.

29. περί ζητημάτων τού νόμου αυτών: not a bad summary of

23.6-9, νόμος being the authoritative basis of the Jewish religion.
Luke no doubt thought it well that the Romans should think of
Christianity as a variety of Judaism, at worst a heretical form of it, at
best its supreme expression.
έγκλημα, another technical term; cf. v. 25(24), v.1.; 25.16. It is a
complaint, or more specifically a charge. See MM 179f. for papyrus
use; add ND 3.66. Paul was accused of disputed matters, but there
was no accusation calling for (worthy of) death or imprisonment.
The charge of bringing Gentiles into the Temple (21.28) appears at
this point to have been dropped; but see 24.19f.
The Western text gives more Christian content to the dispute
evoked by Paul, adding after αυτών, Μωυσέως καί Ιησοῦ τινός.
This could mean ‘the Law of Moses and of Jesus’, but it more
probably means that the dispute concerned their Law, that of Moses,
and a certain Jesus. This addition is found in 614 2147 gig syhmg; the
same MSS add at the end of the verse ἐξήγαγον αυτόν μόλις τή βία;
a fair account of what is narrated in ch. 21 and heightening the
interest at this point.

30. The letter ends with a very stilted sentence. Better than the
genitive absolute would have been the accusative absolute μηνυθέν
followed by the accusative and infinitive (Μ. 1.74); alternatively,
ἐμηνύθη ἐπιβουλή ἔσεσθαι (BDR § 424, n. 2). Preuschen (136)
 57. THE PLOT; PAUL REMOVED TO CAESAREA. 23.12-35. 1085

thinks that two constructions are mixed: μηνυθείσης μοι ἐπιβουλής

ἐσομένης and μηνυθέντος μοι ἐπιβουλήν μέλλειν ἔσεσθαι. There is
a better construction in Thucydides 4.89.2: μηνυθέντος τοῦ
ἐπιβουλεύματος ὑπό Νικομάχου ... ὅς Λακεδαιμονίοις εἶπεν,
ἐκεῖνοι δὲ Βοιωτοῖς. For the latter part of the sentence cf. Her-
odotus 6.98.1, τοῦτο μέν κου τέρας ἀνθρώποισι τών μελλόντων
ἔσεσθαι κακών ἔφηνε ό θεός.
ἔσεσθαι (future infinitive) occurs in the NT only in Acts, and here
it ‘is in an official letter in stilted style’ (Μ. 2.219); cf. BDR § 350,
the future infinitive ‘ist der Volkssprache verlorengegangen’.
With these words Claudius Lysias passes on the information
brought to him by Paul’s nephew, and at the same time passes on his
prisoner to the higher authority, adding that he has also ordered the
Jews who are accusing Paul to appear before Felix.
ἔσεσθαι ἐξαυτής (P74 Β Ψ 36 453 614 2464 pc bo) seems to be the
original text, though ἔσεσθαι ἐξ αυτών ( A E (33) 81 (945) 1175
1739 1891 2495 pc (vg) syh) has good support and could be right—
either of these could give rise to the other; μέλλειν ἔσεσθαι υπό τών
Ι' ουδαίων, ἐξαυτής (<a> (gig syp) sa) is secondary.
λέγειν πρὸς αὐτόν (Β 1175) is surprising, but may be right: the
accusers should speak to him, bring their accusations to his face.
Instead, A 33 pc have λέγειν αυτούς, he told his accusers to speak
themselves, not leaving it to a Roman official to bring a charge in
which he had no interest, λέγειν τὰ πρὸς αὐτόν (Ε Ψ <a> syh), to
speak with reference to his case, is an attempt to improve an
awkward but sufficiently clear phrase.
At the end of the letter X Ε Ψ <a> vgcl sy add the conventional
greeting, ἔρρωσο; P 1241 pm have the less likely plural, ἔρρωσθε.
See 15.29.

31. μὲν οὖν picks up the narrative after the quotation of the text of
the letter. Cf. 15.30.
διά (<a> adds τής; the whole phrase is omitted by P74) νυκτός.
Moule, IB 56 distinguishes the prepositional phrase from the simple
νυκτός: διά v. means in the night, v. by night. Cf. 16.9 (where also
many MSS add τής); and see Zerwick (§ 115). In the night
corresponds with τῇ δὲ ἐπαύριον in ν. 32. Delebecque (113), ‘toute
la nuit’.
For Antipatris see NS 2.167f. ‘A. is to be looked for about eight
miles south of Kfar Saba, in a well-watered neighbourhood, whereas
Kfar Saba itself stands on arid soil.’ This follows Josephus, Ant.
16.142 against 13.390. G. A. Smith, Historical Geography of the
Holy Land (1894, 1968), 124 prefers Ras el-'Ain as the site. The city
was founded by Herod the Great in honour of his father, Antipater
(Josephus, Ant. 16.142f.; also War 1.417). It was a Hellenistic town;
see Sevenster (Do you know Greek? SuppNovT 19 (1968), 97) and
1086 COMMENTARY ON ACTS

Tcherikover (Hellenistic Civilization 46-8). It is mentioned at Gittin

7.7 as between Judaea and Galilee. Hemer (128) refers to S. Dar and
S. Applebaum, ‘The Roman Road from Antipatris to Caesarea’, PEQ
105 (1973), 91-9, and adds that from Jerusalem to Antipatris (via
Lydda) was c. 35 miles. Marshall (372) gives 37 miles, 60 km.;
Haenchen (620) gives 62 km. (= 38.75) miles), Hanson (223) gives
45 miles (= 72 km.). These surprisingly divergent figures reflect
uncertainty about the site (see above) and the existence of alternative
routes, by Lydda and by Bethel—see Begs. 4.295.
The name bears the article, τήν Άντιπατρίδα, possibly as a well-
known station on the route from Jerusalem to Caesarea. See v. 23. It
is very unlikely that the cavalry, with Paul mounted, could reach
Antipatris in the course of a night, absolutely impossible that the
infantry should have done so. Some of the difficulty is removed (also
some of the difficulty regarding numbers) if we may suppose that the
infantry was intended only to get Paul out of the city, and then to
return (cf. Marshall 372); but not the whole of the difficulty. And v.
32 implies (but does not quite state) that it was from Antipatris that
the infantry returned. The details of Luke’s circumstantial and at first
sight convincing narrative tend to crumble, without however destroy-
ing the general picture. He has probably written up imaginatively a
bare historical outline: ‘Paul was transferred to Caesarea.’ Luke’s
knowledge of the geography of Palestine is inexact (Schille 430;
Stählin 292f.).
For the whole journey, Jerusalem to Caesarea, see Josephus, War
1.79 (600 stadia).

32. τῇ δέ επαύριον; see on ν. 31. This confirms that διὰ νυκτός

refers to the activities of a single night. It was (according to Luke) on
the day after leaving Jersualem late at night that the party reached
Antipatris. For the distance see on v. 31. Knowling’s alternative
(475), ‘Not necessarily the morrow after they left Jerusalem, but the
morrow after they arrived at Antipatris’, is strained.
ὑπέστρεψαν: the subject must be the infantry, since the cavalry
are allowed to go on (ἀπέρχεσθαι, to go off, a surprising word,
changed by Ψ <a> to the more conventional πορεύεσθαι).
παρεμβολήν: back to the barracks (the Antonia).
ἐάσαντες presumably refers to the subject of ὑπέστρεψαν, that is,
the infantry; but Luke must mean the centurions commanding the
infantry. Or could ἐάω simply mean, they left them?'

33. The military escort fulfilled its mission, according to instruc-

tions. The journey from Antipatris to Caesarea was about 25 miles.
ἀναδόντες: ἀναδιδόναι is used for delivering letters, e.g. Polybius
29.10(25).7; Diodorus Siculus 11.45.3; cf. IG 14.830, line 22 (LS
103). MM 32 quote papyri, including PTebt 2.448.6-11 (ii/iii AD),
 57. THE PLOT; PAUL REMOVED TO CAESAREA. 23.12-35. 1087

τῷ ἀναδιδόντι σοι τὸ ἐπιστόλιον. A different word, παρέστησαν, is

used for handing over the person.

34. ἐπερωτήσας ἐκ ποίας ἐπαρχείας (so Β* A; the rest rather

more correctly, since επαρχεία is strictly the office of a prefect, have
ἐπαρχία, which is province) ἐστίν. See the long discussion, correct-
ing Mommsen, by Sherwin-White (28-31). He resumes his argu-
ment on p. 55: ‘It was argued earlier that the custom of forum
domicilii [over against forum delicti], that is, of referring an accused
person back to the jurisdiction of his native province, was never more
than optional, and that it was not firmly established in the early
Principate.’ So also Tajra (116f.). Sherwin-White adds (ibidem),
‘But in the case of a Roman citizen, and of a Jewish imbroglio, the
procurator might well have been glad to avail himself of any such
usage.’ Yet he did not. Having ascertained that Paul belonged not to
his own province but to Cilicia he undertook to try the case himself
(v. 35). Why then take the trouble to make the inquiry? Possibly
these were only preliminary inquiries, but Sherwin-White has a
better answer. ‘If Cilicia at the time of the incident of Acts xxiv did
not have a separate imperial legate, Felix’s decision is explained. The
Legate of Syria was not to be bothered with minor cases from Judaea,
though it was his duty to intervene in times of great crisis, and the
status of Cilicia did not require that its natives should be sent to it for
trial, even if the later usage offorum domicilii was in vogue’ (56). On
this view, Felix hoped for an answer that would enable him to slip
out of the whole affair, but found that in fact he could not do so. An
alternative explanation is that his inquiry was intended to ascertain
whether Paul came from a Roman province or from one of the client
kingdoms; in the latter case it would have been unwise not to send
Paul back to his place of origin. In view of the changes (see above) in
regard to forum domicilii and forum delicti Conzelmann’s references
(131) to the Acts of Carpus, Papylus, and Agathonike 24-27 and to
the Martyrdom of Justin 4.7ff lose some relevance; in any case in the
Martyrdom of Justin inquiry is made into the residence of parents
when these are said to have been responsible for making their
children Christians, and thus are themselves liable to a charge.
Cilicia, in the early Principate, was a dependency of the province
of Syria; later, probably in the early years of Nero, it became an
independent province. See Hemer (172, 179, 180, 290f., 381).
ποιας is used for τίνος. Or could it mean, ‘What sort of province,
senatorial or imperial?’
The end of this verse and the beginning of the next are rewritten in
direct speech by 614 2147 syhmg: ἀναγνούς δέ τήν ἐπιστ. καὶ
ἐπηρώτησε τὸν Παύλον ἐκ ποίας ἐπαρχ. εἶ; ἔφη Κίλιξ. και
πυθόμενος ἔφη· ἀκούσομαί σου ... This Western variant is, as
usual, livelier than that of the Old Uncial text. ND 4.173 gives
1088 COMMENTARY ON ACTS

evidence of the use of the form Κίλιξ, but it was current from Homer
onwards. The Western reading is taken up by J. L. North (JTS 47
(1996), 439-63) in the light of evidence that Cilicians had a
reputation for lying.

35. διακούειν (future deponent διακούσομαι) is used for hearing

the various parties in a dispute. See Job 9.33; Plutarch, Themistocles
2(112), Άναξαγόρου τε διακοῦσαι τὸν Θεμιστοκλέα φησί: there is
much non-literary evidence in MM 150. It also means afilli hearing;
contrast v. 34. Page (235) quotes Digest 48.3.6: Qui cum elogio [see
on ν. 25] mittuntur, ex integro audiendi sunt.
ἔφη. BDR § 465.3, n. 5 say that this is not parenthesis but variation
in word-order. Is it not variation that leads to parenthesis?
κελεύσας. Μ. 1.132f. argues that we cannot take this as an
example of subsequent action expressed by an aorist participle. At
most it is coincident action: ‘... “he said ..., meanwhile ordering
him ...,” which may perfectly well mean that Felix first told his
soldiers where they were to take Paul, and then assured the prisoner
of an early trial, just before the guards led him away.’
ἐν τῷ πραιτωρίω τοῦ ( A Ε 33 81 453 945 1739 1891 2464 al;
τω, B 1175; omit, Ρ74 Ψ <a>) Ήρῴδου, the palace built by Herod the
Great and taken over by the Roman administration. We have no
archaeological or other information about it (except that it existed)
and its site remains unknown. At this point Paul becomes formally a
state prisoner (Ehrhardt, Acts 113).
 XV
PAUL AND THE ROMANS
(24.1-26.32)

58. PAUL AND FELIX 24.1-27

(1) After five days the High Priest Ananias with a number of elders and a
barrister Tertullus, came down; these informed the governor against Paul. (2)
When he1 was called, Tertullus began his accusation, saying, ‘Since through
you we enjoy much peace and since reforms for this nation are coming about
through your provident foresight, (3) in all ways and in every place we
welcome this, most excellent Felix, with all gratitude. (4) But in order not to
weary you further, I ask you in your forbearance to hear us briefly. (5) For
we found this man a pest, and one who stirs up riots among all the Jews
throughout the world, and a ringleader of the sect of the Nazoraeans. (6) He
also tried to profane the Temple, and we seized him.2 (8) From him you will
be able yourself, by examining him, to find out about the things of which we
accuse him.’ (9) The Jews also joined in this attack, affirming that these
things were so.
(10) When the governor motioned to him to speak, Paul answered, ‘Since
I know that you have been for many years a judge of this nation I cheerfully
make my defence, (11) since you can find out that it is not more than twelve
days since I came up to Jerusalem to worship. (12) And neither in the
Temple nor in the synagogues nor anywhere in the city did they find me
arguing with anyone or causing the onset of a crowd, (13) nor can they prove
any of those things of which they now accuse me. (14) I do confess this to
you, that it is in accordance with the Way, which they call a sect, that I serve
our ancestral God, believing all things that are according to the Law and the
things that are written in the prophets, (15) having the same hope in God
which they also accept, that there will be a resurrection of both the righteous
and the unrighteous. (16) And so for my part I too exercise myself so as to
have continually a blameless conscience towards both God and men. (17) At
the end of many years I arrived [here]3 to bring alms for my nation and
offerings. (18) While I was engaged in these sacrifices they found me in the
Temple in a state of purity, with no crowd, with no tumult. (19) But some of
the Jews from Asia [allege this],4 who ought to be present before you and
bring an accusation, if they have anything against me. (20) Or let these men

1NEB, the prisoner; NJB, Paul.

2For an insertion made here by many MSS see the note on v. 7.
3Here is not in the Greek.
4Allege this is not m the Greek, where the sentence appears to be incomplete.
1089
1090 COMMENTARY ON ACTS

themselves say what crime they found in me when I stood before the
Sanhedrin, (21) unless it be this one thing that I cried out when I stood
among them, I am being judged this day before you concerning the
resurrection of the dead.’
(22) Felix5 adjourned [the hearing]6 since he had an accurate knowledge of
the Way, and said, ‘When Lysias the tribune comes down I will hear your
case and reach a decision.’ (23) He charged the centurion that he should be
kept7 safe and have relief from prison regimen, and that he should not
prevent any of his friends from doing him service.
(24) After some days Felix arrived with Drusilla his wife, who was a
Jewess, and sent for Paul and heard him on faith in Christ [Jesus].8 (25) As
he discoursed on righteousness, self-control, and the judgement to come
Felix became afraid, and answered, ‘For the present, go your way; when I
have an opportunity I will send for you.’ (26) At the same time also he hoped
that money would be given him by Paul. For this reason also he sent for him
pretty often, and conversed with him. (27) When two years were up Felix
was succeeded by Porcius Festus. Wishing to curry favour with the Jews
Felix left Paul in prison.

Bibliography
C. K. Barrett, FS Bammel, 361-7.
C. K. Barrett, FS Furnish, 161-72.
K. Berger, NTS 23 (1977), 180-204.
Μ. Black, as in (46).
Μ. P. Charlesworth, HThR 29 (1936), 107-32.
F. Cumont, RHR 91 (1925), 1ff.
K. Haacker, as in (56).
J. Jeremias, ZNW 27 (1928), 98-103.
G. Μ. Lee, NovT 9 (1967), 41.
S. Legasse, as in (54).
S. Lösch, ThQu 112 (1931), 295-319.
A. J. Mattill, CBQ 34 (1972), 142.
C. Maurer, TWNT 7.897-918.
J. C. O’Neill, NTS 35 (1989), 219-28 (224).
G. Schneider, in Bammel-Moule (1984) 403-14.
Μ. Stem, in Safrai-Stem 1.74-6.
R. D. Sullivan, ANRW II.8 (1977), 296-354.
C. H. Turner, JTS 3 (1902), 120-3.
W. C. van Unnik, in Kremer, Actes, 37-60.
B. Winter, JTS 42 (1991), 505-31.
P. Winter, NTS 3 (1957), 136-42.
5RSV, put them off.
6reek, them.
7NEB, under open arrest.
8There is textual authority for omitting this word.
 58. PAUL AND FELIX. 24.1-27. 1091

Commentary

The paragraph falls into two parts: Paul's appearance in court before
Felix (vv. 1—23); Paul’s private relations with Felix (vv. 24-27).
In the first part the Sanhedrin, represented by the High Priest and a
number of elders, travel to Caesarea in order to present their case
against Paul. This assumes a situation that has developed since 22.30,
where it seems that Claudius Lysias is making use of the Sanhedrin
as a fact-finding and advisory body. The Jews now bring an
accusation against Paul in the Governor’s court with specific charges
(vv. 5, 6), which however it is left to Felix himself to investigate and
substantiate. The Jews’ advocate, Tertullus, who begins with an
admirable captatio benevolentiae, must of course be supposed to
have made a much longer and more detailed speech. Paul also must
be supposed to have spoken at greater length; but between them the
two speakers make what Luke wished to be understood as the legal
points at issue. Tertullus claims that Paul, as the ringleader of the
Nazoraean sect, has been stirring up unrest among Jews all over the
world, and that he has attempted to profane the Temple. There is thus
at issue a matter of Jewish doctrine (for Paul is leader of a αἵρεσις,
with its own understanding of Judaism), but this leads to the Roman
crime of seditio and to an act which the Romans have agreed shall be
forbidden on pain of death. These accusations are repeated, explicitly
or implicitly, through the remaining chapters, and Schille (436) is not
wrong in describing the present scene as the ‘endgültiges Zusam-
menprall’ between Paul and Judaism. To all the charges Paul, on this
occasion, replies. (1) Of course it is true that he is a Christian, but
what Tertullus calls a αἵρεσις he calls a Way, and the Way consists in
the worship of the ancestral Jewish God and believing all the things
written in the Law and the Prophets: what Jew could object to that? It
includes belief in the resurrection. (2) He has been responsible for no
disturbances, in Temple, synagogue, or city. In Jerusalem there has
been no time since his arrival; as for the rest of the world, there is—
to say no more—no evidence. Jews of Jerusalem cannot supply it,
and the Jews of Asia (who, if there were evidence, might have given
it) were not present. (3) As for the Temple, he was indeed there
because he had come on a charitable mission to his people and was
engaged in sacrifice. He was in a state of purity; there was no
tumult.
No one is likely to maintain that Luke provides his reader with the
very words uttered in Felix’s court by Tertullus and by Paul. The
question that may reasonably be asked is whether Luke has fairly
represented the accusations brought by the Jews and the replies made
by Paul. The question whether the Gospel is or is not the fulfilment of
the Law and the Prophets runs not only through the whole of Acts but
through the whole of the NT. It turns in the end upon a point which is
1092 COMMENTARY ON ACTS

not here specifically mentioned: Was the crucified Jesus now alive or
dead? Was he or was he not the Messiah, that is, the person in whom
the essence of Judaism was crystallized and fulfilled? On the charge
of seditio Paul simply utters a denial. This was justified in the sense
that he had desired no disturbance; what he desired was that all Jews
should peaceably accept his version of Judaism. He had not desired
disturbance, but it could be reasonably maintained that he had
provoked it by insisting on a version of Judaism which the majority
of Jews saw not as the fulfilment but as the destruction of their faith.
The theological question and the social or political question were
closely related. On the question of Temple profanation Paul was on
stronger ground. If the story in Acts is correct, he had not—and
scarcely could have—committed it. If the story in Acts is incorrect,
this is most likely to be on the ground that Paul would not have
entered the Temple to take part in its activities at all.
Thus the Jews seem to be playing down the theological issue;
wisely, for it would not interest Romans, who would not be likely to
condemn a man simply as an erring theologian. They emphasize the
charges that would carry greatest weight in a Roman court. Paul on the
other hand emphasizes his Jewishness and responds to the other
charges with a flat denial and the challenge, ‘Produce your evidence.’
There seems to be a reasonable measure of probability on both sides.
‘Die Szene ist wohl in toto von Lukas gestaltet’ (Lüdemann 256);
not however without some foundation in historical tradition and in
knowledge of fact and procedure. ‘Sherwin-White [48] cites Momm-
sen’s opinion of Paul’s trial as “an exemplary account of the
provincial procedure extra ordinem” ’; so Hemer (129), who agrees.
It must however be added that there is not much procedure in the
narrative. Felix calls Tertullus, Tertullus speaks, Paul speaks, and
Felix postpones a decision.
Weiser (627) thinks that vv. 1-23 were built up by Luke on the
basis of (1) a Jewish delegation including Tertullus got only a
postponement; (2) Paul was accused of desecrating the Temple; (3)
he had come to bring alms (this, it is argued, must be traditional
because elsewhere Luke shows no knowledge of Paul’s collection).
We should add the charge of causing seditio, of which Luke must
have been aware. This means that though the scene is Luke’s creation
it is based upon and enshrines a historical basis. Cf. Roloff (336) and
Preuschen (137).
The second part of the paragraph tells of frequent interviews
between Paul and Felix, in which the latter’s wife Drusilla also took
part. Comparisons with the story of Herod and John the Baptist are
worth little. Felix is depicted much more favourably, and so is Dru-
silla; readiness to take a bribe is of course reprehensible but it is hardly
to be compared with rewarding a dance with a severed head. It is
somewhat more to the point to say that the judge who did not condemn
 58. PAUL AND FELIX. 24.1-27. 1093

Paul must have been (in Luke’s view) a good man, whose judgement
is therefore to be approved, but not so very good—he hoped for a
bribe, and because he did not get one left Paul in prison. Perhaps Luke
felt the need to explain why the judge who did not condemn Paul
nevertheless failed to release him. Felix’s character as depicted by
Josephus and Tacitus suggests a readiness to accept bribes; and he had
good reason to adopt any means of currying favour with the Jews.
There is something to be said for the view that in v. 27 two years
originally, in the information used by Luke, referred to the length of
Felix’s governorship, but was understood by Luke to refer to the
length of Paul’s imprisonment, which interested him much more. See
the note; also Haenchen (632f.); Lüdemann (258-60). The point is
important for the chronology of Paul’s life.

1. μετὰ δὲ πέντε ἡμέρας looks back to 23.35: Felix waits for

Paul’s accusers. A has τινὰς ἡμέρας. P74 reverses the word order:
ἡμέρας πέντε. The number five either has some traditional basis or is
Luke’s attempt to add an appearance of first-hand knowledge;
probably the former, τινάς or ικανάς, rather than a number, would be
in accordance with Luke’s manner.
κατέβη is geographically correct (from the mountains to the
coast), but the word was used as a technical term (cf. ἀναβαίνειν) for
a journey from the capital to any other destination; cf. e.g. 8.15.
ὁ ἀρχιερεύς 'Ανανίας. See 23.2.
πρεσβυτέρων τινῶν; <a> syp have τῶν πρεσβυτέρων, failing to see
the impossibility of all the elders’ appearing in Felix’s court.
<b>ήτορος Τερτύλλου τινός. To judge from the papyri the use of a
<b>ήτωρ, a barrister or advocate (Latin, causidicus), to represent partic-
ipants in a suit was a common practice. MM 563f. quote POxy 1.37.
col. 1.4f. (AD 49), Άριστοκλῆς <b>ήτωρ υπέρ Πεσούριος(?); 2.237.
col. 7.25 (AD 186), Δίδυμος <b>ήτωρ ἀπεκρίνατο μή χώρις λόγου τόν
Σεμπρώνιον κεκεινήσθαι; and refer to other papyri. It should not be
assumed that the practice was uniform or that professional advocates
could be used in every kind of court or every kind of case (cf. Begs.
5.320). Use was not mandatory and probably indicates the importance
or complexity of the case. It is interesting that the case referred to in
the second quotation above bears some resemblance to the complaint
made by the Jews against Paul: it is not (they allege) without ground
that they move against him. It is often supposed that Tertullus was
himself a Jew, no doubt a Greek-speaking Jew of the Diaspora. Knox
(Hell. El. 30, n. 2) compares Josephus’s journey to Rome (Life 13) on
behalf of a number of priests whom Felix had sent there λόγον
ὑφέξοντας τῷ Καίσαρι. The argument that Tertullus was a Jew rests
to a great extent on the first person plural ἐκρατήσαμεν in v. 6. But an
advocate would associate himself with his clients, and the reference to
‘all the Jews’ in v. 5 has been held to show that Tertullus was a Gentile
1094 COMMENTARY ON ACTS

(Stählin 294). One cannot be certain. If the need for an advocate was
linguistic rather than legal it implies that the High Priest and his
colleagues could not (or possibly would not) use Greek well enough to
use it in court. See Lieberman (Greek in Jewish Palestine) and Seven-
ster (Do you know Greek?), neither of whom, however, discusses this
passage.
The name Tertullus is derived from Tertius (as Catullus from
Catius, Lucullus from Lucius). It occurs in Pliny, Epistles 5.15(14).l.
The usus provincialis was regarded as good training for young
lawyers (Cicero, pro Caelio 30 (12)).
οἵτινες for oἵ; cf. 23.33.
ἐνεφάνισαν, here with dative and κατά; see 23.15. The plural
(rather than the singular, which would refer to Tertullus) is used (a)
because the complaint was made by the Jews as a body, and (b)
because it would have been made in the first instance in writing.

2. κληθέντος δὲ αὐτοῦ. The pronoun is omitted by B; for the

omission of the subject in a genitive absolute cf. 21.31. If the copyist
of B took the genitive absolute to be incorrectly used, referring to the
subject of the main verb (When Tertullus was called, he began ...),
he might have thought that the omission would slightly mend the
sentence, or at least make the solecism less evident. But it may be
right to take αὐτοῦ to refer to Paul (When Paul was called, Tertullus
began ...). We cannot say that Luke would never have used a
genitive absolute incorrectly; see e.g. 22.17. But it may be right to
give him the benefit of the doubt here. καλεῖν is used as a technical
term for the summoning of a witness (Plato, Laws 937a: ὁ δ’ εις
μαρτυρίαν κληθείς) or of any other participant in a legal process
(Demosthenes 19.211 (406): καλεῖν ἐμ’ εις τό δικαστήριον). The
active is used of the judge, the middle of the plaintiff; see LS 866 s.v.
14a, b. Begs. 4.297 quotes POxy 9.1204.13, κληθέντος Πλουτάρχου
κρατίστου ’Ισίδωρος (his representative) εἶπ[εν] ...
Tertullus begins his accusation with the customary captatio
benevolentiae; see Quintilian, De Institutione Oratoria 4.1; Cicero,
De Oratore 2.78, 79 (319-325); Lucian, Bis Accusatus 17; for πολύς
in such paragraphs, Thucydides 1.80.1; 2.35.1; 3.37.1; Dionysius of
Halicarnassus 5.1.4 (Conzelmann 131).
πολλής ειρήνης. It requires some exaggeration, or indeed imag-
ination, to find any peace in Palestine in the time of Felix. It could
however be claimed that many bandits had been put down (Josephus,
War 2.253-263).
διὰ σοῦ. There are parallels in which διά is used, but with the
accusative. Μ. 1.105 quotes PMagdola 16 and 20 where the peti-
tioner asks ἵνα διὰ σὲ βασιλεῦ τοῦ δικαίου τύχω, and comments that
‘if the humble petitioner had meant “through you”, he would have
addressed the king as a mere medium of power: referring to a
 58. PAUL AND FELIX. 24.1-27. 1095

sovereign power, the ordinary meaning “because of you” is more

appropriate.’ This belongs to the third century BC; die accusative
appears also in a close parallel to Acts in POxy 41 (Μ. 1.106; iii/iv
AD), πολλών αγαθών ἀπολαύομεν διὰ σαί [= σέ]. But at the
beginning of his speech Tertullus will not have intended to represent
Felix as a mere intermediary. In English, however, ‘through you’
need not imply disparagement.
διορθωμάτων: <a> has κατορθωμάτων, κατορθ. is a Stoic techni-
cal term for good actions, the opposite of αμαρτήματα. See SVF 4.81
(Index, s.v.) and 3.136 (Frag. 501; Chrysippus, in Stobaeus, Eclogae
2.96.18 (Ed. Heeren, 2.192)), ... κατορθώματα μὲν τὰ τοιαῦτα.
φρονεῖν, σοφρρονεῖν, δικαιοπράγειν, χαίρειν, εὐεργετεῖν,
εὐφραίνεσθαι, φρονίμως περιπατεῖν, πανθ’ ὅσα κατά τόν ορθόν
λόγον πράττεται. αμαρτήματα δ’ είναι ... Phrynichus 225 some-
what surprisingly rejects κατόρθωμα in favour of ἀνδραγάθημα.
διόρθωμα, on the other hand, means setting right, reform, often in
connection with laws. Thus Plutarch, Numa 17 (71), τό περί τόν
νόμον διόρθωμα. It is probable that διόρθωμα was the original
reading and reform its meaning, though to what reforms benefiting
the έθνος (here undoubtedly the Jewish nation; cf. vv. 10, 17; 10.22)
Tertullus refers we do not know. Delebecque (114): 'cette nation; le
démontratif semble indiquer que Tertullus n’est pas Juif. See
however above.
The participles τυγχάνοντες (since we enjoy; cf. 26.22; 27.3; Lk.
20.35 for this Attic use) and γινομένων (and since there have been)
are causal and introduce v. 3.
διὰ τῆς σῆς προνοίας. πρόνοια is prudentia; see 2 Macc. 4.6;
Rom. 13.14; (12.17). πρόνοια is one of the standard virtues of the
Hellenistic ruler. ‘Rühmend von der πρόνοια des Herrschers (nach
Dio Chrysostom Oratio 3.43 eine πρόνοια ανθρώπων κατά νόμον),
Feldherm oder Staatsmannes zu sprechen, wird im Hellenismus
stehender Brauch, vgl 2 Makk 4.6 ..., Epistle of Aristeas 30:
πρόνοιας βασιλικῆς οὐ τέτευχε, Diodorus Siculus 29 fragment 19
(von Hannibal)... PHerm 119B3.3 ...’ (J. Behm, in TWNT4.1006).
See also S. Lösch, ‘Die Dankesrede des Tertullus’, Theol. Quartals-
chrift 112 (1931), 310ff.; Μ. P. Charlesworth, ‘Providentia et
Aetemitas’, HThR 29 (1936), 107-32. There is further papyrus
evidence in ND 3.143. It is clear that Tertullus (or Luke?) knows the
proper style to use.
It is possible to connect πάντη τε καί πανταχοῦ (ν. 3) with this
verse, but the sentences seem better balanced if they are joined to
what follows. Not so, however, Page (236): great peace and
improvements everywhere.

3. πάντη (adverbs in –η appear in the NT only here and at 21.28)

τε καί πανταχοῦ, in every way and in every place. For the spelling
1096 COMMENTARY ON ACTS

see Μ. 2.84 and for the paronomasia see BDR § 488.1, n. 2 (quoting
Plato, Menexenus 247a, διὰ πάντος πάσαν πάντως προθυμίαν
πειρᾶσθε ἔχειν, but not referring to this verse). See also as examples
Aristotle, Nicomachean Ethics 1.10.15 (1101a); Philo, De Opificio
Mundi 61.
ἀποδεχόμεθα, we welcome—the peace and reforms mentioned in
ν. 2. It is (Begs. 4.297) a courteous word, and this is reinforced by
ευχαριστία, for which see I. Assos 7.22, an honorific decree quoted
in ND 1.83. There is a different Jewish estimate of Felix in Josephus,
Ant. 20.182. πάσης continues the paronomasia as well as the captatio
benevolentiae. Cf. 28.31.
κράτιστε: cf. 23.26. Φῆλιξ: so accented by NA26 in agreement
with Blass 251, ‘Φῆλιξ scrib., non Φήλιξ, quia ι et υ ante ξ
corripiuntur (κῆρῦξ, Φοῖνιξ).’ Against this, and it seems rightly,
Begs. 4.297: ‘Φήλιξ, not Φῆλιξ, pace Blass, for Felix is a Latin word
in which the last syllable is long (felῖcem), so that the analogy of
κῆρυξ and φοῖνιξ proves nothing.’
4. Tertullus continues the polite, ingratiating, manner of his
introduction. Other examples of an orator’s self-limitation are col-
lected by Betz, 118, n. 1.
ἐγκόπτω is normally hinder. This it can hardly be here, unless in
the sense of preventing Felix from attending to other business. The
Syriac and Armenian give the meaning weary; this is supported by
the fact that ἔγκοπος (Anthologia Palatina 6.33; but the use is
Septuagintal: Job 19.2; Isa. 43.23) means weaned. Begs. 4.298
translates detain, ἐγκόπτω is the reading of Ac B E L <a> co; κόπτω
is read by P74 A*vid (Ψ) 33 1175 1241 1891 pc. κόπτω means to
strike, in one sense or another; here it could perhaps mean to annoy.
τῇ σῇ ἐπιεικεία (B* spells ἐπεικεία; on various questions of
spelling in this word see Μ. 2.89, 314, 348). Moule (IB 45) notes but
does not explain the unusual dative. Clearly Felix is being asked to
listen and to act with έπιεικεία—towards the Jews. The word is hard
to translate. See H. Preisker (TWNT 2.585-7); further parallels in
Betz (209, n. 3). It denotes reasonableness, fairness, in general and
especially perhaps in a judge, who is prepared not to break the laws
but to give them an understanding, non-legalist interpretation. Most
significant is the personification of ἐπιεικεία in Plutarch’s Caesar 57
(734); καὶ τό γε τῆς ἐπιεικείας ιερόν οὐκ ἀπὸ τρόπου δοκοῦσι
χαριστήριον ἐπί τῇ πραότητι ψηφίσασθαι. Note the parallel with
πραότης.
5. The logical and grammatical connection implied by γάρ is not
immediately clear; the word may in part account for the long
interpolation between vv. 6 and 8—see on v. 8. The best explanation
of γάρ is that its sentence provides the basis of the appeal for a
patient hearing made in v. 4: You should heed our appeal for a
 58. PAUL AND FELIX. 24.1-27. 1097

careful hearing, for this is a very serious matter of grave concern to

the Empire. εὑρόντες looks back to the events of chs. 21, 22, but has
in mind also die whole experience of the Jews with the trouble-
making prisoner.
λοιμόν properly means a plague; it is transferred to human beings
both adjectivally (in the LXX especially as the rendering of various
combinations of e.g. 1 Kdms 1.16; 2.12) and substantially
(Demosthenes 25.80 (784)), pestilential, a pest. Here it could be
either a noun or an adjective complement, without significant
difference in meaning.
καί κινοῦντα: he is a pest, in that he stirs up. For καί used to
introduce the sense of a preceding word cf. BDR § 442.9b.
στάσεις (<a> sy sa have στάσιν): again ch. 22 is chiefly in mind, but
(from the Jewish point of view), other events earlier in Paul’s career
would be relevant, e.g. 17.6. Theological questions (such as resurrec-
tion) are now set aside; the issue now is seditio, which a Roman
court, unconcerned with theology, would be bound to take very
seriously, πᾶσιν (om. P74) ... κατά τήν οἰκουμένην emphasizes that
it is not the Jerusalem incident alone (see v. 6) that is complained of.
The Jews had become a universal people, to be found in every part of
the Empire, and a disturbance that threatened their peaceful relations
with the Roman administration would constitute the basis of a
movement that Felix could not ignore. Sherwin-White (51f.) draws
attention to this, referring (following F. Cumont, RHR 91 (1925),
3-6) to the letter of Claudius to the Alexandrians (see Η. I. Bell,
Jews and Christians in Egypt (1924), 23-6). The relevant passage
(lines 98-100) runs: ... εἰ δέ μή, πάντα τρόπον αυτούς
ἐπεξελεύσομαι καθάπερ κοινήν τεινα (sic) τῆς οικουμένης νόσον
ἐξεγείροντας. Luke’s λοιμός recalls νόσος in the letter. Sherwin-
White comments, ‘The similarity is deliberate. It is evident that the
narrative of Acts is using contemporary language. The charge was
precisely the one to bring against a Jew during the Principate of
Claudius or the early years of Nero. The accusers of Paul were
putting themselves on the side of the government. The procurator
would know at once what the prosecution meant.’
πρωτοστάτην τε. Again, the new clause elucidates the proceed-
ings, indicating the role in which Paul (according to the allegation)
had acted as trouble-maker. πρωτοστάτης was originally a military
term (a front-line man; so e.g Thucydides 5.71.2, ὁ πρωτοστάτης
τού δεξιού κερῶς; Job 15.24, LXX), here in the derived sense of
chief or leader. For αἵρεσις see on 5.17; the accusation makes Paul
the head of a dissident and troublesome Jewish party. See Fitzmyer
(Essays 276, n. 11). Delebecque (115), on πρωτοστάτης ‘... c’est le
soldat du premier rang; il fait partie des troupes de choc (cf.
Xénophon, Cyrop. 6.3.24). Paul est présenté comme un soldat
d’élite, donc dangereux.’ Maddox (70) goes further: ‘... it is
1098 COMMENTARY ON ACTS

precisely as such that Luke himself wants to portray him. He is more

important for what he represents than for his own sake.’ Papyri using
the word are cited in ND 4.244, but add nothing relevant.
The party is that τών Ναζωραίων. At 2.22; 3.6; 4.10; 6.14; 22.8
(see Rüger as cited on this verse); 26.9. Acts uses the term ’Ιησοῦς
(Χριστός) ό Ναζωραίος; here only do we have the plural describing
not Jesus but the Christians. If the evidence of Acts alone is
considered it is natural to suppose that the adjective Nazoraean was
first applied to Jesus (probably with the meaning ‘man coming from
the town of Nazareth’) and then came to be attached to his followers
(cf. ‘Christian’, 11.26). It is however important to bear in mind also
the patristic evidence for the existence of a Jewish sect bearing this
or a similar name (see especially Epiphanius, Panarion 29.6), Jewish
references to and the possible etymology of the Hebrew
word. See, inter alia, Black, Scrolls 66-74; P. Winter, NTS 3 (1957),
136-42; B. Gärtner, Die Rätselhafte Termini Nazaräer und Iskariot
(Horae Soederblomianae IV; 1957), 5-36; Kosmala (315). As the
last named points out, there is an interesting parallel between the
words of Acts 24.5 (τής τών Ναζωραίων αἱρέσεως) and the
combination of and in the additional ‘test’ Benediction
formulated c. AD 85 (see St John 127; Background 210f.).
suggests the plural of the Qal participle of which means to guard
or to observe; it is used with Torah in e.g. the saying of R. Meir
(c. 150) in Berakoth 17a: Keep my Torah in thy
heart. The suggestion thus lies ready to hand that there was a group
of Observants, headed perhaps by the Chief Observant. Such a group
could have existed independently of the Christian movement, which
was subsequently identified with it; or the name could have been
given to an early Christian, Jewish Christian, group which rever-
enced the Law. Later the name may have suggested a connection
with Nazareth. Such suggestions are of interest and have the
advantage that they may be used to explain Epiphanius’s belief that
there was a pre-Christian Jewish sect of Nazoraeans. They have the
disadvantage that Jesus does not appear to have lived in such a way
as to attract to himself the title of Law-keeper in Chief. There is
perhaps no better view than that the adjective (also in the form
Ναζαρηνός) was first attached to Jesus in view of his connection
with Nazareth, and was then passed on to his followers. It is however
probable that in the course of time various (false) etymologies were
attached to the name. See Η. H. Schaeder in TWNT 4.879-84; BA
1077.
Schille (432) makes the point that the reference to Nazoraeans is
intended to show that all the proceedings described are relevant not
only to Paul individually but to all Christians.
For πᾶσιν ... οικουμένην, gig has ‘non tantum generi nostro sed
fere universo orbe terrarum et omnibus Judaeis’, ‘doubtless the
 58. PAUL AND FELIX. 24.1-27. 1099

“Western” rewriting, and wholly in accord with the glossator’s

method elsewhere* (Ropes, Begs. 3.223).
6. ὃς καί adds a further charge. In addition to fomenting trouble
throughout the Jewish world Paul had tried to profane the Temple
(by bringing Gentiles into it, 21.28—where the word is the more
Jewish κοινοῦv).
ον καί, anacolouthon. These two words would be better omitted:
Having found this man ... we seized him. The author has forgotten
εὑρόντες (v. 5), and supposes that he had written ηὕρομεν.
The charge of desecrating the Temple could be brought into a
Roman court because the Romans had accepted the provision for
maintaining its sanctity (Stählin 295). Disturbances caused by Paul
as leader of the Nazoraeans and desecration of the Temple: ‘Beide
Punkte enthalten religiöse und politische Brisanz.’
On the text of this verse see on v. 7.
7. A number of MSS (Ε Ψ 33 (323 614) 945 1739 (2495) pm gig
vgcl sy(p)) contain a considerable paraphrase given in some printed
copies of the NT as the end of v. 6, v. 7, and the beginning of v. 8.
The whole sentence runs as follows.
καί κατά τόν ἡμέτερον νόμον ἠθελήσαμεν κρῖναι (κρίνειν,
614 2495 pc), παρελθών δέ Αυσίας ό χιλίαρχος μετά πολλής
βίας ἐκ τῶν χειρῶν ἡμών ἀπήγαγεν, κελεύσας τούς
κατηγόρους αὐτού ἔρχεσθαι ἐπί (πρός, Ε 2464 pc) σε.
See Clark (xlvii); Ropes (Begs. 3.225); Delebecque (115).
This describes the events of 22.22—23.30 from the Jewish point
of view and in very compressed form. An editor no doubt thought
that some such cross-reference was necessary in order to explain
what was taking place. Luke’s narrative, written from a non-Jewish,
or anti-Jewish, point of view certainly suggests something other than
a peaceful Jewish trial according to law—a lynching, rather; and,
according to Luke’s story, the tribune was not the first to use
violence. Is Luke (or his editor) deliberately setting out to represent
the Jews as guilty of falsehood? Is he himself guilty of misrepresent-
ing in his narrative Jewish intention fairly set out in Tertullus’s
speech? Cf. the letter ascribed to Claudius Lysias. Probably neither;
he is using traditional material which reflects two points of view.
Dibelius (151, n. 32), cautiously suggests that the long text is
original.
For μετά ... βίας cf. 5.26 and see BDR § 198.4, n. 5 (classical
usage would have required βίᾳ, or πρὸς βίαν).
ἐπί σε: cf. 23.30, ἐπὶ σοῦ. These words are contrasted with κατά
τόν ἡμέτερον νόμον.
8. The sequence of relatives is continued: ὅς καί ... ον καί ...
1100 COMMENTARY ON ACTS

(v. 6)... παρ’ σὖ ... Tertullus leaves cross-examination to the judge

(αὐτός ἀνακρίνας). The verb is used as at 4.9; 12.19. Ί1 s’agit d’un
procès régulier “extra ordinem” devant le procurateur, qui doit
entendre accusateurs, accusé, témoins, et faire lui-même son enquêt,
avec son pouvoir discrétionnaire’ (Delebecque (115), pointing out in
the same context that περὶ πάντων τούτων depends on ἀνακρίνας
not ἐπιγνῶναι). κατηγορεῖν takes two genitives, as for example in
Demosthenes 21.5 (515).
The long Western addition to the text (see on v. 7). has the effect
of causing παρ’ σὗ to refer not to Paul but to Lysias. This is defended
by Clark (xlvii), who observes that ‘no evidence could be so cogent
as that of the Roman official who had made the arrest, while Paul
would be a suspect witness when speaking about himself’. Haenchen
(625), on the other hand, thinks that the Western editor thought it so
foolish that the truth should be arrived at by the examination of Paul
that he made the change that introduces Lysias as witness. Probably
both underestimated the power of an examining magistrate to get the
truth out of an unco-operative witness, ἀνακρίνειν in fact describes a
process that would be applied to Paul rather than to the tribune.
Haenchen (629), Schneider (2.344), Weiser (626), and others,
quote the opinion of Lösch that Tertullus’s speech is ‘ein Meister-
stück von ausgesuchter rhetorischer Kleinkunst’ (op. cit. (p. 1090).
p. 317).
9. συνεπέθεντο, joined in the attack. The word is a military term;
its use implies that Paul is regarded as a public enemy.
καί οἱ ’Ιουδαῖοι suggests, but does not prove, that Tertullus was
not himself a Jew. Haenchen draws this conclusion, but it is possible
to take the sentence as, So said Tertullus, their representative, and so
said the Jews themselves. See on v. 1.
ταῦτα οὕτως ἔχειν. For the expression see 7.1; 17.11.
614 2147 syh** fill out the picture, having, instead of συνεπ. δέ,
εἰπόντος δὲ αὐτοῦ ταῦτα συνεπέθεντο.
10. Paul replies when the governor gives him leave to do so.
νεύειν means to give a signal, which will be interpreted in accord-
ance with the circumstances. Here it is followed by an infinitive,
λέγειν; the same construction is used by Euripides, Hecuba 545.
Homer, Iliad 8.246 has accusative and infinitive.
In place of the simple λέγειν, syhmg has defensionem habere pro se
statum autem assumens divinum dixit
There is no other witness to these words, but Clark (153, 232) takes
them to represent the Western text, which he gives in Greek as
ἀπολογίαν ἔχειν ὑπέρ εαυτοῦ· ό δέ σχήμα ἔνθεον ἀναλαβὼν ἔφη.
If Clark is right in this, but wrong in thinking the Western text
original, the view is confirmed that sees in the Western text some
analogies to the development of apocryphal acts; cf. Acts of Paul and
 58. PAUL AND FELIX. 24.1-27. 1101

Thecla 3 (L.-B. 1.237), ποτὲ δὲ αγγέλου πρόσωπον εἶχεν. Cf. 26.1

(syhmg again); 6.15.
Like Tertullus (v. 2 and note), Paul begins with a captatio
benevolentiae, not unduly straining the truth. For Felix, see on 23.24.
ἐκ πολλῶν ἐτών may seem to exaggerate. A two-year term in
senatorial provinces was usual, and Hemer (173) draws attention to
the relevance of the ‘two years’ of v. 27, if this is taken to refer to
Felix’s proconsulship. There is however some ground (see above, pp.
1080f.) for thinking that Felix had been active in Palestine, as a
junior colleague of Cumanus, before his appointment as procurator.
For any exaggeration cf. Claudius’s ἐκ πολλῶν χρόνων in PLond
1912, line 22 (Bell, p. 23); see also Schille (433). That he had been a
judge is no exaggerated statement; δίκαιον, inserted after κριτήν by
Ε Ψ 323 614 945 1175 1739 2495 pm syh, is not an original part of
the text. ὄντα σε ... ἐπιστάμενος: accusative and participle are here
correctly used, whereas in Hellenistic Greek the participle was often
replaced by the infinitive, or a ὄτι clause (Μ. 1.229). On the tense of
ὄντα see Moule, IB 101; it is due simply to the paucity of participles
of εἶναι.
εὐθύμως, cheerfully; εὐθυμότερον (<a>) adds a little liveliness to
the narrative.
τὰ περὶ έμοῦ my case, is an accusative of respect: I make my
defence (for ἀπολογεῖσθαι see 19.33) in relation to my case. The
argument, according to Conzelmann (132), is not that Christianity
and Judaism are essentially the same, but that there is no case to
answer. Calvin (2.248) defends Paul’s defence. He had no time to
expound the Gospel positively and confined himself to answering the
false assertions that had been brought against him.

11. δυναμένου σου, causal, since you are able to perceive ...
δτι is best taken (as in the NA26 text) to mean that, introducing an
object clause in dependence on ἐπιγνῶναι. An alternative possibility
is to place a stop after ἐπιγνῶναι, which then has no object—you are
able to come to a judgement; ὅτι is then for.
The construction of the ὅτι clause is obscure, though the general
drift is clear. It is not more than twelve days since Paul reached
Jerusalem (21.15)—little enough time for the crimes of which he is
accused. With πλείους must be understood ἡμέραι: there are not to
me (dative of possession: BDR § 189.1, n. 1) more [days] than
twelve days. The comparison ought to be expressed either by ή with
the same case (here nominative) or by the genitive. In fact we have
the nominative without ή. The error is caused by the emphasis in the
writer’s mind on the twelve-day period. One might imitate: It is not
more than—well, it is (only) twelve days since ... For the omission
of ή before a numeral Bruce (1.424) compares the Latin omission of
quam; the parallel though interesting is probably no more than a
1102 COMMENTARY ON ACTS

coincidence. Moule (IB 202) notes that it would be possible to take

ἡμέραι as subject and understand ήμερων with δώδεκα; possible,
but hardly probable. On the calculation of the twelve days, see
below.
ἀφ’ ἧς stands for ἀπὸ τής ημέρας ή ...
ἀνέβην, the word of pilgrimage; see on 11.2.
προσκύνησών, future participle expressing purpose; cf. 8.27. This
cannot be said to be a complete statement of Paul’s purpose in
coming to Jerusalem; he came as the bearer of a gift from the Gentile
world (cf. ν. 17), and probably with a view to some form of
consultation with the leaders of the Jerusalem church.
εἰς ’Ιερουσαλήμ may be taken with ἀνέβην: I came up to
Jerusalem; more probably, εἰς is used for ἐν, as frequently: ... to
worship in Jerusalem. Calvin (2.249) has great difficulty with Paul’s
intention to worship in the Temple. Luke has none: partly because he
is not as systematic and logical a theologian as Calvin, partly because
it is his intention to represent Paul the Christian as continuing to be a
good Jew. On his attitude to the Temple see FS Bammel 361-7.
Others find difficulty in the chronology of the twelve days and
their relation to the narrative up to this point. Bauemfeind (263), for
example, thinks that the number may come from a source different
from that of the preceding narrative, or that it may be the result of
adding together the five of ν. 1 and the seven of 21.27. Wettstein
(2.621) is still worth noting: ‘Primus dies is est, quo Hierosolyma
venit, secundo lacobum vidit. IV sequentes circa purificationem fuit
occupatus, biduo detentus captivus Hierosolymis, V. postquam Cae-
saream venisset die, caussam dicit i.e. decimo tertio’—which pre-
sumably Wettstein reckons a twelve-day interval. Pesch (2.257) finds
the simple 5 + 7 = 12 calculation unsatisfactory and gives the
following (Bruce 2.443 is similar):
1. Arrival in Jerusalem (21.17)
2. Visit and talk with James (21.18)
3-9. Seven-day purification process (21.27)
9. Arrest (21.33)
10. Paul before the Sanhedrin (22.30)
11. Plot against Paul (23.12)
12. Arrival in Caesarea (23.32)
It is questionable whether the last three items can be accommodated
within three days, and the timetable takes no account of the five days
in Caesarea. Further, the question must be considered whether the
twelve days refer to time up to the moment of speaking or up to the
point at which Paul came under Roman guard (21.33)—the Romans
would know very well what crime, if any, he had committed since
then. Thus Begs. 4.300, ‘In spite, therefore, of the εὲσί it is possible
that the phrase means “I had not been twelve days in Jerusalem
 58. PAUL AND FELDC 24.1-27 1103

when the trouble arose” there is much to be said for this, but it is
not clear why Begs, continues, ‘and the number is merely a literary
addition of “seven” in 21.27 and “five” in 24.1.’ Surely Luke
cannot have failed to see that the five days of 24.1 are placed after
‘the trouble arose’. Bauemfeind (see above) may be right in thinking
that ‘twelve’ is derived from another source, or it may be a Lucan
approximation.
12. On the sequence ούτε ... ούτε ... ούτε ... ουδέ (ν. 13) see
BDR § 445.2, η. 3; the final negative gives the sense ‘noch
überhaupt’. But the second ούτε and the third are correlated with
οὔτε ἐν τω ἱερῷ, and in English it is probably best to put them
together: Neither in the Temple, nor in the synagogues, nor in the
city did they find me ... The reference to the Temple is clear. The
synagogue could refer to the synagogues of the Dispersion, but the
reference in v. 11 to the brevity of Paul’s stay in Jerusalem, and the
reference to the city that follows, suggests that it is Jerusalem
synagogues that are in mind. κατὰ τήν πόλιν is throughout the city;
with the negative, nowhere in the city.
πρός τινα διαλεγόμενον, arguing with anyone—the first step
towards creating a disturbance.
Zerwick (§ 227) insists upon the full force of the active ποιούντα,
provocantem concursum (the middle would mean simply taking part
in such a gathering). For ἐπίστασιν, <a> has ἐπισύστασιν; the former is
a word of various and uncertain meaning; followed by όχλου it must
mean the collecting of a crowd, or perhaps a little more—the onset of
a crowd (see LS 659; BA 607—Andrang, Ansturm). Cf. (also for the
variant) 2 Cor. 11.28 (with the note in 2 Corinthians 301).
13. περί ὧν for ταῦτα περὶ ὧν: they cannot prove, substantiate, to
you those things concerning which ... The accusations are of course
those of vv. 5, 6. Paul says nothing about the charge of causing
world-wide disturbance (a) because no proof was offered and (b)
because Felix could try only crimes committed within the area of his
judicial authority. So Pesch (2.257f.).
ουδέ, in B 81 pc. P74 A Ε Ψ <a> have ούτε, following the
threefold occurrence of that word in v. 12, but missing the point
made there (see the reference to BDR).
For νυνί, P74 E <a> have νυν. In Acts νυν occurs 25 times, νυνί
twice (including this verse). Copyists changed the less into the more
familiar.
14. ομολογώ, I confess, or admit, more often with an infinitive
(e.g. Aristophanes, Knights 296, ὁμολογῶ κλέπτειν) than as here
with ὅτι. But ομολογεῖv here may be not quite confess. The ὅτι
clause (for details see below) contains a double assertion, one half of
which could be regarded as an admission whereas the other protests
1104 COMMENTARY ON ACTS

innocence, or indeed makes an affirmation (cf. 23.8). The word σοι

may be significant: To them (the Jews) I maintain that I am nothing
but a good Jew; to you I don’t mind admitting that...
κατὰ τὴν οδόν... τφ πατρῴῳ θεῷ. The beginning and end of this
clause are both emphatic. On the one hand, it is the ancestral Jewish
God, the God of the fathers, the God of the OT, that Paul worships,
no new deity whom he has discovered or invented for himself. This
will be underlined in the following words. On the other hand, he
worships the old God in the new way, κατά τήν οδόν, that is, in
conformity with the (Christian) way. For οδός in this sense see 9.2.
λατρεύειν is used in the summary of the OT at 7.7, 42; also at 26.7;
27.23. Behind these occurrences is the Jewish For τῷ πατρῴῳ
θεῷ cf. Sophocles, Antigone 838, θεών πατρώων; Vergil, Aeneid
9.247, di patrii; cf. Thucydides 2.71.3.
Paul (Luke) accepts the word οδός as a description of Christianity.
Indeed we see here (so Haenchen 630) why Luke is glad to use the
word. ‘Dieser Begriff bezeicnnet die neue Jesusreligion als eine
eigene Grösse und reisst sie trotzdem nicht vom Judentum los: er
erinnert ja aufs stärkste an alttestamentliche Wendungen wie “die
Wege des Herrn”, die das Judentum als die gelobte wahre Religion
hinstellen. Dieser Weg hat Paulus nicht aus dem Judentum hinausge-
führt ...’ αἵρεσις however is a word that belongs to Jewish
opponents: ἣν λέγουσιν αἵρεσιν, used for example by Tertullus (v.
5). The implicit disavowal of αἵρεσις means that Christianity regards
itself not as a sect or group within the people of God; it is the people
of God, and its way is the halakah for all Israel. ‘Hanc appelationem
Paulus corrigit, non quo tum esset odiosa, sed quod non satis digna’
(Bengel 477).
That the God Paul worships is indeed the ancestral, OT, God is
proved by the fact that Paul believes πᾶσι τοῖς κατά τόν νόμον καὶ
τοῖς ἐν τοῖς προφήταις γεγραμμένοις. The general sense of this
clause is clear, but its construction is affected by the omission of ἐν
τοῖς by A <a>. The shorter text is somewhat simpler; the first τοῖς
must be linked with γεγραμμένοις: all the things that are written
according to the Law and by (in) the prophets. If the longer text is
read it means that a distinction is being made (as well as an
association) between all the things that are according to the Law and
the things that are written in the prophets (i.e., in the prophetic
books— as a section of the OT). That Paul should believe what
was written in the prophets gives rise to no great difficulty; the next
verse gives the outstanding example. Could he have claimed to
believe all the things according to the Law? Only if he were allowed
to give some of them a new interpretation, to understand circumci-
sion, for example, as circumcision of the heart, in spirit not letter
(Rom. 2.29). On the representation in Acts of Paul as a faithful and
devout Jew see Introduction, pp. xc, ci.
 58. PAUL AND FELIX. 24.1-27. 1105

Stauffer (Theologie 220) takes this passage with 14.15; 17.23ff.;

22.14 as ground for the proposition, ‘Die Herrlichkeit und Allein-
herrschaft des Einen Gottes kommt durch das Christusereignis erst
voll zur Offenbarung und Durchsetzung.’ This is certainly true in the
sense that Paul (Luke) never thinks of two Gods, one of the OT and
one of the NT, or that the God spoken of in the OT is anything other
than the God revealed in the New. But Stauffer (228) goes too far in
seeing in this verse a ‘Bekenntnis zur Kirche’, ‘ein Treubekenntinis
zum verfolgten Schar Christi’, if he means more than that Paul
acknowledges that he is a member of the persecuted group.
Again (see on v. 11) Calvin (2.251) is in great difficulty, com-
pounded by the fact that he does not see that οδός refers to
Christianity. Weiser (629) sees Luke’s intention clearly: ‘Durch
diese Argumentation soll deutlich werden, dass der “Weg”, d.h. das
Christentum, weder eine verengte oder fehlgeleitete und von Israel
abgefallene jüdische Sondergruppe ist, sondern wahres Israel.’
15. Not for the first time (cf. 23.6), Paul singles out belief in the
resurrection as the central feature of his faith, which it shares with
(Pharisaic, not Sadducean) Judaism. It is implied that he ought not to
be persecuted for believing what all good Jews believe. There is no
reference here to the resurrection of Jesus; it is the general resurrec-
tion that is in mind.
The word ελπίς has slightly different meanings as it is governed
by ἔχων (cf. Thucydides 8.48.1, πολλὰς ἐλπίδας εἶχον), when it
means hope itself, the confident expectation that something will
happen, or by προσδέχονται, when it means the thing hoped for (id
quod speratur—Blass 254); they themselves (the Jews) await,
expect, the realization of hope. They and Paul believe that there is to
be (μέλλειν ἔσεσθαι, correct use of the future infinitive with μέλλειν;
the NT often uses the present infinitive) a resurrection (Ε Ψ <a> sy
unnecessarily add νεκρών). On this occasion it is explicitly added
that the resurrection will be of both the righteous and the unright-
eous, that is, for judgement. In many passages of the NT the fate of
the unrighteous is left unclear because only the righteous are spoken
of and only their resurrection is explicitly affirmed. Cf. Dan. 12.2;
Rev. 20.11-15. See also StrB 4.2.1166-1198, the Excursus, Allge-
meine oder Teilweise Auferstehung der Toten?—There was a great
variety of opinion.
16. ἐν τούτφ is difficult to interpret precisely. The antecedent of
τούτῳ may be the preceding clause, that is (the fact that) there is to
be a resurrection of righteous and unrighteous, ἐν is probably causal
(because of this; see e.g. Xenophon, Cyropaedia 1.3.14) though it is
hardly necessary to suppose that it is in any direct sense dependent
on the use of the Hebrew Cf. Jn 16.30. Many take the phrase this
way; thus Bruce (1.425) (= therefore); Schneider (2.348) (darum,
1106 COMMENTARY ON ACTS

aus diesem Grund); Zerwick (§ 119) (propterea, cf. 7.29); BDR §

219.2, n. 2 (deswegen). It may however be better not to give τούτῳ a
precise antecedent. Moule (IB 79, 132, 197) renders, and so, that
being so (cf. the use of the similar expression with the relative, ἐν ᾧ;
also 26.12, ἐν οἷς).
καί αυτός: ‘ “I also”, as well as my accusers and the Jews whom
they represent’ (Page 239). ασκώ (ND 3.153 refers to H. Dressier,
The Usage of ἀσκέω and its cognates in Greek Documents to 100 AD
(1947); see also FS Furnish 161-72) refers in the first instance to the
training of the body for various skills and athletic pursuits but was
easily adapted for intellectual and moral training; e.g. Xenophon,
Memorabilia 1.2.19-28: ὁρῶ γὰρ ώσπερ τὰ του σώματος έργα
τούς μὴ τὰ σώματα ἀσκοῦντας οὐ δυναμένους ποιεῖν, οὕτω καὶ τὰ
τής ψυχής έργα τούς μή τήν ψυχήν ἀσκούντας οὐ δυναμένους
(19). Neither the verb, nor the noun ἄσκησις (used by Xenophon in
the passage referred to), occurs elsewhere in the NT; the thought of 1
Cor. 9.27 is not exactly parallel. Nearer are the passages in which
Paul exhorts his readers to imitate him, as he imitates Christ (1 Cor.
4.16; 11.1; 1 Thess. 1.6; cf. 2 Thess. 3.7,9). Cf. also passages in Paul
and elsewhere in which the Christian life is compared to the training,
discipline, and effort of the athlete (e.g. 1 Cor. 9.25f., 1 Corinthians
217f.). The relation between moral discipline and faith is a theme
that is not worked out in Acts, or indeed by Paul himself in the
certainly genuine letters.
ἀπρόσκοπον συνείδησιν. For Paul’s clear conscience cf. 23.1;
also 20.20, 27, 33; also however 1 Cor. 4.4, οὐκ ἐν τούτῳ
δεδικαίωμαι. ἀπρόσκοπος may mean not stumbling or not causing
to stumble. At 1 Cor. 10.32 the latter meaning is undoubtedly correct
(άπρόσκοποι... Ίουδαίοις ... Έλλησιν ... τή ἐκκλησία); at Phil.
1.10 the former is probably though not certainly correct. The present
passage is nearer to 1 Cor. 10.32, but that passage does not introduce
the word συνείδησις, as this one does. The meaning seems to be that
Paul’s intention, the aim of his self-training, is that his conscience
shall not accuse him of offending against God or man. Since the
conscience (see Romans 50f.; C. Maurer in TWNT 7.897-918) was
thought of as an agent that passed judgement on actions already
committed this seems to be the probable meaning of ἀπρόσκοπος;
alternatively it would be possible to think of a conscience that guided
him without stumbling in his relations with God and men. Begs.
4.302 refers to papyrus evidence for the meaning unharmed; see MM
72, also ND 1.55: ‘In the papyri ἀπρόσκοπος εἰμί is an idiomatic
way of saying, "I’m all right”.’ Perhaps simply, therefore, a good, or
sound, conscience. If ἔχων (<a> gig) is read, the balance of sentence
and thought changes. For πρὸς τόν θεὸν καί τούς ἀνθρώπους, cf.
Philo, Abraham 208.
διά παντός is rather surprisingly added at the end of the sentence;
 58. PAUL AND FELIX. 24.1-27. 1107

E 614 1739v.1. 2495 pc gig improve the word order by placing it after
ἔχειν. The meaning is clear.

17. Paul resumes his defence. He has pointed out that his beliefs
are those of a good Jew and that in practice he keeps his conscience
clear. It has been alleged (vv. 5f.) that he is a trouble-maker and has
profaned the Temple. What are the facts?
He has been absent from Jerusalem for a long time. On διά in the
sense of after see Moule, IB 56; the meaning is ‘common in Greek
authors’, e.g. Aristophanes, Plutus 1045, διὰ πολλοῦ χρόνου. Cf.
Mt. 26.61; Gal. 2.1. πλειόνων, comparative, is used in the sense of
the superlative, very many years. The observation in Begs, that in
Modern Greek the comparative with the article is used to express the
superlative (as in French) is scarcely relevant since here there is no
article. No explanation is necessary. ‘Used alone, [the comparative
degree] often expresses excess or tendency, and may be rendered by
too, very, rather, somewhat' (Simonson, § 1340). So e.g. Schneider
(2.348), ‘im klassischen Sinn, “nach etlichen Jahren" ’. It is not
clear from what point the ‘many years’ are calculated. Paul reached
Jerusalem at 21.15; his previous visit had been that of ch. 15, or
perhaps at 18.22 (if this verse does refer to a visit to Jerusalem; see
the note). For the dates of these events see Introduction, pp. liv-lxii;
neither interval could be naturally described as covering ‘many
years’, and it may be said that what is in mind is the interval since his
conversion, when Paul broke off his earlier relations with official
Judaism in Jerusalem, πλειόνων may be an exaggeration; cf. ἐκ
πολλῶν ετῶν in ν. 10. Or it may be a real comparative and refer to
this πολλών: My last visit to Jerusalem was before you, Felix, took
office.
παρεγενόμην, I arrived—in Jerusalem.
ἐλεημοσύνας ποιήσων. The future participle expresses purpose; cf.
ν. 11. ἐλεη. ποι. is not characteristically Greek, but probably reflects
both and In late Hebrew (cf. Aramaic)
came more and more to have the meaning of alms (see Jastrow
1263f.). ἐλεη. ποι. occurs in Tobit and Sirach; see also Gen. 47.29
(where the Hebrew is Paul has now visited Jerusalem in
order to bring and distribute alms—
εἱς τὸ ἔθνος μου. εις (connecting not with ποιήσων but with
έλεημοσύνην—Delebecque (117); but see also Weiser 629f.) must
mean for the benefit of ; that it should thus replace the dativus
commodi is discussed by Deissmann (BS 117f.) with OT and papyrus
parallels (see also MM 186f.) and reference to Mk 8.19f. and—
especially important because these deal with Paul’s collection—
Rom. 15.26; 1 Cor. 16.1; 2 Cor. 8.4; 9.1, 13. See also Μ. 3.236 and
Zerwick (§ 51). In Acts we have heard nothing (since 11.27-30, a
mission completed in 12.25) of any intention on Paul’s part to give
1108 COMMENTARY ON ACTS

alms to his nation, the Jews, or, for that matter, to anyone. We know
however from the epistles (Romans, 1 and 2 Corinthians) that at this
point in his ministry he was deeply concerned in a plan to bring relief
to the poor saints in Jerusalem, that is, to the Jewish Christians of the
mother church. There can be no doubt that this collection (see D.
Georgi, Die Geschichte der Kollekte des Paulus für Jerusalem
(1965); K. F. Nickle, The Collection (1966); Romans 2f., 255f.; 1
Corinthians 23, 385-7; 2 Corinthians 25-8, 216-42) was a matter
of the greatest importance to Paul, and it is very surprising that Acts
should pass it by in almost complete silence, noticing it only in this
inaccurate reference. We must conclude either (1) that Luke was less
well informed about Paul’s work than one would expect a companion
to be, or (2) that he found the collection less interesting than Paul did,
or (3) that there was some good reason for suppressing it. Thus Bruce
(2.445): ‘Luke evidently knew about the collection but, equally
evidently, he is very reticent about it. This may have been because
the enterprise ended in disaster; another possible reason is that at
Paul’s trial it was misrepresented as an improper diversion of money
that ought to have swelled the Jerusalem temple tax, and Luke judged
it wise to refer to it only in the most general terms.’ In view of Luke’s
silence elsewhere Roloff (338) is right to infer that this reference
comes from tradition; it is not Luke’s invention.
Paul came to Jerusalem with alms and προσφοράς, in the LXX
but seldom elsewhere, (sacrificial) offerings. This is the meaning at
21.26—Paul paid the expenses for the sacrifices due from the men
who had taken vows. It is to be noted however that those sacrifices
were asked of him by James and his colleagues after he reached
Jerusalem; there is nothing to suggest that he came to Jerusalem with
the intention of making (ποιήσων) offerings, though silence of
course does not in itself prove that he had no such intention. His
epistles say nothing to suggest that he would have been likely to plan
participation in Temple worship. The words καί προσφοράς, at the
end of the sentence, look rather like an addition or afterthought; they
are however unquestionably part of the text.

18,19. ἐν αἷς] ἐν οἷς, L 323 326 1241 pm. The neuter will refer to
the preceding sentence, and the circumstances it describes, in general
and is thus the easier reading: ‘Meanwhile, they ...’ αἷς, feminine
plural, can only refer to προσφοράς: While I was engaged upon
these sacrifices ...’.
ἡγνισμένον, not so much ‘having been purified’ as ‘being in a
state of purity’ (perfect participle). See 21.29 (also 21.24, ἁγνίσθητι
σὺν αύτοῖς).
There was no crowd and no tumult. Cf. v. 12.
‘δέ after τινές is opposed to the emphatic οὐ μετά ὄχλου ουδέ μ.
θ.—“I was not creating disturbance, but certain Jews from Asia
 58. PAUL AND FELIX. 24.1-27. 1109

(brought an accusation to that effect).” Before stating what the

accusation was Paul proceeds parenthetically to comment on the
absence of his accusers as indicating the falsity of their charge.
Strictly he ought to have gone on to state what the accusation was,
instead of doing so however he breaks off, and, turning to the
Sanhedrists, says “or let these now say what I was proved guilty
of", the force of “or” being this—“The absence of my original
accusers shows that they had no case, or, if this inference is objected
to, then let these men themselves (though their evidence is only
second-hand) say what...” ’ (Page 239f.). This exposition assumes
punctuation (as in NA26) with a full stop at the end of v. 19. This
means that the nominative τινες ’Ιουδαῖοι is left without a verb. A
less probable alternative is to place a comma at the end of v. 19 and
use the verb εἰπάτωσαν out of v. 20: Let the Jews of Asia speak, or
failing them, let those Jews now present themselves speak. Dibelius
92 (cf. Ropes, Begs. 3.225) draws attention to the possibility that a
few lines may have dropped out at this point. After θορύβου, perp2
and some Vulgate MSS add et apprehenderunt me clamantes et
dicentes, tolle inimicum nostrum, an addition which ‘seems to be
proved ancient by the reference in Ephrem’s commentary’ (Ropes).
It is however no more ancient than many other Western additions,
and the text of all Greek and almost all other authorities should be
accepted.
οὓς ἔδει (δει, L 1241 pm sa) ἐπὶ σου παρεῖναι. This was a valid
point.

The Roman law was very strong against accusers who aban-
doned their charges. Claudius himself had been busy with
legislation aimed at preventing accusers within the system of
ordo from abandoning their charges. He made a speech about
the matter in the Senate, and his proposals were later completed
by the SC Turpilianum of AD 61, under Nero. This laid down
penalties for the offence which the lawyers call destitutio ...
Once again, the author of Acts is well informed. But there is
more to it than that. The disappearance of one set of accusers
may mean the withdrawal of the charge with which they were
particularly associated. The Asian Jews had accused Paul of two
things: one, teaching everywhere, i.e. throughout the ‘world’,
the oikoumene, against the Hebraic law, and two, of bringing
Hellenes into the Temple. Charge one was taken over by the
Jewish clergy. Charge two, according to Acts, could not be
substantiated. ‘They had seen Trophimus with Paul in the city,
and thought he had been taken into the Temple.’ Hence when
the Asian Greeks [sic Sherwin-White; but surely he must mean
Jews?] withdrew from the case, Paul had a sound technical
objection to put forward. (Sherwin-White 52f.)
1110 COMMENTARY ON ACTS

Luke cannot however report Paul’s release. The convincing legal

argument may have been Luke’s, not Paul’s. And could not the Jews
of Jerusalem have said, ‘Jews are Jews, whether from Asia or
Jerusalem; it is Jews who bring charge one, and we are representative
Jews’?
On the absence of ἄν in the apodosis Moule (IB 149) writes, ‘ἄν is
usually, if not always, omitted with verbs whose very sense implies
obligation, necessity, possibility'.
εἴ τι ἔχοιεν. The optative gives, and was doubtless intended to
give, the impression of literary effect (Zerwick § 356). It is not
however correct; better would have been εἴ τι ἔχουσιν or ἐάν τι
ἔχωσιν (BDR §385.2, η. 4). Έη bon grec, εἴτις est l’équivalent d’une
relative exprimant une totalité; mais l'optatif peut surprendre’ (Dele-
becque 118). Similarly Radermacher (132), with the interesting
observation, ‘Die Reaktion gegen die Volkssprache, die den Optatif
aufgibt, hat die Literaten dazu geführt, ihn unter ihre besondere
Protektion zu stellen.’ Knowling (485), however, takes the optative
seriously: ‘the optative of subjective possibility, representing the
subjective view of the agent—if they had anything against me (in
their own belief)’.
20. Paul proposes in this verse to introduce (and in the next verse
introduces) a new charge that might possibly, but unjustifiably, be
brought against him. For the grammatical relation of this verse to the
preceding see on v. 19. If the Jews of Asia choose not to appear in
order to bring the accusation which only they can bring since they are
the supposed witnesses of the supposed offence, let those Jews who
are present speak of what happened when I appeared before their
own Council. Let them say what offences they found (in me, ἐν ἐμοί,
is added by C Ε Ψ (945 1739 1891) <a> lat sy bo—sensible but
somewhat superfluous and not easily combined with the genitive that
follows) when I stood before (ἐπί, in the presence of) the Sanhedrin.
See 22.30-23.10.
For ἤ at the beginning of the verse P74 AC pc have εἰ, a phonetic
itacism.
στάντος: Calvin (2.253) notes the tense, complaining that Erasmus
and the Vulgate (cum stem) treat it as a present. They should have
cum steti.
21. This verse recalls 23.6. The Jews have nothing to complain of
but one cry (φωνή) that Paul uttered, in which he asserted that the
real issue in his trial was that of resurrection; this he affirmed (with
no explicit reference to the resurrection of Jesus, though, as 25.19;
26.23 show, this was in fact vital), thereby associating himself with
one part of the Sanhedrin (the Pharisees) and setting himself against
the other (the Sadducees). True, this produced an uproar in the
Sanhedrin; but was that an αδίκημα, even though Paul had expressed
 58. PAUL AND FELIX. 24.1-27. 1111

himself more pointedly than he reports (Φαρισαῖός εἰμι, υιός

Φαρισαίων)? If to be a Pharisee was a crime, half the Sanhedrin was
guilty.
ἤ after τί = ἀλλ’ ή; cf. Xenophon, Oeconomicus 3.3.
μιᾶς ταύτης φωνής: According to BDR § 292.3, n. 5 the absence
of the article means that ταύτης is predicative. The words are
equivalent to ή φωνή ἦν μία αὕτη, the cry (that I uttered) was this
single one. But Μ. 3.193 is probably correct in taking the view that
though in earlier Greek (except in the epic poets and sometimes in
the tragedies) the attributive use of οὗτος would have required the
article, we should probably take the use of αὕτη in the present phrase
to be attributive. ‘The def. art. was being carelessly used, as time
went on, in these connections.’ Cf. Philo, De Specialibus Legibus
3.165: ἢ δι’ ἓν τούτο μόνον, ὅτι συγγενεῖς εισι.
ἐκέκραξα. This form of the aorist occurs here only in the NT (the
only other occurrence of the aorist, in Acts 7.60, has ἐκραξεν), but is
common in the LXX; see Thackeray, Grammar 1.273 (with the
reference there to other parts of the book). There are other traces of a
reduplicated form of the verb, κεκράγ(ζ)ω, though this form itself
apparently does not exist. See LS 988f., s.v. κράζω; 935, where
κέκραγμα, κεκραγμός, and κεκράκτης are listed; also BDR § 75, n.
2, ‘Neubildung nach dem als Futur umgedeuteten κεκράξομαι; vgl.
das präsentische κέκραγα.’
ύφ’ υμών ( E <a> lat syh) judged by you, is the easier reading and
certainly a ‘correction’ of ἐφ’ υμών (A B C Ψ 33 81 104 2464 pc s
syp), judged before you, that is, in your court.
‘If the Jews speak the truth they must admit that they had no case
against him except theological differences, which in the eyes of Felix
would be none at all. This was always Paul’s position’ (Begs. 4.304).
It would be prudent to add the words, ‘as represented by Luke’.

22. ἀναβάλλεσθαι, middle, to put off, to adjourn, occurs here only

in the NT. Felix like other Roman officials, has no intention of being
drawn into an internal Jewish dispute, or of doing injustice to a man
whose only offence lay in what other Jews regarded as unorthodox
theology. He makes use of standard procedure. When a case was put
off for fuller hearing in the light of new evidence or prolonged
consideration the judge would say ‘Amplius’. Thus Cicero, Brutus
22 (86), Cum consules, re audita, amplius de consilii sententia
pronunciavissent ...; cf. In Verrem 2.1.29 (74). Knowling (486):
ἀνεβάλετο αυτούς = ampliavit eos. Cf. Plutarch, Themistocles 18
(120). MM 30 quotes PTebt 1.22.9, ἀναβαλλόμενος εἰς τὸν
φυλακίτην, ‘referring the matter to the inspector’.
Felix was, according to Luke, well informed about Christianity
(the Way). No other evidence confirms this claim. ἀκριβέστερον: for
the use of the comparative cf. 17.22. It is probably elative (cf.
1112 COMMENTARY ON ACTS

Zerwick § 148, certissime sciens), but it may be a genuine compar-

ative—he had a more accurate knowledge than the Jews had, or
perhaps better, a more accurate knowledge than they thought he had.
They could not pull wool over his eyes. How did Felix acquire his
knowledge? It is sometimes said, Through Drusilla. Bruce (2.446)
properly comments, ‘But what opportunity had she of knowing about
it?’
For οδός see on 9.2. τα περί τής ὁδοῦ, the facts relating to
Christianity.
εἶπας: the participle refers to time coincident with that of the main
verb. He put them off with the words ... So Μ. 1.133 (also for
διαταξάμενος, v. 23).
καταβῇ, from Jerusalem to Caesarea; see v. 1.
διαγνώσομαι. See 23.15; cf. 25.21 (διάγνωσις).
The Old Uncial text (in P74 (A) B C Ε (Ψ 2495) 33 (81) 945
1175 1739 1891 2464 pc latt sy bo) begins abruptly with ἀνεβάλετο.
The connection is smoothed by 093vid sa, which have ἀκούσας δέ
ταῦτα ὁ Φήλιξ ἀνεβάλετο αυτούς.
In Ρ74 Ε ψ 614 (1241) 2495 pc Felix’s direct speech is introduced
by δτι.

23. διαταξάμενος. For the tense and time reference see on v. 22.
The dismissal of the Jewish accusers was doubly made in the words:
I will examine your case ...; Centurion, see that the prisoner is kept
... The various actions (ἀνεβάλετο ... εἴπας ... διαταξάμενος) are
all coincident.
On the use of the passive infinitive after a verb of commanding
BDR § 392.4 write ‘Bei Verben des befehlens steht in mehr
lateinischer als klassisch griechischer Weise der AcIPass [= Accusa-
tive and passive infinitive] (statt Inf. Akt. mit Akk. Obj.), wenn einer
Person etwas geschehen soll, ohne dass der Ausfuhrende genannt
ist.’ On the present verse they add, ‘der Centurio bewacht ihn nicht
selbst’. Delebecque (118) says that the passive is used ‘pour signifier
un ordre impersonnel, que l’on ne discute pas, donné par une autorité
officielle’. This passage is the more striking because of the combina-
tion of passive (τηρεῖσθαι) with active (ἔχειν, and, with a different
subject, κωλύειν). The mixed construction grates somewhat on the
ear, but it is in fact perfectly clear: Paul is to be kept; Paul is to have
ἄνεσις; the centurion is to prevent no one ...
τηρεῖσθαι may mean simply ‘to be kept as a prisoner’ and
understood in malam partem; he will be deprived of his liberty. But it
may (and the context suggests this as at least an additional if not an
alternative meaning) refer to protective custody. Paul will be kept
under Roman guard and not allowed to fall into the hands of the
Jews. Cf. 25.4, 21.
ἄνεσις occurs in the NT at 2 Cor. 2.13; 7.5; 8.13; 2 Thess. 1.7;
 58. PAUL AND FELIX. 24.1-27. 1113

relief, in this case lightening of the circumstances of his prison

régime. It would be custodia libera.
οἱ ἴδιοι, his own people, would normally suggest his relatives (and
these may be included here—see 23.16) and friends (ND 3.148
quotes POxy 46.3314.15, true friends); here, fellow Christians. In
antiquity, and indeed until modern times, conditions in prison would
be made tolerable by the access of friends who brought provisions
and comfort. Cf. Lucian, Peregrinus 12; Josephus, Ant. 18.203f.,
235. This corresponds to the fact that for Romans (such as Felix; not
for Greeks) imprisonment was not a punishment but a means of
keeping people available for trial or for actual punishment. Digest
48.19.8,9 (Ulpian): carcer enim ad continendos homines, non ad
puniendos haberi debet.
ύπηρετεῖν will here refer to the services that a man unable to
provide for himself would need, especially no doubt food.

24. παραγενόμενος, Felix came—where? It is possible (though

Luke does not say so) that he had been absent and returned to
Caesarea. Otherwise he came to wherever Paul was. Marshall (381)
thinks of the prison. Schneider (2.352) with perhaps greater probabil-
ity says that παραγεν. ‘bezieht sich auf den betreffenden Raum im
Prätorium (23.35), in den Felix den Gefangenen rufen liess’.
σὺν Δρουσίλλη. See on 23.24. This was not Drusilla the grand-
daughter of Cleopatra and Antony (Tacitus, Histories 5.9) but
Drusilla the sister of Herod Agrippa II (Josephus, Ant. 20.141-144);
this is made clear by the addition of οὔση Ίουδαίᾳ. Not that the
Herods were fully Jewish, though they did maintain at least a façade
of Jewishness. Most editors accept the reading of B C2 33 36 81 1175
pc, τή ιδία γυναικί, his own wife. Whose wife might be expected to
accompany him? It is true that this Drusilla had previously been the
wife of Azizus, but Luke shows no interest in this story. C* 093 <a>
have τή γυναικί; P74 *2 Ε Ψ 945 1739 1891 2464 al have τή
γυναικί αύτού; A al have τή ιδία γυναικὶ αὐτοῦ, which is
manifestly conflate. Μ. 1.88, 90, observing that in the NT ίδιος
though on the way to exhaustion has not yet reached that point, is
inclined ‘for once to prefer to B. A modern textual critic would
not write ‘for once’, with its implication that B is very nearly
inerrant. G. Μ. Lee (NovT 9 (1967), 41) defends the reading of B on
unconvincing grounds; in his quotation from Babrius (Fable
45.12-14) there were excellent metrical reasons why the author
should not have written the αὐτῶν which the sense seems to require,
but Luke was writing prose, and if he meant ὁ Φήλιξ αὐτὸς σὺν
Δρουσίλλη τή γυναικί there was no reason at all why he should not
have written it. The reading of B is difficult enough to be accepted as
the origin of the variants, and not so difficult as to be impossible.
There is no doubt that the force of ίδιος was weakening, and we may
1114 COMMENTARY ON ACTS

conclude that the reading of B is correct, that the present use shows a
stage of weakening more advanced than is common in the NT, and
that the words should be rendered his wife.
After Ίουδαίᾳ, syhmg (boms), possibly representing the Western
text, or at least one branch of it, add: quae petebat ut videret Paulum
et audiret verbum; volens igitur satisfacere ei. Cf. the variant in v. 27.
Both are probably secondary ; the editor may have wished (a) simply
to ‘improve’ the story; (b) to supply a reason for v. 24b; (c) to build
up a comparison between Drusilla and Herodias; (d) to justify the
mention of Drusilla. Clark (155, cf. 381) reconstructs the Western
Greek text as: ἥτις ἠρώτησεν ἰδειν τόν Παύλον καί ἀκούσαι τόν
λόγον. Θέλων οὖν χαρίζεσθαι αυτῇ ... Calvin (2.255) dismisses
Felix’s apparent interest in Paul; he hears him only on account of
Drusilla, and she is merely curious.
Luke knows that if Paul is to be heard faith will be one of his
themes; cf. 20.21. But if v. 25 is to be regarded as summarizing the
content of Paul’s discourse on faith it must be said that faith is here
understood in a sense different from that of the epistles. See on v.
25.
Ίησούν is omitted by A Cvid Η P 614 1241 pm syp sams. It is hard
to see any reason why the name should have been omitted, easy to
see why it might have been added. There is much to be said for the
short text.
There seems to be no good reason why Felix could not at this point
have dismissed the case; cf. Lucian, Peregrinus 14.
Pesch (2.260) observes that vv. 26, 27 join up well with v. 23; this
suggests that vv. 24, 25 are a Lucan insertion. Weiser (632) thinks
that vv. 24-26 were formed by Luke.

25. Paul is of course obliged to obey Felix’s summons, no doubt

willing to discuss and dispute (διαλέγεσθαι) with him as with others
(e.g. 17.17).
Paul’s themes (surprisingly under the heading of faith—see v. 24,
also below) are δικαιοσύνη, εγκράτεια, and κρίμα. δικ. is undoubt-
edly a major theme is the epistles, but the word seems to be used in a
different sense here. See 10.35 and 17.31, which suggest the double
meaning of righteous behaviour in men and righteous judgement in
God, who will duly reward those who practise righteousness. The
only other passage in Acts in which the word is used is 13.10.
εγκράτεια (see W. Grundmann in TWNT 2.338-40) occurs at Gal.
5.23 as part of the fruit of the Spirit, elsewhere in the NT only at 2
Pet. 1.6 (but cf. 1 Cor. 7.9; 9.25, ἐγκρατεύεσθαι; Titus 1.8,
εγκρατής). In all these passages ethical behaviour is in mind. In itself
the word means mastery over something or someone, but its sense
seems to have been determined by Plato’s εγκράτεια ἐαυτοῦ
(Republic 390b) and similar uses. It means mastery over oneself,
 58. PAUL AND FELIX. 24.1- 27 1115

over one’s pleasures and desires (Republic 430c); thus discipline,

especially self-discipline. It is not for man to estimate his own
success in righteousness and self-discipline, hence τό κρίμα τὸ
μέλλον, the judgement to come in which God will make his own
estimate, κρίμα occurs nowhere else in Acts, κρίσις only at 8.33,
which is clearly not relevant. Of 22 occurrences of the verb κρίνειν,
20 refer to human judgements of one kind or another. The only
relevant parallel, but that a close one, is 17.31 (note here also the use
of δικαιοσύνη). There is some kinship between these passages in
Acts and 1 Thess. 1.9f., where the coming wrath from which Jesus
delivers men is certainly the wrath that is the negative aspect of
judgement, but neither in Athens nor before Felix does Paul get as far
as the deliverance. For these themes cf. Acts of Paul and Thecla 5;
Acta Johannis 84; Actus Petri cum Simone 33-35 (L.-B. 1.238;
2.1.192; 1.85-87); also Aristeas 278.
<a> is no doubt mistaken (textually, not in sense) in adding ἔσεσθαι
after μέλλοντος; C 36 1175 pc have μέλλοντος κρίματος.
Felix is frightened (έμφοβος, clearly passive in sense, as in
Theophrastus, Characters 25(26).1; earlier active, e.g. Sophocles,
Oedipus at Colonus 39), as well he might be, by the thought of
judgement. But the words of his response are ambiguous; cf. 17.32.
Does he mean that he intends to converse further with Paul, or is he
making a polite gesture? V. 26b suggests the former.
In τὸ νῦν ἔχον the last word ‘is not very easy to explain’ (Moule,
IB 160). Moule compares Tobit 7.11 (B A), τό νυν ἔχων ἡδέως
γίνου, but this is somewhat easier. Begs. 4.305 describes the phrase
as ‘entirely idiomatic in Hellenistic prose’, and quotations in Wett-
stein (2.624; e.g. Lucian, Cataplus 13: τὸ δὲ νυν ἔχον, μή διάτριβε)
confirm this. Perhaps ἔχον should be taken as an absolute participle
and τό νυν as corresponding to an adverb, as in the familiar
construction ούτως ἔχειν, to be thus. ‘It being now.’
καιρόν μεταλαβών, when I get an opportunity, similarly with the
variants παραλαβών (A), λαβών (093 81), and εὐρών (P74). καιρῷ
δὲ ἐπιτηδείω on a suitable, or convenient, occasion (E latt) is a more
commonplace expression (though καιρ. μετ. also is described in
Begs, as entirely idiomatic and probably a secondary reading).
On faith (v. 24) in this context: it is possible to argue that faith is
related to salvation (cf. 16.31 ) and thus to the forgiveness of sins. It is
in the light of judgement that this becomes a serious matter.
‘Entsprechend wird in 24.24f. deutlich, dass mit der Verkündigung
des “Glaubens an Christus Jesus’’ vor Heiden die Predigt vom
“künftigen Gericht’’ untrennbar verbunden ist, auch vor einem
römischen Prokurator’ (Delling, Kreuzestod 91).

26. The ambiguity (see ν. 25) in the description of Felix’s motives

reappears in this verse, the first half of which suggests nothing but a
1116 COMMENTARY ON ACTS

corrupt and mercenary interest whereas the second half is open to a

more favourable interpretation. Men do have mixed motives, and
there is no reason to suppose that Felix could not have kept Paul
available to himself for a mixture of good and bad reasons; it seems
however unlikely that Luke would have at his disposal information
about the governor’s inward thoughts; his account will reflect at best
the different things that people said about Felix, at worst his own
imaginative reconstruction. It is probably true that he did wish to
present Felix in a double light: as the ‘good’ Roman, who recognizes
Paul’s innocence, and as the ‘bad’ Roman who does not release him
because he hopes for a bribe. But Haenchen’s conclusion (634) is not
necessarily true: ‘Der Mann, der Paulus besucht, um von Jesus zu
hören, und der Mann, der Paulus besucht, um Geld zu sehen, ist nicht
ein und derselbe.’ They might be; not every man is a logically and
morally consistent whole.
ἅμα, at the same time, that is, at the same time that he was
promising to send for Paul and hear what he had to say he was
nourishing the thought that the prisoner, who evidently had friends
who were interested in his fate (v. 23), might prove a useful source of
revenue. ‘En bon grec, ἅμα καί, et aussi ἅμα δέ, indique le vrai
motif, déguisé sous un prétexte’ (Delebecque 119). αὐτῷ is omitted
by B p* vgst; after Παύλου, <a> co add ὅπως λύση αὐτόν, ‘unwilling
to leave anything to the reader’s imagination’ (Metzger 492).
The intention ascribed to Felix was of course illegal, specifically
proscribed by the Lex Julia de repetundis of 59 BC, contained in
Digest 48.11 (... praecipit ... ne ... quis ob hominem in vincula
publica conjiciendum, vinciendum, vincirive jubendum, exve vincu-
lis dimittendum ... aliquid acceperit). The practice was illegal, but it
nevertheless existed. For the illegalities of another procurator,
Albinus, see Josephus, War 2.273; Ant. 20.215.
πυκνότερον can hardly be a true comparative, since there is
nothing available for comparison; it will be elative, pretty often. See
Μ. 3.30; Turner (Insights 90). Cf. v. 22.
ὁμιλεῖν with a dative can mean to speak to; this is not impossible
here, but much more probably is either the general sense to consort
with, to hold converse with, or perhaps even a sense nearer that of
Xenophon, Memorabilia 1.2.15, 39, where LS’s (1222) ‘frequent a
person’s lectures’, ‘be his pupil’ are too strong for Acts but on the
right lines. The variant διελέγετο (C 36 453 pc) gives the required
meaning.
‘Sic thesaurum evangelii omisit infelix Felix’ (Bengel 478).

27. διάδοχος is usually (LS 393) an adjective, and it should

probably be taken here as predicative: Felix received Porcius Festus
in succession to himself, as his successor. Little is known of Porcius
Festus (it is not logical to assume that this means that he was
 58. PAUL AND FELIX. 24.1-27. 1117

governor for only a short time). His work as governor is summed up

by Josephus in War 2.271: ‘Festus, who succeeded Felix as procura-
tor, proceeded to attack the principal plague of the country; he
captured large numbers of the brigands (ληστών) and put not a few to
death.’ A few more details are given in Ant. 20.182-200. The λησταί
were called σικάριοι from the weapons, like the Roman sicae, that
they used. He was reasonably tolerant to the Jews over the wall they
built in the Temple with a view to preventing Agrippa from
observing the sacrifices. Festus seems to have died in office; he was
succeeded by Albinus.
The dates of Festus’s governorship are given by the Armenian
version of Eusebius’s Chronicle as 54-60; according to Jerome,
whose version of the Chronicle is on the whole more trustworthy,
they are 56-60. Between Festus and Albinus there was an inter-
regnum during which total confusion reigned. It was at this time that
James, the brother of Jesus, was put to death (Josephus, Ant. 20.200;
cf. Hegesippus in Eusebius, HE 2.23.4-18). The martyrdom is
usually dated in 62 (rejecting Hegesippus, who seems to think of a
date immediately before the outbreak of war in 66). 60-62 makes a
long interregnum, but 62 is confirmed by Josephus, War 6.300, 305
(Albinus was in office four years before the war) and 308f. (seven
years and five months before the last stages of the siege). More
important is the question when Festus succeeded Felix. The fullest
discussions are to be found in Begs. 5.464-7 and in NS 1.465f. (note
42), but these come to different conclusions. Begs, lays stress on the
fact that after his recall Felix was accused by the Jews, and ‘he would
undoubtedly have paid the penalty for his misdeeds against the Jews
had not Nero yielded to the earnest entreaty of Felix’s brother Pallas,
whom at that time he held in the highest honour (μάλιστα δὴ τότε
διά τιμής άγων εκείνον)’ (Josephus, Ant. 20.182). But Pallas fell
from favour early in Nero’s principate, so that 55 or 56 would be the
latest possible date for Felix’s recall. Moreover (see above) Euse-
bius’s Chronicle gives 54 or 56 for the accession of Festus. Thus K.
Lake in Begs. NS, where the date is 60 is preferred, replies by casting
doubt on the trustworthiness of Eusebius’ Chronicle (and on our
knowledge of it), and by pointing out that Pallas had fallen from
power at a very early date (by 13 February 55—the 14th birthday of
Britannieus—at latest), so that it must be inferred ‘that Pallas in spite
of his dismissal remained influential—a conclusion which agrees
completely with Tacitus’ statement’ (466). The statement in question
occurs in Annals 13.14: Nero ... demovet Pallantem cura rerum quis
a Claudio impositus velut arbitrum regni agebat; ferebaturque degre-
diente eo magna prosequentium multitudine, non absurde dixisse ire
Pallantem ut eiuraret. For the eventual death of Pallas see Tacitus,
Annals 14.65. Nero apparently grew tired of waiting for a share of his
great wealth. NS does not observe that this qualification of the
1118 COMMENTARY ON ACTS

position of Pallas to some extent spoils the case against Begs.

Pallas’s influence is much likelier to have lasted till summer 55, or
even 56, than till 60. This is not a matter on which certainty can be
attained (unless fresh evidence appears), and to some extent a
decision will turn upon the meaning of the next words (διετίας δέ
πληρωθείσης) to be considered. The early date has encountered
much criticism but there is still a good deal to be said for it. See in
addition to the two works mentioned, E. Schwartz (NGG 1907,
263-99, especially 284-7); J. Jeremias (ZNW 27 (1928), 98-103);
Haenchen (80-84); R. Jewett (Dating Paul's Life (1979), 41-4, and
the literature there cited); Hemer (173: ‘The view of Lake ... is
simply unacceptable’).
διετίας δὲ πληρωθείσης, when a two-year interval was completed.
It is very often supposed that this period refers to Paul’s imprison-
ment in Caesarea; this is assumed, for example, without discussion in
NS (loc. cit.). Haenchen (632) makes the valid point that Luke ‘nur
für die Gefangenschaft des Paulus Interesse hat’; cf. Schneider
(2.353). On this view, and if Festus was appointed in 55 or even 56, it
is impossible to harmonize this date with those arrived at above (p.
871; see also Introduction, pp. lvif.) for Paul’s residence in and
departure from Corinth: the activities of Acts 18-21 could not be
brought to a close by 53 or 54. This would constitute a strong
argument in favour of a later date for the supersession of Felix by
Festus (though not a date as late as 60). But διετίας πληρωθείσης is a
genitive absolute and its connection with the rest of the sentence is an
open question. Grammatically it is at least as likely that it refers to
Felix’s governorship as that it refers to Paul’s imprisonment, and
though it is correct, as Haenchen observes (see above) that Luke’s
overall interest is in Paul, at this point he is speaking of the
procuratorial succession. And this way of taking the clause suits the
chronology admirably (see Romans 4f.). The case is further strength-
ened by the fact that διετία represents the Latin biennium (BDR
§ 5.3; cf. Acts 28.30) and that biennium is sometimes (by no means
always) used for a two-year period of office (e.g. Cicero, In Verrem
2.3.93 (216), Biennium provinciam obtinuit; Suetonius, Galba 7,
Africain proconsul biennio obtinuit. So H. G. Kippenberg, Kleine
Pauly 4.1059.
The change of ruler might have seemed a suitable time for Paul’s
release, but Felix left him in prison (δεδεμένον). According to the
majority of MSS Felix’s motive was to curry favour with the Jews.
χάριτα κατατίθεσθαι is a not uncommon phrase, ‘to lay up a store of
gratitude or favour’ (LS 917, κατατ. II 4); or simply in Hellenistic
use ‘do a favour’ (Begs. 4.306). Page (241): καταθέσθαι ‘is strictly
used of “depositing with a banker’’; its metaphorical use is clas-
sical’; he cites Thucydides 1.33.1; 1.128.3. Delebecque (119) adds
Herodotus 6.41.3; Plato, Cratylus 391b; Demosthenes 19.240 (416);
 58. PAUL AND FELIX. 24.1-27. 1119

59.21 (1351); Xenophon, Cyropaedia 8.3.26. Cf. 25.9, where Festus

does the same. The difference between the two passages is that in
25.9 the accusative of χάρις is χάριν (as almost always in the NT ),
whereas here it is χάριτα (so A B C 33 104 1175 pc; χάριν in
E L 323 945 1739 1891 2495 al; χάριτας, <a>). See Μ. 2.132. In the
idiom χαρ. κατατ., χάριτα often appears and this usage was
probably in Luke’s mind (or possibly in a source). The Western text,
represented by 614 2147 syhmg, ascribes to Felix a different reason for
leaving Paul in prison: τòν δέ Παύλον εἴασεν ἐν τηρήσει διά
Δρούσιλλαν. The probable origin of the reading is (a) the desire to
find a reason for the reference to Drusilla in v. 24, and (b) the thought
that there was a parallel between John the Baptist’s suffering at the
hands of Herodias and Paul’s at the hands of Drusilla.
For κατέλιπε, P74 A L 81 104 453 1175 2464 pc have κατέλειπε.
For divergent views on this verse see the good account in Weiser
(635).
 59. PAUL APPEALS TO CAESAR 25.1-12

(1) So Festus arrived in the province and after three days went up to
Jerusalem from Caesarea. (2) The chief priests and the leading men of the
Jews informed him against Paul and begged him, (3) asking a favour against
[Paul],1 that he would send him to Jerusalem, for they were making an
ambush to kill him on the way. (4) Festus answered that Paul was being kept
at Caesarea, but that he himself would shortly be2 going [there]. (5) ‘So’, he
said, ‘let the eminent men among you come down with me and, if there is
anything wrong in the man, let them accuse him.’
(6) He spent not more than eight or ten days among them, went down to
Caesarea, and on the next day sat down in the place of judgement and
commanded Paul to be brought. (7) When Paul arrived the Jews who had
come down from Jerusalem surrounded him bringing many weighty charges,
which they were not able to prove. (8) Paul in his defence said, ‘Neither
against the Law of the Jews, nor against Caesar, have I done any wrong.’ (9)
Festus, wishing to curry favour with the Jews, said in response to Paul, ‘Are
you willing to go up to Jerusalem and be tried concerning these matters there
before me?’ (10) Paul said, 'I am standing in Caesar’s court, where it is right
that I should be tried. I have done no wrong to the Jews, as you well know.
(11) If then I am a wrongdoer and have done something worthy of death I am
not refusing to die, but if there is nothing in the things of which they accuse
me no one can make a present of me to them. I appeal to Caesar.’ (12) Then
Festus spoke with his council and replied, ‘You have appealed to Caesar, to
Caesar you shall go.’

Bibliography
G. D. Kilpatrick, EThL 53 (1977), 107-12.
S. Legasse, as in (54).
H. Sahlin, NovT 24 (1982), 188.

Commentary
A new governor has taken office in Judaea. He is a man of energy.
Three days are enough for him to settle in Caesarea before setting out
for his second capital city, Jerusalem, where not more than eight or
ten days are needed for a preliminary survey, which includes a
request from the Jews to bring Paul to Jerusalem. The old plot of
assassination (23.12) is revived: Paul will be murdered on the way.
Whether because he suspects the plot or on general principles Festus
1Greek, him.
2There is not in the Greek; NEB, leaving Jerusalem; NJB, going back there.
1120
 59. PAUL APPEALS TO CAESAR. 25.1-12. 1121

refuses. He is going to Caesarea; let the Jews do so too so that there

may be a hearing. There is a hearing; and it is inconclusive. Indeed, if
Luke’s account is followed it is incomplete; witnesses are not called.
The Jews present their accusations; Paul affirms his innocence.
Instead of reaching a decision Festus asks Paul if he is willing to go
to Jerusalem and be tried there. Why does Festus not complete the
legal business in Caesarea? The answer given is that he wishes to do
the Jews a favour, but this desire is qualified by the question to Paul,
which is meaningless if Paul is not free to answer, No, I am not
willing to go to Jerusalem. In fact he does so answer, and in doing so
demands more than the resumed trial in Caesarea that Festus by
implication offers him. Paul had as much to fear from acquittal as
from condemnation. Released from custody in Caesarea with no
Roman soldiery at hand to protect him he would have been an easy
victim to the assassin’s knife. By appealing to the Emperor he took
the case not merely out of Jerusalem but out of Judea and assured
himself (so far as there could be any assurance) of a safe passage to
Rome.
The historicity of this narrative has often been assailed. Objections
are well summed up by Haenchen (641): ‘Bei Lukas bleibt unver-
ständlich: 1. warum nach Abschluss der Verhandlungen kein Urteil
erfolgt, sondern eine Verlegung des Prozesses ins Auge verfasst
wird, 2. warum Paulus nicht einfach auf Fortführung des Prozesses in
Cäsarea besteht, sondern an den Kaiser appelliert, 3. warum Festus
einen des crimen laesae maiestatis verdächtigen Mann nicht selbst
richten (oder nach Rom senden) will.’
Haenchen’s first point may be answered by the observation that, in
Luke’s narrative, the proceedings had not reached an Abschluss.
Each party had made an introductory statement. The second has been
answered in the observation that any verdict reached in Caesarea
would have been highly dangerous to Paul; to transfer the case to
Rome was for him an excellent move. The third is hypothetical: that
Paul was accused of causing seditio throughout the Empire and thus
of the crimen laesae maiestatis was true, but it was not unreasonable
on Festus’ part to wish to form his own judgement on the substance
of the charge brought by the Jews. The integrity of Festus has also
been called in question. Why should he imperil a man (still presumed
innocent) by sending him from Caesarea to a prejudiced trial in
Jerusalem? To some extent the force of this complaint rests on the
meaning of ἐπ’ ἐμοῦ (ν. 9; see the note), but in any case the attitude
of Festus, as an honourable man, is, given his position, intelligible.
He is a new governor with an unusually intractable native population
to deal with; of course he would like to do something that would win
favour with the Jews. It is to his credit that he is aware that justice
demands that he should also bear in mind the rights of his prisoner;
hence, ‘Are you willing ...?’
1122 COMMENTARY ON ACTS

The story as Luke tells it hangs together. This does not prove that
it is historically true. It might be intelligent fiction. Various views on
this question have been held. It is certainly true that the story
achieves at least one end that Luke would hope to achieve. ‘An dem
Weltmissionar Paulus hat der römische Staat als an dem hervorra-
genden Exempel ein für allemal erkannt und demontriert, dass kein
Grund vorliegt, an der politischen Loyalität der Christen zu zweifeln’
(Schmithals 217; cf. the less positive estimate in Schille 454).
Lüdemann describes vv. 1-5 as a doublet of ch. 24, and vv. 6-12 as
even more clearly a doublet (261). Pesch (2.264) on the other hand
argues that Luke drew the main features of this paragraph from his
Quellenbericht; the tensions in the paragraph are due to his desire to
present the Romans in a good light.
Weiser (638) gives ten reasons for the belief that the present
paragraph ‘sowie die Komposition der Szenen und Reden Kap
25-26 stammen ... erst von Lukas’. Most of these consist of
parallels between the ‘trial’ of Paul and that of Jesus in Lk. 23. They
are as follows. (1) The hearing before the Roman Governor is
followed by one before the Jewish king. (2) The connection between
the two scenes recalls the use of Ps. 2.1f. in Acts 4.27. (3) The
combination fulfils the prediction made to Paul in 9.15f. (4) In both
hearings leading Jews bring accusations but no condemnation fol-
lows. (5) The planned assassination of v. 3 repeats that of 23.12-15.
(6) Acts 25.7 repeats 24.5f. as Lk. 23.10 repeats 23.2. (7) Acts 25.7
recalls Lk. 23.14 in that neither accusation is effective. (8) Acts
25.13-23, explaining why Paul appears before Agrippa II, recalls
Lk. 23.6f., 11. (9) Paul, after being found innocent before Festus and
Agrippa, remains in custody (26.30, 32); so Jesus, though found
innocent by Pilate and Herod (Lk. 23.24f.), is not released. (10)
Paul’s speech in Acts 26 is made up from elements previously used
and contained in the narrative.
The last point is not relevant to discussion of the present paragraph
(see below, pp. 1144ff.). The points of 1, 4, 6, 7, 8, 9 do bring out a
measure of parallelism between the story of Jesus and the story of
Paul; it is of course not to be denied that Luke is responsible for both
stories in the form in which we read them. He does nothing however
to indicate awareness of the parallelism, as he probably would have
done had it been important to him, and does not repeat his earlier
reference to Ps. 2.1f. It was of central importance to Luke (and he
constantly emphasizes the belief) that both Jesus and Paul were
innocent of the charges brought against them, and of nearly equal
importance that both appeared before both Jews and Romans, of
whom the latter were more favourable to them. Luke seems to have
had a special interest in, perhaps a special knowledge of, the Herods
(see on 13.1). It is not surprising that he should have written up the
story of Paul’s trials in the way that he does, and a number of details
 59. PAUL APPEALS TO CAESAR. 25.1-12. 1123

may well be his own contribution; this however does not affect the
main outline of events. The essential question we have to ask is
whether Paul was a Roman citizen (see on 16.37; 22.25) and used his
citizenship as a means of having his case transferred to Rome. Given
citizenship, the appeal makes very good sense, and there is little to be
said for the argument of Schmithals (219) that Paul travelled to
Rome as a free man. The We-narrative returns in ch. 27 with the
journey to Rome, and Philippians is best understood (but we must
beware of arguing in a circle) as representing a Roman imprisonment
consistent with the position of a prisoner on appeal.

1. Little is known of Porcius Festus. For the date of his accession

see on 24.27. Josephus records that he had to deal with the sicarii
(Ant. 20.185-187) and with an impostor who led followers into the
wilderness (188). He allowed the Jews to send an embassy to Nero
about a wall built in the Temple (193,194); he died in office, and was
succeeded by Albinus (197). The precise sense of ἐπιβάς will be
determined by the meaning of ἐπαρχεία and the variant ἐπαρχείω
(both of which have orthographical variants in which ει is replaced
by ι). ἐπαρχεία ( B C Ε Ψ <a>) will refer to the province, so that
ἐπιβάς τῇ ἐπαρχεία will mean, Having arrived in the province (cf.
Ant. 20.185, ἀφικομένου δέ εις τήν Ίουδαίαν Φήστου ...).
ἐπαρχείω will according to Μ. 2.157 be short for τή ἐπαρχείω (the
adjective is of two terminations) ἐξουσία, and the clause will mean,
Having entered upon his (provincial) office. This however may be
too simple, for it may be right to supply not ἐξουσία but χώρα, so
that the meaning would again be, Having arrived in the province.
Commentators and others differ in their preference, but Begs. 4.306,
after a very full note on έπαρχεία—έπαρχείω, rightly concludes,
‘There is probably no difference in the meaning of these two
phrases.’ Festus would enter upon his office as governor of the
province when he entered the province. Cf. 23.34. Page (241) points
out, ‘Strictly Judaea was not a “province”, but a department of the
province of Syria, but the term is used loosely.’ ND 2.85 draws
attention to a narrower meaning (district) in PMich 659.45, but this
belongs to the 6th century AD. Festus may have landed at Caesarea
itself, or at Seleucia, completing the journey by land. The procurator
usually resided at Caesarea and went up to Jerusalem when occasion
required. It would be natural to do so at the beginning of one’s term
of office and Festus (according to Luke) permitted no delay (after
three days) in going up to his second capital, ἀνέβη is the pilgrimage
word (see 11.2), but here suggests nothing more than actual ascent.
It is not clear what logical force οὖν can have. This is no doubt
why, in P74 pc, it is replaced by δέ.

2. ἐνεφάνισαν: for this word see on 23.15. ND 2.104 cites a

1124 COMMENTARY ON ACTS

parallel from an inscription in Beroea. The construction here is

identical with that in 24.1 (with dative, and κατά with genitive).
οἱ αρχιερείς is hardly to be expected and the variant ὁ ἀρχιερεύς
(H P 049 189 326 pm) is not surprising. But the plural occasions no
difficulty; see e.g. 4.23; 5.24 and many passages in the gospels. The
High Priest at this time was Ishmael b. Phiabi (Josephus, Ant.
20.179).
οἱ πρῶτοι τῶν ’Ιουδαίων occurs again at 28.17; cf. 13.50; 28.7. It
recalls the use of for leaders, but is intelligible in itself. It may
be right to compare the δυνατοί in v. 5; see on that verse.
In what follows Luke does not use his pronouns with perfect
clarity (αὐτῶ ... αὐτόν ... αὐτοῦ ... αὐτόν), but there is little doubt
what he means.
The Jewish leaders have a request to make (παρεκάλουν).
3. The request (v. 2) is specified, αἰτούμενοι χάριν κατ’ αὐτού
χάρις is here used in the concrete sense of a favour, somewhat
unusually described not as done on behalf of someone (though this is
of course a secondary implication) but as done against Paul.
Behind the favour asked for is the same kind of plot as had failed
earlier (23.15f.); at 23.21 the verb ἐνεδρεύειν is used. Festus is to
send for Paul to Jerusalem, where the Jewish leaders now are; the
Jews will make an ambush and kill Paul on the road between
Caesarea and Jerusalem, μεταπέμπεσθαι is Lucan: 10.5, 22, 29;
11.13; 20.1; 24.24, 26—and nowhere else in the NT. ἐνέδραν
ποιοῦντες: the active is used, not the middle, because the Jewish
leaders will cause the ambush to happen but will not themselves take
part in it (Zerwick § 227: ‘... “insidias paraturi”, non necessarie
“ipsi insidiaturi” ’). Contrast Thucydides, 3.90.3, ... δύο φυλαί...
ἐνέδραν πεποιημέναι. κατά τήν οδόν also is Lucan. ‘L’expression
est propre à Luc: Lc 10.4; Act 8.36 et 26.13’ (Delebecque 120).
There is a variant in this verse not recorded in NA26. Ropes (Begs.
3.229): ‘The Greek translated in the gloss of hcl.mg may have run
somewhat as follows: οι ευχήν ποιησαμενοι oπως επιτυχωσι του
γενεσθαι αυτoν εν ταις χερσιν αυτων. But the paraphrase probably
involved other changes, no longer recoverable, from the B-text, and
the Syriac is perhaps not a perfectly literal rendering. No other trace
of the gloss is known. The paraphrast seems to have overlooked the
lapse of two years since 24.12.’ Or perhaps he did not understand the
two years of 24.27 to refer to Paul’s imprisonment!
4. This is not Luke’s characteristic μὲν συν; it does not begin a
new incident, and there is an answering δέ. So rightly Begs. 4.307;
Haenchen (636) agrees with Begs, that μέν should stand before
Παῦλον: ... τòν μὲν Παῦλον ... εαυτόν δέ ... Festus is not to be
hoodwinked but behaves with a proper responsibility towards his
prisoner. He is another of Luke’s good Roman officials.
 59. PAUL APPEALS TO CAESAR. 25.1-12. 1125

τηρεῖσθαι, Paul was (being) kept, and would for the present still
be kept.
εἰς Καισαρείαν: ‘L’acusatif avec εις implique la venue à Césarée:
Paul y a été conduit pour y être gardé’ (Delebecque 120). But it is
more likely that here as frequently εις is used for έν.
ἑαυτόν δέ: αυτός might have been expected, since the pronoun
refers to the subject of the verb ἀπεκρίθη; ‘att. potius αυτός ... v.
tarnen Kühner Gr. II2 595sq.’ (Blass 258). Μ. 3.147: ‘Luke may have
deviated from class. usage into the accus. because he wished to
coordinate the new subject with Παῦλον.’ According to BDR §
406.1, n. 1 what Luke does is in fact classical when a contrast is
intended, though αυτός also would be possible.
If the Jews wish to make a case against Paul they must go to
Caesarea, where he is; Festus himself is setting out for Caesarea
shortly. Festus ‘will über Paulus nicht ohne ordentliche Gerichtsver-
fahren entscheiden’ (Schmithals, 217). See also on vv. 5, 16.
For ἐν τάχει see 12.7; 22.18; cf. Lk. 18.8—a Lucan expression.
5. μητί marks a change from indirect to direct speech; cf. 1.4.
οἱ ἐν ὑμῖν δυνατοί may be taken to be equivalent to the αρχιερείς
and πρώτοι of v. 2. δυνατοί is not uncommon with the meaning
eminent men (cf. 1 Cor. 1.26). Thus Thucydides 1.89.3, οἰ δυνατοί
τῶν Περσῶν; Plutarch, Romulus 27 (34), τούς ἐν 'Ρώμη δυνατούς;
Philo, Moses 1.49; Josephus, War 1.242, Ιουδαίων οἱ δυνατοί. Cf.
however Bengel (479): qui valent, ad iter faciendum; also Blass
(258), qui possunt. But the former meaning is more probable; so
Page (242); Schille (441), ‘δυνατός ist ein politischer Terminus
technicus.’ Festus does not intend that the courtroom shall be
cluttered with a large number of Jews, and perhaps suspects a violent
attempt on Paul’s life. For the same reason, probably, he wishes the
Jews to accompany him (συγκαταβάντες). Cf. ἀνέβη, v. 1.
The Jews are to accuse Paul εἴ τί ἐστιν ἐν τω ἀνδρί ἄτοπον. This
is straightforward; ἄτοπον is a mild word for crime used either
because Festus is showing proper legal caution or because Luke
wishes to show Paul in the best possible light. Cf. 28.6. In the text
however ἄτοπον is read by A B C E 33 81 945 1175 (1739 1891)
2464 al lat; in place of this τούτω is read by <a>; τούτῳ ἄτοπον by Ψ
36 453 614 2495 al syh sa? bo. The majority of (late) MSS thus have
an expression εἴ τί ἐστιν ἐν τῷ ἀνδρὶ τούτῳ, to be rendered literally,
If there is anything in this man, and evidently meaning, If there is
anything evil, criminal, in this man. This does not apper to be a
regular Greek idiom, but cf. Jn 14.30, ἐν ἐμοί οὐκ ἔχει ούδέν.
Cf. Schmithals (217) quoted on v. 4.
6. Festus did not stay long among the Jews (ἐν αὐτοῖς, omitted by
P74) in Jerusalem—in days, ού πλείους οκτώ ή δέκα. So (Ρ74 B ) A
C 33 36 81 453 945 1175 1739 1891 pc latt bo, probably rightly. But
1126 COMMENTARY ON ACTS

Ψ <a> have πλεἱους οκτώ ή δέκα; 2147 pc have οκτώ ή δέκα. There
can be little doubt that the text of the old uncials is correct; it gives a
natural kind of approximation.
Festus returned to Caesarea, and on the day after his arrival took
the matter in hand, ordering Paul to be brought (for the passive
infinitive, ἀχθῆναι, after a verb of commanding, see 24.23) to the
courthouse. He sat ἐπί τοῦ βήματος. For the word βήμα see on 18.12
(with the reference to Dinkier); see also vv. 10, 17 in this chapter.
‘Da eben das Sitzen des Richters zum amtlichen Charakter des
Rechtsprechens gehört, so wird mit dem Ausdruck καθίζειν ἐπὶ
βήματος nicht nur summarisch das Niederlassen auf dem Richter-
stuhl, sondern zugleich der Beginn der Amtshandlung ausgedrückt’
(Dinkler, Signum Crucis 122). Cf. the Latin sedens pro tribunali.
τῇ ἐπαύριον. ‘Les indications chronologiques sont ici très pré-
cises, plus cependant quand il s’agit de Césarée que de Jérusalem’
(Delebecque 120).
7. πολλά καί βαρέα. The καί is classical (BDR § 442.7a), and
probably bears witness to an attempt to write in literary style. Blass
(258) could write, ‘αἰτίωμα pro αἰτίαμα nusquam exstat’. nusquam
is no longer strictly correct; αἰτίωμα occurs in PFay 111.8. This does
not alter the fact that Luke, having just used the classical καί now
abandons classical use completely. Whether these many weighty
charges included more than had already been brought by the Jews on
earlier occasions Luke does not tell us; he is confident that they were
incapable of proof.
ἴσχυον] ἴσχυσαν P74 *. The aorist is constative: all they said did
not amount to proof. The imperfect suggests continuous but unsuc-
cessful attempts to prove. One might translate, They could not prove,
however hard they tried. ND 4.86 has an inscription to illustrate the
use of ἀποδεῖξαι with the meaning to prove.
8. Paul’s defence is no more explicit than the Jewish accusations.
Luke no doubt feels that he has by now presented the legal material
in sufficient detail, and moves rapidly on to the great dramatic
dénouement in v. 11. ἀπολογεῖσθαι is the technical term for
delivering the defence speech; cf. 22.1.
The three points touched on, but not discussed, do represent the
major issues, or rather the major fields in which the issues lay. As
regards the Law, Luke maintains that Paul’s interpretation, the
Christian interpretation, is in fact correct. He has not abandoned the
Law, he has not counselled his fellow Jews to abandon the Law; he
has revised the conditions for the admission of Gentiles. The Temple
offence has been dealt with, both by narrative (which showed Paul
engaging in Temple worship in a proper way) and by exposing the
slender basis of the charge: Jews of Asia—now unavailable as
witnesses—had seen Paul with Gentiles in the city and thought he
 59. PAUL APPEALS TO CAESAR. 25.1-12. 1127

had taken them into the Temple (21.28f.; 24.6, 12, 18). As for the
offence against Caesar, he was not the Egyptian, or any other rebel
(21.38). He was a Roman citizen, and a loyal one. ἥμαρτον takes on
a slightly different sense in each setting—Law, Temple, Caesar.

9. χάριν καταθέσθαι; see 24.27. There χάριτα seems to be the

right reading; here it is very scantily attested and χάριν must be
original. See Μ. 2.132; also v. 3. Is Luke following here a different
written source from one used in ch. 24? The data are insufficient to
warrant such a hypothesis.
Though the Jews, or their most important representatives (v. 5),
had accompanied Festus to Caesarea and had brought their charges
before his court there, they evidently still wished to try Paul in
Jerusalem. No doubt (according to Luke) they would have liked best
to conduct the trial on their own, but a trial in Jerusalem with the
Procurator in the chair would have been better than nothing, though it
is hard to see what they would have gained by thus transferring a
Roman trial from Caesarea to Jerusalem unless some form of the plot
(v. 3) and an appeal to violence were in mind. This assumes that ἐπ’
ἐμοῦ means before me, though according to Tajra (141f.) it means
only in my presence, so that Paul would be in Jewish hands, and Jews
would pass sentence on him. But see BDR § 234.2, n. 4; Blass
(258f.); ‘Apud verbum κρίνεσθαι necessarie iudicem designat, cf.
etiam 15.2 [Blass must be referring to the Western text]; 24.21. Sic
Chrysost.: ὡσεὶ ἔλεγεν οὐχί αὐτοῖς δίδωμί σε, άλλ’ αὐτὸς κριτής
ἔσομαι.’ See also Begs. 4.308: ‘The position of ἐπ’ ἐμοῦ is emphatic.
It seems to imply that Festus undertakes not to give up Paul to Jewish
jurisdiction.’ See further below.
It does not seem possible to do more than guess at the motivation
that lies behind this verse. At one extreme, Calvin (2.262) thinks that
Festus may have been motivated by passion for gain and been ready
to gratify the Jews for this reason. Hanson (232) recognizes that ‘the
question why Festus proposed to Paul an adjournment to Jerusalem
cannot be answered’, but adds (233) that Festus may have been
attempting a compromise; Paul may, rightly or wrongly, have
thought this. Roloff (343) attacks Festus’ integrity. The question that
leaves Paul to decide whether or not to go to Jerusalem cloaks
dishonesty. If Paul is innocent, he should release him. To hand Paul
over to the Jews was to bend the law. Others think differently.
Bengel (479): ‘Poterat Festus Paulo non rogato decemere: sed
conscientia eum retinebat, et res divinitus gubemabatur, ut Paulo
daretur causa appellando Bengel adds, however, on ἐπ’ ἐμοῦ, ‘Hoc
Festus speciose addit.’ Bruce (2.452) takes the view that from
Festus’s side the suggestion of a trial in Jerusalem but with him as
judge was a reasonable enough proposal, and Bauemfeind (265) asks
what could be more natural than that the new Governor should seek
1128 COMMENTARY ON ACTS

the answer to a Jewish theological problem in Jerusalem. Knowling,

who (492) holds that ἐπ’ ἐμοῦ means ‘ “me praesente”, for the
Sanhedrists would be the judges; otherwise where would be the
favour to the Jews?’ adds (493): ‘It is possible that Festus may have
been quite sincere in his proposal: his words at least showed that in
his judgement there was no case against Paul of a political nature,
and he may have thought that religious questions could be best
decided before the Sanhedrim (sic) in Jerusalem, whilst he would
guarantee a safe-conduct for Paul as a Roman citizen.’
These various opinions all rest upon the assumption that we have
in this paragraph a factual account and that we ought to be able to
find in it, or behind it, the various motives, personal, moral,
theological, that made the actors act as they did. This is perhaps a
mistaken assumption. The point is made forcibly by Schüle in the
discussion that begins on p. 441 with the observation that there is no
ground in law for Festus’s suggestion. Schille takes as the acceptable
alternative that Luke is creating a literary motivation for the appeal.
It is ‘nicht geschichtliche, aber literarische Logik’ that determines the
story. This is certainly true to the extent that the first question to ask
is not, ‘What happened?’ (though that is a very proper question and
should be asked), but, ‘What does Luke wish to convey in this
paragraph?’ Undoubtedly he wishes to show a representative Chris-
tian as innocent on all counts and the victim of malice. But when
Haenchen (640) states, ‘Dann ist aber nicht ersichtlich, warum Festus
die Zustimmung des Paulus wünscht’ it is quite possible that there is
a historical reason, and that it is one that is creditable to the Roman
official.
The whole of this discussion would be unnecessary if we could,
with 33 pc. read ἤ before ἐπ’ ἐμοῦ. This makes very good sense in
that it presents a clear pair of alternatives (‘Would you like to be tried
in Jerusalem or before me?’), and does not involve the surprising
picture of Festus presiding in a rabbinic court. But the short text is
not only the more difficult, which would invite change, but also that
of almost all MSS. It must certainly be accepted.

10. Grammarians are agreed that though in Hellenistic Greek the

periphrastic perfect is normally a simple equivalent to the ordinary
perfect, yet here the periphrastic construction (especially if those
MSS are followed that separate ἑστώς from εἰμι—see below) adds
‘great emphasis’ (Μ. 3.88). BDR § 352.3, ‘Zuweilen dient die
Umschreibung dem rhetorisch kräftigeren Ausdruck: Apg 25.10...’.
Zerwick (§ 360), ‘ἑστὼς ἐπί τοῦ βήματος Καίσαρός εἰμι (Α 25.10)
rem multo fortius et magis pictorice exprimit quam simplex ἕστηκα
ἐ. τ. β. Καίσαρoς.’
The text of Paul’s opening words is given in three forms. (1) ἐπὶ
τού βήματος Καίσαρoς ἐστώς εἰμι, οὗ ... (P74 Ε Ψ <a>); (2) εστώς
 59. PAUL APPEALS TO CAESAR. 25.1-12. 1129

ἐ. τ. β. Κ. εἰμι, συ ... ( 453 1175 pc). These two forms have the
same meaning, I am standing in Caesar’s court, where ..., but (2) is
more emphatic; see above. (3) ἑστὼς ἐ. τ. β. Κ. ἑστώς εἰμι σὗ ... (Β
only), Standing in Caesar’s court I am standing where ... This is
more vivid still, and it may be what Luke wrote, copyists regarding
one ἑστώς or the other as superfluous. See Begs. 4.308, T think B
ought to be followed.’ If the reading of B is rejected it is probably
right to accept (1) with its numerous and ancient support. H. Sahlin
(NovT 24 (1982), 188) considers ἐπὶ τ. β. K. ‘eine sekundäre Zutat’.
In this sentence βήμα presumably denotes a place where one may
stand (as Dinkier, Signum Crucis 122, seems to agree), though there
is no reason to suppose that trials would always and necessarily be
held in that place. The significance of the place is, however, that
trials are held in it; the English court seems to have a similar force,
meaning both the place and the judicial institution that makes use of
the place. Paul, as the context will show, means that only Caesar (or
his representative) has the right to pass judgement on him. This is not
without difficulty; see Conzelmann (134). Can Paul mean that it is in
Caesar’s court that the question can be settled whether or not his
interpretation of Torah (conditioned by the death and resurrection of
Jesus) is correct? Scarcely. Hitherto he has insisted (23.6; 24.21) that
the real issue behind his trial is resurrection; in what sense is Caesar a
judge of that issue? Caesar may be said to have a right to judge
whether Paul has acted as a rebel or rioter or disturber of the peace
within a Roman province, but this is only a third of the charge as Paul
has stated it in v. 8. The simple fact is that (in Luke’s narrative) Paul
feels safer with Romans than with Jews and thinks that involvement
with the former is a small price to pay for freedom from the latter.
The Romans provide a way of escape from the Jews who have made
up their minds to have his blood; what is more, pursuit of the Roman
process offers a means by which the intended journey to Rome may
be achieved (19.21; 23.11). The Jews are his accusers and therefore
cannot act as judges of the truth of his claim, ’Ιουδαίους ούδέν
ἠδίκησα (ἠδίκηκα, Β (81)). Seditio is the charge that is left, and
that means a Roman court, and even if Festus presided in Jerusalem
(see ν. 9, ἐπ’ ἐμοῦ) the court would be Jewish rather than Roman; or
so Paul feared. Conzelmann’s judgement (134) of Festus (‘Festus
allein(!) ist schuld daran, dass der Prozes weitergeht. Paulus ist
gezwungen, an den Kaiser zu appellieren’) is not fair. Paul made use
of the privilege of appeal (v. 11) in order (a) to escape the Jews, (b) to
make the journey to Rome.
Paul believes that he can even cite Festus as a witness. He knows
the facts of the matter well. κάλλιον, comparative, may be used as a
superlative; so BDR § 244.2, n. 3 (κάλλιον = ἄριστα). Cf. 1 Tim.
1.18. Not all agree. Wertstem 2.626: ‘Multo melius, quam ego
possum exponere’; Bengel 479: melius quam alii. More to the point
1130 COMMENTARY ON ACTS

is that the verb is ἐπιγινώσκειν (cf. 24.8). Knowling (493) translates,

‘As thou also art getting to know better.’ He adds that otherwise an
‘ungracious and unjust’ retort is ascribed to Paul. Delebecque (121)
similarly: ‘... comme tu es en train de le mieux découvrir’.
According to Begs. 4.308 κάλλιον ‘is the “intensive” comparative
... it is however possible that κάλλιον ἐπιγινώσκεις ... is ... the
comparative of politely qualified expression’. Cf. 17.22.

11. As in ν. 4, μὲν συν is not Luke’s characteristic formula; there

is a δέ (Begs. 4.308).
αδικώ is to be compared with ἠδίκησα (or ἠδίκηκα, v. 10) and
with πέπραχα (Wettstein 2.626 quotes Moeris: πεπραγώς,
Άττικῶς· πεπραχώς, Έλληνικῶς); the present tense has the effect
of summing up the perfects which refer to particular acts of
wrongdoing (BDR § 322.1, n. 1). It is in fact a perfective present and
can be rendered, I ama wrongdoer, or, I am in the wrong. ἀδικεῖv is
used here as in Xenophon, Memorabilia 1.1.1, αδικεί Σωκράτης ...
οὐ νομίζων ... εἰσφέρων ... διαφθείρων, Socrates is guilty ... of
not believing... of bringing in... of corrupting... The use in Plato’s
version of the charge against Socrates is somewhat different (Apol-
ogy 19b), Σωκράτης αδικεί καί περιεργάζεται... ζητών ... ποιῶν
.. .διδάσκων ... Socrates does wrong and goes about... seeking ...
making ... teaching ... The latter use is more common; cf. Aristotle,
Rhetoric 1.10.3, ’Έστω δὴ τὸ αδικεῖν, τό βλάπτειν ἔκοντα παρά
τόν νόμον.
παραιτούμαι has two meanings, to entreat for something, but also
to refuse, to decline, to avert by entreaty (LS 1310f.). It clearly has
the latter meaning here. Cf. Plutarch, Coriolanus 20 (223), ...
κολάσεως παραιτεῖσθαι; Euripides, Heraclidae 1026, κτεῖν’, ού
παραιτούμαι σε, and especially Josephus, Life 141, θανεῖν μἐν εἰ
δίκαιόν έστιν, ού παραιτούμαι. If Paul can be found guilty he will
not attempt to buy off the appropriate penalty—τό ἀποθανεῖν; the
article is anaphoric (BDR § 399.1, n. 2) referring back to θάνατος
immediately above.
ὧν: assimilation of the relative. If there is nothing in the things of
which they accuse me. It is clear that what Paul fears is that his
Jewish opponents should be given a free hand. As long as he is in
Roman hands he is safe; among Jews his life would be worthless. He
knows his rights: οὐδείς με δύναται αὐτοῖς χαρίσασθαι. Cf. ν. 16.
Begs. 4.309 says 'One can hardly say, “make a present of me” ’. In
the course of 60 years the expression has, I think, become usable.
Paul knows his rights and proceeds to make them as secure as
possible by appealing over Festus’ head to Caesar.
LS 635f. and BA 595f. give no other example of ἐπικαλεῖσθαι
Καίσαρα. They quote Plutarch, Marcellus 2 (299), τούς δημάρχους
ἐπικαλούμενος, appealing to the tribunes of the people, and Tiberius
 59. PAUL APPEALS TO CAESAR 25.1-12. 1131

Gracchus 16 (832), ἐπικαλεῖσθαι τὸν δήμον ἀπό τών δικαστών, to

appeal from the judges to the people. For Roman law on appealing to
the Emperor see Sherwin-White (68-70) (with his reference on 59
to A. Η. Μ. Jones); also Tajra (144-7). At this point in the
development of Roman practice the institution was not the later
appellaiio, by which (as in modern English law) a condemned and
sentenced person might apply to a higher court to have verdict, or
sentence, or both, changed, but provocatio, which was an appeal
before trial to a higher court which would then take the whole case,
trial, verdict, and sentence out of the lower court. It is clearly this
procedure that is described in Acts. Attempts of one kind and another
have been made to try Paul, but in none has a verdict been reached.
Paul knows, or believes, that trial before Jews would be fatal,
perhaps thinks that Festus is weakening, and therefore plays his
trump card. Cf. Schneider (2.356f.) (against Haenchen): ‘Für Lukas
ist das genannte Motiv für die Appellation des Paulus mit der
Befürchtung verbunden, der Stadthalter könne den Juden weiter
entgegenkommen und ihn schliesslich ausliefem.’ Henceforward he
will appear before the Governor, but only in fact-finding inquiries,
designed to provide Festus with material for the report he would be
obliged to send.

12. μετά τοῦ συμβουλίου, συμβούλιον: LS 1677, s.v. II, give 'a
council of advisors or assessors, PTebt 286.15 (ii AD), Plutarch,
Lucullus 26 (509); esp. freq. of the consilium of a Roman magistrate,
governor, etc.'. Numerous examples are given, including inscrip-
tions; interesting as showing the linguistic equivalent is Plutarch,
Romulus 14 (25), κωνσίλιον γὰρ ἔτι νυν τό συμβούλιον καλοῦσι.
Many examples of such consilia can be quoted, e.g. Horace, Carmen
4.5.4; Satire 1.7.23; Cicero, Pro Sex. Roscio Amerino 52 (151); CIL
10.7852, in consilio fuerunt Μ. Iulius Romulus leg. Propr., T. Atilius
Sabinus, etc. It was established custom that the Governor should (a)
consult his council, and (b) make his own decision (Tajra 148f.).
It is not clear what in this case Festus would have to discuss with
his council. Assuming that Paul was a Roman citizen (and it was
unnecessary for Luke to mention this matter again after 22.25-29)
the Governor was bound to grant the request that he made (see
Sherwin-White 63f., against Lake in Begs. 5.311, 317, 318). He
could not disallow the appeal, under the Lex Iulia de vi publica
(passed in the time of Augustus), of which the relevant part is given
in Digest 48.6, 7 (Ulpian): Lege Iulia de vi publica tenatur qui cum
imperium potestatemve haberet civem Romanum adversus provaca-
tionem necaverit verberaverit iusseritve quid fieri aut quid in collum
iniecerit ut torqueretur. Another form in the Sententiae Pauli 5.26.1,
2 adds, condemnaverit vive publica vincula duci iusserit. Accord-
ingly Festus ἀπεκρίθη: he answered (Paul’s appeal). Here however
1132 COMMENTARY ON ACTS

the word probably has the sense of giving an official opinion. The
cognate noun ἀπόκριμα is sometimes used with this meaning: so LS
204 (‘esp. of the answers given by Emperors to legationes'); see also
Josephus, Ant. 14.210 (plural; the Latin version has responso); 2 Cor.
1.9. Festus may have consulted the consilium with a view to his
report to the Emperor.
Καίσαρα ... πορεύση: an impressive and justly famous epigram.
Paul preferred a Roman to a Jewish court. In this sense, ‘Der
Appell an den Kaiser ... symbolisiert... die Scheidung der entste-
henden Kirche vom Judentum’ (Pesch 2.267). ‘Paulus ist damit
endgültig dem jüdischen Bereich entrissen’ (Roloff 344); but it is
equally true that Paul’s first act on reaching Rome was to call
together the leaders of the Jewish colony (28.17).
Bruce (1.433) thinks it possible that Paul was appealing to Caesar
in the person of his representative; Festus used the words that Paul
had spoken as an easy way out of a difficulty, ‘If you appeal to
Caesar you shall have Caesar himself, not his representative in
Judaea.’ But Lüdemann’s conclusion is probably correct (263):
‘Höchst wahrscheinlich hat vor Festus in Cäsarea ein Prozess unter
Beteiligung der jüdischen Führung Jerusalems gegen Paulus stattge-
funden, in dessen Folge Paulus unter Berufung auf sein römisches
Bürgerrecht an den Kaiser appelliert.’
This is a very important point in Luke’s book. ‘Die Berufung an
den Kaiser—und damit im Sinn des Lukas: die Eröffnung des Weges
nach Rom—ist der Höhepunkt, auf den alles Bisherige hinzielt’
(Weiser 642).
 60. FESTUS AND AGRIPPA 25.13-22

(13) When some days had passed King Agrippa and Bernice came to
Caesarea and greeted Felix. (14) As they stayed there many days Festus
referred Paul’s case to the king, saying, ‘A man has been left a prisoner by
Felix (15) concerning whom, when I was in Jerusalem, the chief priests and
elders of the Jews laid information asking for a sentence against him. (16) I
answered them that it was not the Romans’ custom to hand over any man
before the accused has had his accusers face to face and received the
opportunity of making a defence against the charge. (17) So they came here
with me and I made no delay but on the next day sat in court and commanded
the man to be brought. (18) Concerning him his accusers stood and brought
no charge of those evil things that I suspected, (19) but they had disputes
with him about their own religion1 and about a certain Jesus who had died
and whom Paul alleged to be alive. (20) For myself, I was at a loss over the
inquiry about these things and asked if he would be willing to go to
Jerusalem to be tried on these matters. (21) But when Paul appealed to be
kept for the Emperor’s decision I gave orders that he should be kept until I
should send him up to Caesar. (22) Agrippa said to Festus, 'I could wish to
hear the man myself.’ ‘Tomorrow’, said Festus,2 ‘you shall hear him.’

Bibliography
C. D. Chambers, JTS 24 (1923), 183-7.
J. Dupont, Études, 527-52 (= RScR 49 (1961), 354-85).
W. F. Howard, JTS 24 (1923), 403-6.
S. Légasse, as in (54).

Commentary
Weiser (see above, p. 1122) holds that the episode involving King
Agrippa was introduced by Luke into the story of the judicial
proceedings against Paul in order to create a parallel with the story of
Jesus (Lk. 23.6-12). There is much to be said for this view; the
historicity of the incident is not easy to defend (see Hemer 348f.). It
is however worth while to observe two points of difference. The
Agrippa section is far longer than the Herod section: 47 verses
(25.13-26.32) against 7. There is far more detail. And in Acts,
though she does not contribute to discussion or to the logical
1RSV, superstition.
2Greek, he.
1133
1134 COMMENTARY ON ACTS

development of the story, Bernice is on stage all the time and in this
sense plays a substantial part. She has no counterpart in Lk. We have
reason to think that Agrippa and Bernice did travel together, and
made courtesy visits to other rulers. At Josephus, Life 49, they went
to Berytus to visit (ὑπαντήσαι) Cestius (Governor of Syria). At War
2.309 Agrippa visited Alexander at Alexandria, congratulating him
on his appointment in Egypt. The following sections (310-314) do
not expressly say that Bernice would have accompanied him if she
had not been preoccupied in Jerusalem, but this is an easy inference.
There is thus nothing impossible in the statement that the couple
visited Festus early in his governorship, and if Luke, hunting in
Jerusalem for memories of Paul, discovered the fact, he had at once
the basis of his story. Whether he had further information there is no
means of judging; he must be himself responsible for the conversa-
tion between Festus and Agrippa, which has the effect of presenting
Festus as an honourable Roman. ‘So hat der kleine Abschnitt viel
geleistet: Festus persönlich ist vor dem Leser rehabilitiert’ (Haen-
chen 646). Perhaps on the whole more fairly: .. wollte [Lukas]...
zeigen: so hat ein Römer die Sache des Paulus gesehen’ (Stählin
302). This Roman view completely omits the serious Roman charge
of seditio and laesa maiestas; because there was no evidence?
Throughout the behaviour of Festus is capable of a better inter-
pretation than is often given to it; but of course we are viewing
Festus as Luke intends that we should see him, in contrast with the
Jews.

13. Ήμερων ... τινῶν. All one can say of the period is that it
seemed more natural to measure it in days than in years (Luke and
other biblical writers do not often use weeks and months to express
intervals). The story gives the impression that Festus has not made up
his mind what to do with the prisoner whom he has just consigned to
Rome.
Άγρίππας ὁ βασιλεύς. For the use of the article cf. 12.1 and see
BDR § 268.1 (without reference to this passage). This (Herod)
Agrippa was the son of the Herod (Agrippa) of 12.1. See NS
1.471-483. Claudius would have permitted him to succeed his father
when his father died (see I.573f.), but allowed himself to be
persuaded not to do so. Agrippa, who had been educated in Rome,
remained there at least till the death of his uncle, Herod of Chalcis, in
48. Shortly after this event Claudius conferred on him his uncle’s
kingdom together with charge of the Temple in Jerusalem and the
right to appoint the high priest (Josephus, Ant, 20.222,223). When he
entered upon his kingdom is not certainly known; according to NS
1.472 it was not till after 52. Later his kingdom was extended to the
north. He took the Roman side in the Jewish War, after which he
retained no territories west of Jordan. In inscriptions (see NS 1.475,
 60. FESTUS AND AGRIPPA. 25.13-22. 1135

n. 15) he is described as βασιλεύς μέγας φιλόκαισαρ ευσεβής καί

φιλορώμαιος. According to Photius, Agrippa II died in AD 100 (the
third year of Trajan). The matter is discussed at length in NS
1.481-3, with the conclusion that 92 or 93 is a more probable date.
Festus and Agrippa seem to have been on good terms (Josephus, Ant.
20.189-196). Pesch (2.269, 273) argues against Conzelmann (135)
that Luke would not have sought out a pair like Agrippa and Bernice
to provide an occasion for an additional appearance of Paul.
There must have been a pre-Lucan source which at least alluded to
Agrippa. ‘Folglich ist unsere Szene wahrscheinlich nicht völlig “frei
entworfen".’
Βερνίκη was Agrippa II’s sister. See NS 1.474-6, 479. She had
been married to Herod of Chalcis and bore him two sons. After
his death she lived with her brother, and much scandal was talked
about the relation between them (Josephus, Ant. 20.145, φήμης
ἐπισκούσης, ὅτι τἀδελφῷ συνείη; see also Juvenal, Satire
5.156-160). She was also said to have been mistress of the Emperor
Titus (Suetonius, Titus 7: insignem reginae Berenices amorem, cui
etiam nuptias pollicitus ferebatur; also Tacitus, Histories 2.2; cf. Dio
Cassius 65.15.3f.; 66.18.1). To dispel the scandal about her relation
with Agrippa she married Polemo, King of Cilicia, but before long
returned to her brother (Josephus, Ant. 20.146). See a long note
in Hemer (173f.); also Begs. 4.309. On her name see Hemer (238)
and BDR § 42.1. It is more correctly written Βερενίκη (= Attic
Φερενίκη); so 1175 pc sa; C*v,d has Βερηνίκη.
ἀσπασάμενοι. The visit was a courtesy call. The tense of the
participle has been much discussed. The future participle
(ἀσπασόμενοι) might be expected as a normal manner of expressing
purpose, and this is read by Ψ 36 81 323 1739 1891 2495 pm lat sy
sa. In the versions this may well be a matter of correct interpretation
(e.g. Vulgate, ad salutandum Festum), but in Greek texts it must be a
‘correction’. C. D. Chambers (JTS 24 (1923), 183-7) and W. F.
Howard (JTS 24 (1923), 403-6) argued that here the aorist participle
was used to express purpose; so also Williams (260). It is probably
correct to speak of coincident action; so Μ. 1.132f., observing that
some classical precedent can be found for this in Pindar, Pythians
4.189 (λέξατο πάντας ἐπαινήσαις), which Bruce (1.435) renders
‘mustered and praised them’, adding that the phrase in Acts in
equivalent to κατήντησαν καί ἠσπάσαντο. Cf. Metzger (492), who
gives a full bibliography on the grammatical question. Haenchen
(643) comments, ‘Begrüssung und Besuch fallen zusammen’. Preu-
schen (143) somewhat differently: ‘Die von den meisten Hss gebo-
tene Lesart ἀσπασάμενοι setzt voraus, dass die Begrüssung in
Jerusalem stattfand.’ See also Μ. 3.80. Zerwick (§ 264) notes the
view that understands ‘adventum et salutationem quasi ut actionem
unam quae esset “visitatio (adventus) salutaria" (actio coincidens)’,
1136 COMMENTARY ON ACTS

but adds (§ 265) that the action may be subsequent. See also Moule
(IB 100, 202); BDR § 339.1, n. 4 (‘wobei sie begrüssten’); § 351.1,
η. 2.
Agrippa, Bernice, and the men of 25.23, are like the council of
assessors in 25.12. Begs. 4.310 compares Josephus, Ant. 17.93
(Quinctilius Varus, with Herod the Great and his sister Salome, and
others; in War 1.620 Salome is not mentioned). Cf. also Ant. 16.30.

14. Again, as in v. 13, there is a quite imprecise note of time; with

this the imperfect διέτριβον is consistent; cf. 14.28 and see BDR §
332.1, n. 2.
ἀνέθετο is most naturally taken in the technical legal sense of
refer, remit. Surprisingly LS 123 and MM 38 both prefer impart,
communicate; the accompanying expression τά κατά τόν Παύλον
(see Μ. 3.15), Paul’s case, has a definitely legal sound and carries
ἀνέθετο with it. Of course the reference here is an informal one.
Agrippa and Bernice do not constitute a higher court. The only court
to which Paul’s case could after his appeal be formally remitted
was the Emperor’s. But Festus felt himself in need of consultation
beyond that which he had had with his συμβούλιον (25.12), and took
the opportunity afforded by the neighbourly visit. Festus begins
to sum up the case so far as it had gone. With this summing up
Ehrhardt (Acts 119f.), compares Pliny, Epistles 10.74 (the case of
Callidromus).

15. γενομένου μου εἰς 'Ιεροσόλυμα, when I was in Jerusalem.

This use of εἰς (where ἐν would be expected) is described by Μ.
3.254 as ‘especially Semitic’. It cannot be so here, and Turner in the
context shows that the distinction between the two prepositions
became blurred in Hellenistic Greek. The present is one of many
examples in Acts of this blurring.
ἐνεφάνισαν, they laid an information. See on 23.15; 25.2.
oἱ αρχιερεῖς καί οἱ πρεσβύτεροι. See 4.23; 23.14.
αἰτούμενοι καταδίκην, asking for sentence against him. The
variant δίκην (Ε Ψ <a>), in view of κατ’ αὐτοῦ, implies no different
sense: they were asking for judgement against him. Cf. Herodotus
1.2.3,... πέμψαντα ... κήρυκα αἰτέειν τε δίκας ...

16. Cf. 25.4f. Festus represents himself as having taken a high

moral tone with the Jewish accusers. Luke as usual shows Roman
officials in a good light; see J. Dupont, ‘Aequitas Romana’, in Études
527-52. Roman custom (on έθος as the source of law see Tajra 155)
must be observed; but the Roman custom is oddly described in the
words χαρίζεσθαί τινα ἄνθρωπον. The meaning of χαρίζεσθαι
must be the same as in 25.11, but here there is no dative. To translate,
‘... to give up anyone’ would fail to convey the sense of doing a
 60. FESTUS AND AGRIPPA. 25.13-22. 1137

favour. The lack of a dative was felt by the copyists of C 945 1739
1891 pc, who write τινι for τινα; the deficiency was made up
differently by <a> gig sy sa, which add εις απώλειαν after άνθρωπον.
The dative must simply be understood: It is not our custom to make a
present of any man to anyone. An alternative possibility might be to
translate χαρίζεσθαι as to favour, but for this the verb should be
followed by a dative, not an accusative. There is some slight Latin
authority for damnare (instead of donare).
πρὶν ἢ ... ἔχοι... λάβοι. This construction, as indeed the whole
speech, shows Luke’s ability to put suitably classical Greek (‘kunst-
mässig durchgeführte indirekte Rede’, Radermacher 131) on the lips
of educated men. Cf. Introduction, p. xlv. On notera les deux
optatifs d’une langue très pure’ (Delebecque 122). The optatives
replace the subjunctives that would be used in direct speech (cf. Lk.
2.26). See Μ. 1.169; Moule, IB 133; BDR § 386.3, n. 4; Zerwick (§
346). This is the only example in the NT of this use of the optative.
See Schille (444).
Accused and accusers must come face to face. κατὰ πρόσωπον
‘used adverbially, as in Acts 25.16; 2 Cor. 10.1; Gal. 2.11, is
certainly not Semitic, but the prepositional use in Lk. 2.31; Acts 3.13,
though not uncommon in Greek (cf. Xenophon, Cyropaedia 6.3.35,
τὴν κατὰ πρόσωπον τῆς ἀντίας φάλαγγος τάξιν, “the post imme-
diately in front of the enemy’s phalanx”), is suggested by the OT
idiom’, Μ. 2.466; similarly BDR § 217.1c, n. 3.
The metaphorical use of τόπος (= opportunity) seems to be
mainly biblical. At Thucydides 6.54.4, LS 1806 think τρόπῳ a
probable conjecture. Preuschen (143) speaks of τόπος as a ‘ver-
breiteter Latinismus’ (= locus) but cites only Sirach 4.5 (where locus
is not used in the Vulgate!); 1 Macc. 9.45; Josephus, Ant. 16.258. See
however Wettstein 2.628: ‘τόπον ... απολογίας λάβοι] ex Latino
sermone formata locutio, locum respondendi, i.e. facultatem, potesta-
tem accipere'.
απολογία and ἔγκλημα are both technical forensic terms; see
22.1; 23.29. Properly an ἔγκλημα is a written accusation, but there is
no need to suppose that Luke had this in mind. MM 139f. quote only
papyri; ND 3.66 (cf. 4.86) adds Dittenberger, OGIS 1.229.41.
For the Roman principle appealed to see Digest 48.17.1: et hoc
iure utimur, ne absentes damnantur neque enim inaudita causa
quemquam damnari aequitatis ratio patitur. Cf. Justin, 1 Apology 3;
Tertullian, Apology 1.3; 2.2; Appian, Bellum Civile 3.54: ὁ μὲν
νόμος, ὦ βουλή, δίκαιοῖ τὸν εἠθυνόμενον αυτόν ἀκοῦσαί τε τής
κατηγορίας καὶ ἀπολογησάμενον υπέρ αὐτοῦ κρίνεσθαι; Tacitus,
Histories 1.6.
‘Das sagt Festus, obgleich er sich nicht an den Grundsatz gehalten
hatte (siehe VV 9-11)’ (Schneider 2.363). This seems unfair; we do
not know what Festus would have done if Paul had not appealed.
1138 COMMENTARY ON ACTS

Hanson (235) raises the question whether the Jewish court could
have tried Paul in a capital charge and executed him; see Jn 18.31.
17. συνελθόντων: in the great majority of MSS the genitive
absolute is completed by the addition of αυτών; only B pc omit the
pronoun. It is probably correct to omit; it is hard to see why if αυτών
originally stood in Luke’s text it should have been dropped, whether
accidentally or intentionally. And BDR § 423 has (without reference
to this verse), ‘Der Gen. eines Ptz. kann auch ohne Subst. im Gen.
stehen, wenn es sich ohne weiteres ergänzen lässt.’ Metzger (492f.)
defends the short text. Cf. Lk. 12.36; Acts 21.31.
αναβολήν ... ποιησάμενος, to make a delay, as at Thucydides
2.42.3, ... αναβολήν τοῦ δεινοῦ ἐποιήσατο; Plutarch, Camillus 35
(147), ... ἔγνω μή ποιεῖσθαι τής τιμωρίας αναβολήν. Luke (like
Thucydides and Plutarch, uses the middle of the verb; the active is
often used with the same meaning. See 24.22.
ἐπὶ τού βήματος: cf. 25.6,9,10. It is usually said that the meaning
here must be on the (judgement-)seat. This is not necessarily so,
since καθίζειν can mean to sit as judge (LS, s.v. II 3; 854); e.g.
Herodotus 1.97.1; Plato, Laws 659b; Philo, De Ebrietate 165
(καθίσας οὖν ὁ νοῦς ἐν τῷ εαυτοῦ συνεδρίῳ ... ). Cf. the English,
‘Judge X is sitting at the Old Bailey’.
For the passive infinitive after a verb of commanding see on
22.24.
18. It would be natural to suppose that a man for whom severe
punishment was demanded would be accused of serious crime. A
Roman official could hardly be expected to understand that to ‘deny
the root’ of Jewish religion was the most serious crime
imaginable—the crime of Esau, for example (Baba Bathra 16b).
Luke is emphasizing Paul’s innocence from the Roman point of
view.
αιτίαν φέρειν (or ἐπιφέρειν, with 6 104 1241 <a>) is a standard
forensic expression: Herodotus 1.26.3, ἄλλοισι ἄλλας αιτίας
ἐπιφέρων; Thucydides 5.75.2; Philo, De Josepho 184, αιτίαν ἡμῖν
ὡς κατασκόποις ἐπέφερεν; Plutarch, Aratus 45 (1048).
ὑπενόουν is probably I supposed, though to suspect is a fairly
common meaning of the verb. Sherwin-White interprets differently:
‘ “The accusers brought no charge against him of any evil act that I
could understand’’, ὧν ἐγὼ ὑπενόουν. The word is pejorative, and at
its strongest means “suspect”. This phrase may well correspond to
the formula “any act of which I was prepared to take cognizance”,
“de quibus cognoscere volebam”.’ Hemer (131) is more positive:
ὧν ἐγὼ ὑπενόουν, ‘ “of which I could take cognizance,” reflecting
the legal formula de quibus cognoscere volebam'. But there is
nothing in Luke’s Greek corresponding to volebam. In fact, ὧν ...
πονηρών, expressed thus by attraction of the relative, is equivalent to
 60. FESTUS AND AGRIPPA. 25.13-22. 1139

τῶν πονηρῶν ἃ ἐγὼ ὑπενόουν (Zerwick § 18), and means ‘they

brought no accusation of those evil things that I supposed (or
suspected).’ Copyists may have been confused by the sentence.
πονηρών is read by Β E 81 104 pc; πονηράν by P74 A C* Ψ 36
614 945 1175 1739 1891 2495 al gig vgww (sy); πονηρά by C2 w.
The word is omitted by and may be a gloss. The sentence
makes sense without it: no charge of those things that I supposed
(suspected).
19. Paul was accused not of crime but of heresy, matters of
dispute, ζητήματα, concerning their own religion. To Paul the Jew
and Christian the religion of Athens was δεισιδαιμονία (17.22); to
Festus the religion of Christians and Jews was δεισιδαιμονία. On
this word and its cognates see on 17.22.
So far Paul would have agreed with Festus’s assessment; he is
reported as claiming (and in the epistles did in fact claim) that
Christianity was true Judaism, in the sense that it was the fulfilment
of Judaism. He would also have agreed (though in each case he
would have expressed the matter differently) with Festus’s second
point: the dispute between Paul and the Jews turned upon a certain
Jesus who had died but was affirmed by Paul to be alive. Cf. Rom.
10.9: Jesus is Lord; God raised him from the dead. ‘Nul n’a mieux
exprimé l’essence de la religion nouvelle et ce qui sépare doctrinale-
ment Paul des Juifs’ (Delebecque 122). More circumspectly, one
must recognize that Luke can only have guessed at the contents of
any conversation between Agrippa and Festus; but it was a good
guess—at least, one creditable to Festus. There is an interesting—
but not important—verbal parallel in Lucian, De Syria Dea 6:
πρῶτα μὲν καταγίζουσι τῷ Άδώνιδι, ὅκως ἐόντι νέκυϊ. μετά δέ τῇ
ἑτέρη ήμερη, ζώειν τέ μιν μυθολογέουσι. ‘Es geht nicht um
“Anerkennung des Christentums durch den Staat”, sondern um
“Ausscheidung aus dem Geschichtsverfahrung” ’ (Schneider
2.363).
The construction is changed by P74vid sa, which place ἥν after
δεισιδαιμονίας.
20. The drift of the verse is clear, but there are several difficult
questions of grammar, ἐγώ is clearly emphatic: the Jews (including
Paul) thought they knew all about the affair, but as for me I was quite
at a loss. ἀπορούμενος (middle) is unusually followed by an
accusative, rather than by ἐν or περί (εις is introduced here by C E L
Ψ 33 36 323 614 945 1175 1739 1891 2495 al). The active
sometimes takes an accusative; here, unless the accusative is treated
as one of respect, we have an example of an intransitive verb used
transitively. See BDR § 148.2, n. 4; Μ. 3.244.
ζήτησις is here (judicial) inquiry. This is illustrated by an
inscription quoted in ND 4.86.
1140 COMMENTARY ON ACTS

ἔλεγον: ‘impf. quia condicionem fert, quae res per se est imper-
fecta, donee accedat alterius consensus, cf. 21.12 παρεκαλοῦμεν al.’
(Blass 261).
εἰ βούλοιτο. Μ. 3.127: Other clauses introduced by εἰ and
dependent on a verb like ζητεῖν are virtually indirect questions, a
class. survival: Acts 17.11; 25.20.’ This is well enough in 17.11,
where the verb is ἀνακρίνειν; it cannot be so easily applied here,
where the verb is λέγειν, which is not like ζητεῖν and seems never to
introduce an indirect question. Most modern versions (but contrast
Vulgate dicebam, Syriac translate ask, but this is not the
meaning of λέγειν. It seems in fact impossible to defend the
construction, or to improve on Alford’s comment (ad loc.; 2.273),
‘There is a mixed construction between “7 said, Wilt thou?” as in
ver. 9, and “7 asked him whether he would.It ". '.is. not the only
imperfectly constructed sentence in Acts, or the worst; it is unfortu-
nate that it occurs where Luke is evidently intending (see the
optatives in v. 16) to write in a good style. Moule (IB 154) gives
‘more loosely and idiomatically’ the rendering, 'I said, would he like
...?’ This loose but idiomatic English represents the Greek well, but
the Greek, though loose, is scarcely idiomatic.
Festus does not here, as he does, or may do, or appears to do, in
25.9, say that he himself would preside over a hearing in Jerusalem,
but it is evident that Luke is making no attempt to reproduce the
original proposal verbatim.
For περὶ τούτων, H P 049 323 1241 pm have περί τούτου.

21. τοῦ δὲ Παύλου ἐπικαλεσαμένου, genitive absolute, though

Paul reappears in the accusative case (αὐτόν) later in the sentence,
and indeed twice. In the first case the reflexive pronoun might have
been expected, or better αυτός (Delebecque 123). See BDR § 283.1,
n. 3.
τηρηθῇναι, τηρεῖσθαι. The passive infinitives are Latin rather
than Greek: BDR § 392.4, n. 14. Cf. v. 17. τηρηθῆναι is constative,
τηρεῖσθαι describes the process; each leads to a conclusion, εἰς,
έως.
<o> Σεβαστός is the usual Greek term for the Emperor (Augustus);
contrast Lk. 2.1.
διάγνωσις correponds to the Latin cognitio; cf. διαγινώσκειν,
23.15; 24.32. In 7G 14.1072.4f. the office a cognitionibus Augusti is
rendered ἐπὶ ... διαγνώσεων τού Σεβαστού. There is papyrus
evidence in MM 147; ND 1.47; 4.86. Delebecque (123) translates
jusqu'à la décision; ‘la preposition εἰς implique une captivité qui doit
durer jusqu’à Rom.’
ἀναπέμψω: many MSS (H L P 049 323 pm) have πέμψω, but
ἀναπέμπειν is the technical term for the remission of a case to a
higher court and is very probably correct.
 60. FESTUS AND AGRIPPA. 25.13-22. 1141

For ἐπικαλεσαμένου, Ρ74 Ψ 1739* pc have the less appropriate

present participle ἐπικαλουμένου.
The Old Latin MS gig (see Ropes, Begs. 3.231) has a different
form of the verse: tunc paulus appellavit cesarem et petiit ut
reservaretur ad augusti cognitionem; cumque eum non possem
judicare jussi eum reservari ut remittam eum cesari.—an example,
probably of the sort of rewriting that lies behind the Western text. On
this basis Clark (158) reconstructed the Western text thus: τότε ὁ
Παῦλος ἐπεκαλέσατο Καίσαρα καὶ ἠτήσατο τηρηθῆναι αυτόν εις
τὴν τού Σεβαστοῦ διάγνωσιν, ἐπειδή τε αὐτόν οὐκ ἐδυνάμην
κρῖναι, ἐκέλευσα ... Metzger (493) expresses no opinion.

22. Άγρίππας ... Φῆστον. Ρ74 C Ε Ψ <a> gig vgcl sy add ἔφη, but
the short text is no doubt correct. For the ellipse of the verb see BDR
§ 480.5a, n. 6 (with Hellenistic and papyrus parallels).
ἐβουλόμην, desiderative imperfect, ‘possibly a fossilized relic of a
conditional clause ... If so, it has lost its ἄν’ (Moule, IB 151, cf. 9). It
takes the place of the older βουλοίμην ἄν (Zerwick § 356), but
occurs also ‘in der guten Literatursprache’ (Rademacher 128,
quoting Dionysius of Halicarnassus, De Demosthene 1087 (42),
ἐβουλόμην ἔτι πλείω παρασχέσθαι παραδείγματα). See also Aris-
tophanes, Frogs 866 (890), ἐβουλόμην μὲν οὐκ ἐρίζειν ἐνθάδε.
Begs. 4.312 translates, I had wished.
αὔριον is ‘Vorzugsvokabel des Lukas’ (Schneider 2.364).
 61. FESTUS, AGRIPPA, AND PAUL 25.23-26.32

(23) On the next day, when Agrippa and Bernice came with great pomp and
entered the audience chamber, along with the tribunes and the leading men
of the city, and when Festus had given the word of command, Paul was
brought. (24) Festus said, ‘King Agrippa, and all you gentlemen who are
present with us, you behold this man, concerning whom the whole people of
the Jews petitioned me, both in Jerusalem and here, shouting out that he
ought to live no longer. (25) I however did not see that he had done anything
worthy of death, and when he himself appealed to the Emperor I decided to
send him. (26) I have nothing definite about him to send in writing to the
Emperor;1 for this reason I have produced him before you, and especially
before you, King Agrippa, in order that, when examination has taken place, I
may get something that I may write. (27) For it seems to me nonsense to send
a prisoner and not at the same time to signify the charges against him.’
(1) Agrippa said to Paul, ‘It is permitted you to give an account of
yourself.’ Then Paul stretched out lus hand and began his defence. (2) 'I
consider myself fortunate, King Agrippa, that I am today about to make my
defence, concerning all the things of which I am accused by the Jews, before
you, (3) especially since [I know]2 that you are familiar with all the customs
and disputes current among the Jews. So I beg you to hear me patiently. (4)
All the Jews know my manner of life from my youth, which was from the
beginning within my own nation and in Jerusalem. (5) They have known
from of old, if they are willing to testify, that I lived in accordance with the
strictest party within our religion, a Pharisee. (6) And now I am standing trial
for the hope of the promise made by God to our fathers, (7) [the promise]3 to
which our twelve-tribe people, zealously worshipping [God]4 night and day,
hope to attain. It is for this hope that I am being accused, O King, by Jews.
(8) Why is it judged incredible with you that God raises the dead? (9) I
myself thought that I ought to do many things contrary to the name of Jesus
the Nazoraean. (10) And this I did in Jerusalem, and I shut up in prison many
of the saints, having received authority from the chief priests; and when they
were being killed I cast my vote against them. (11) And often in all the
synagogues I tried by punishing them to make them blaspheme, and being
exceedingly mad against them I persecuted them even to foreign cities. (12)
And as I was travelling to Damascus with authority and commission from the
chief priests, (13) at midday, on the road, I saw, O King, a light from heaven,
beyond the brightness of the sun, shining round me and those who were
travelling with me. (14) We all fell to the ground and I heard a voice saying
to me in the Hebrew5 language, “Saul, Saul, why are you persecuting me? It
1Greek, the lord (i.e., the Emperor).
2I know is not in the Greek.
3The promise is not in the Greek.
4God is not in the Greek.
5Aramaic is probably intended.
1142
 61. FESTUS, AGRIPPA, AND PAUL. 25.23-26.32. 1143

is hard for you to kick against the goad.” (15) I said, “Who art thou, Lord?”
The Lord said, “I am Jesus whom you are persecuting. (16) But get up and
stand on your feet. For this is why I have appeared to you, to appoint you a
minister and witness both of the things that you have seen and of those in
which I shall appear to you, (17) rescuing you from the people and from the
Gentiles, unto whom I now send you, (18) to open their eyes, so that they
may turn from darkness to light and from the authority of Satan to God, so
that they may receive forgiveness of sins and a lot among those who have
been sanctified by faith in me.” (19) Consequently, King Agrippa, I did not
prove disobedient to the heavenly vision, (20) but first to those in Damascus
and in Jerusalem, and in all the land of Judaea and to the Gentiles, I
proclaimed that they should repent and turn to God, doing works worthy of
repentance. (21) Because of these things Jews seized me when I was in the
Temple and tried to make away with me. (22) So, having obtained the help
that comes from God, I stand to this day, testifying to small and great, saying
nothing but the things which the prophets and Moses said were to happen,
(23) that the Christ was to suffer, and that he first would, on the basis of
the resurrection of the dead, proclaim light to both the people and the
Gentiles.’
(24) While Paul6 was saying these things in his defence, Festus said with a
loud voice. ‘Paul, you are mad; your great learning is leading you into
madness.’ (25) Paul said, ‘I am not mad, most excellent Festus, but uttering
words of truth and sober-mindedness. (26) For the king, to whom 1 speak
with boldness, knows about these matters, for I am persuaded that none of
them escapes him; for this was not done in a comer. (27) King Agrippa, do
you believe the prophets? I know that you believe.’ (28) Agrippa said to
Paul,7 ‘With little trouble you are trying to persuade me to play the
Christian.’ (29) Paul replied, T wish to God that, with8 little trouble or much,
you and all who hear me today would become such as I myself am—apart
from these bonds.’ (30) The king rose up, and the governor and Bernice and
those who were sitting with them. (31) When they had withdrawn they spoke
to one another, saying, ‘This man is doing nothing worthy of death or
imprisonment.’ (32) Agrippa said to Festus, ‘This man could have been
released if he had not appealed to Caesar.’

Bibliography
C. K. Barrett, as in (48).
H. Braun, ThR 29 (1963), 175.
H. J. Cadbury, JBL 48 (1929), 421f.
J. Dupont, FS Schnackenburg, 125-43.
J Dupont, Nouvelles Études 446-56 (= FS Barrett, 290-9).
A. Fridrichsen, Coniect. Neot. 3 (1939), 14f.
K. Haacker, as in (56).
6Greek, he.
7RSV, In a short time you think to make me a Christian; NEB, You think it will not
take much to make me a Christian; NJB, A little more, and your arguments would
make a Christian of me.
8RSV, whether short or long; NEB, Much or little; NJB, Little or much.
1144 COMMENTARY ON ACTS

P. Harlé, NTS 24 (1977-8), 527-33.

C. J. A. Hickling, Kremer, Actes 499-503.
S. Legasse, as in (54).
A. J. Malherbe, Paul and the Popular Philosophers (1989), 147-3.
R. F. O’Toole, The Christological Climax of Paul's Defense, AnalBib 78
(1978).
J. Μ. Reynolds, JRS 68 (1978), 111-21.
L. Schmid, TWNT 3.665-7.
W. Stegemann, as in (43).
A. Vögeli, ThZ 9 (1953), 415-38.
R. D. Witherup, as in (55).

See also other works cited in (23) (I.438f.) and in (55) (II.1030f.).

Commentary
For the meeting between Festus and Agrippa see above, in the
introduction to Section 60. Such a meeting is not to be dismissed as
impossible or even unlikely, but it is hard to see what trustworthy
information about it Luke can have found. The new paragraph is
somewhat different in that it purports to relate a more public event
held in some sort of audience chamber (ἀκροατήριον, v. 23) and
attended by the retinues of Festus and of Agrippa and Bernice and a
miscellaneous and undefined company of local notables. At least it
was not, as represented by Luke, a private conversation. Paul’s long
speech, which includes the third account of his conversion, is the
main feature of the narrative; this will be discussed below. The
speech is interrupted by Festus’ shout, ‘You are mad, Paul’. A brief
conversation between Paul and Agrippa follows before the scene is
wound up by the explicit affirmation, by both rulers, that Paul is an
innocent man. The only historical question that arises and may
perhaps have an answer is whether a Christian writer composing
freely would introduce into his story the assertion that Paul was mad.
The answer is probably, Yes, he might, since as a writer he is
evidently able to allow Paul to reply, and even to have the last word.
In fact he presents here what he takes to be Paul’s reply to a charge
brought against Christians in his own (Luke’s) time, at the same time
constructing a climax in his picture of the heroic Paul. The picture
begins with the third account of Paul’s conversion (cf. 9.1-19;
22.4-21). The second and third accounts differ from the first partly
through Luke’s liking for variation but partly also through adaptation
to the settings in which they are delivered. The second was addressed
to a Jewish crowd in the Temple (see p. 1031). The third is spoken to
Gentiles. Thus (for details see the notes below) Paul says that the
voice from heaven addressed him as Saul, and adds the explanation
 61. FESTUS, AGRIPPA, AND PAUL. 25.23-26.32. 1145

that the voice spoke in Hebrew (Aramaic). The voice asks as in the
other accounts, ‘Why are you persecuting me?’ But immediately the
Greek proverb is added, 'It is hard for you to kick against the goad’.
There will be no need after this for dispute, and the story goes on
immediately to Paul’s commission to go to the Gentiles, which is not
deferred (as in ch. 22) to a later occasion in the Temple. Paul the Jew
is to evangelize the Gentiles, and is assured of God’s protection as he
fulfils his mission. To this vocation Paul had not been untrue; this
was why the Jews hated him so bitterly and had tried to kill him,
though, as he goes on to point out, his Gospel simply asserts the
fulfilment of what had already been written by Moses and the
prophets, and was addressed to all (τῷ τε λαῷ καὶ τοῖς ἔθνεσιν, ν.
23). Festus’s interjection gives him the opportunity of asserting his
sober truthfulness and dependence on the prophets, and of uttering as
his last word the declaration that makes him, beyond any other, the
model of a Christian preacher (v. 29).
Haenchen (662) correctly sums up the effect of the passage. ‘Der
Leser soll nur den Eindruck mitnehmen: Paulus war ohne alle
Schuld, obwohl er nicht frei kam!’ This is undoubtedly the effect that
Luke intended that the passage should have, and he wrote so as to
achieve it. Yet his narrative is not entirely fictional. At the end of the
proceedings in Palestine Paul was not free, and the only reason why
he was not free was that Roman officials thought that there was at
least enough to be said for him to make it worth while to take the
case to a higher court. That Paul’s Gospel was true was of course for
Luke not in question. And the conclusion was one with which Paul
himself was probably not displeased. He was as well protected from
the Jews as he could hope to be, and would soon be out of reach of
Palestinian Judaism, enjoying a free passage to Rome, the place he
had made his objective.

23. τῇ οὖν ἐπαύριον takes up αὔριον in 25.22. ἐλθόντος, the

singular participle, is followed by the double subject τοῦ Άγρίππα
καὶ τῆς Βερνίκης; a very natural construction. Bernice is here
something of an afterthought; when later in the verse brother and
sister are both clearly in mind the plural participle εἰσελθόντων (P74
pc have ἐλθ.) is used.
φαντασία in the sense of parade, ostentation, show (as distinct
from literary and philosophical meanings) is cited by LS 1916 only
from relatively late authors. MM 664 have no papyrus evidence to
quote.
ἀκροατήριον is a place designed, or used, for the purpose of
hearing; often a lecture room, here perhaps audience chamber? For
the ending –τήριον see Μ. 2.342 and BDR § 109.8, n. 10. It is by no
means necessarily a law court, and the use of the word suggests, or at
least is consistent with, an informal hearing; this indeed it must have
1146 COMMENTARY ON ACTS

been. Once the appeal to the Emperor’s court had been made and
allowed no lower court had any right to try Paul. The main
impression is that of a show put on to gratify Agrippa, though it is
doubtless true that Festus would be glad to gain additional informa-
tion that would help him to write a more adequate account of the case
he was sending to Rome. Begs. 4.312: ‘Festus was merely showing
off an interesting prisoner to entertain Agrippa, the chief local
dignitaries, and the officers of the Mess.’ Begs. 4.313: ‘... to obtain
material for a report on a case which puzzled him’.
σύν τε χιλιάρχοις. τε might have been expected to follow the noun
χιλ. which is coordinated with ἀνδράσιν, but attraction to follow the
preposition is classical (BDR § 444.5, n. 6).
τῆς πόλεως, presumably Caesarea.
κελεύσαντος. The verse is based on a complicated pattern of
absolute genitives—έλθόντος, εἰσελθόντων, κελεύσαντος. After
these the last words, ἤχθη ὁ Παύλος, stand out with impressive
simplicity. This perhaps supports the view of Ehrhardt (Acts 120f.),
against Cadbury (History 43): Luke is not in this elaborate picture
indulging in a kind of snobbery.
The Syriac Harclean margin contains (the equivalent of) the words
τοῖς καταβεβηκόσιν ἀπό τῆς ἐπαρχείας; Ropes (Begs. 3.233)
observes that it is not clear in the MS (which lacks the usual
diacritical marks) whether this is an addition or replaces τοῖς κατ’
ἐξοχήν τῆς πόλεως. Clark (381) disagrees with Ropes’s description
of the words as a gloss.
Stählin (305): from this point onwards and throughout ch. 26 there
is a parallel between Jesus and Paul, with the difference, ‘dass Jesus
die ganze Szene mit seinem Schweigen, Paulus dagegen mit seiner
Rede beherrscht.’ It is a quite considerable difference. ‘Die Szene vor
Agrippa und Bemike ist eine Dublette derjenigen vor Felix und
Drusilla’ (Lüdemann 262; cf. 25.1,6). But the counterpart to Felix is
Festus.

24. θεωρείτε, indicative. So Bengel (480), rightly.

ἐνέτυχον. The plural verb follows properly upon the noun of
multitude πλήθος, but Β Η Ψ 104 945 al thought that the singular
(-εν) would be more appropriate, ἐντυγχάνειν means to approach
someone with a request, often in favour of a person (e.g. Rom. 8.27),
hence to intercede; here, against.
πλήθος followed by a national name is often not multitude, but
'people in the official political sense’. So Deissmann (BS 232f.),
citing 1 Macc. 8.20 and inscriptions (IMAe 85.4 and 90.7, both τὸ
πλῆθος τὸ 'Ροδίων). It is suggested that the move against Paul was
not an occasional riot but a move on the part of the Jewish people as
a whole. That (as Deissmann points out) πλήθος was used also of
religious communities is in the light of the nature of Judaism
 61. FESTUS, AGRIPPA, AND PAUL. 25.23-26.32. 1147

consistent with this. For the position of τε (εν τε) see the note on v.
23.
βοώντες (Ρ74 A B 81 pc) is strengthened to ἐπιβ. by C Ε Ψ
<a>).
μήκετι at the end of the sentence emphatically doubles the
negation.
Ropes (Begs. 3.233) translates the Harclean Margin of vv. 24, 25,
26 as follows: et in Hierosolymis et hic, ut traderem eum iis ad
tormentum sine defensione. non potui autem tradere eum, propter
mandata quae habemus ab Augusto. si autem quis eum accusaturus
esset, dicebam ut sequeretur me in Caesaream, ubi custodiebatur: qui
quum venissent, clamaverunt ut tolleretur e vita. quum autem hanc et
alteram partem audivissem, comperi quod in nullo reus esset mortis.
quum autem dixerem: Vis iudicari cum iis in Hierosolyma? Cae-
sarem appelavit. de quo aliquid certum scribere domino meo non
habeo. This is given in Greek στίχοι by Clark (381f.).
25. This verse contains the strongest assertion so far of Paul’s
innocence. The Roman magistrate has found in him nothing worthy
of death, which evidently was the penalty that his Jewish accusers
sought. μηδὲν ... πεπραχέναι: πράσσειν can be used in a general
sense but it is used ‘meistens von Handlungen, die nicht löblich sind’
(BA 1399f., with many examples from the NT and elsewhere).
Many copyists ( * Ψ <a> gig s syh) changed the finite verb
κατελαβόμην into the participle καταλαβόμενος and thus produced
the neat sentence, I having apprehended ... and he having appealed
... I decided to send ...
For the aorist participle ἐπικαλεσαμένου, P74 2464 pc have the
present, ἐπικαλουμένου, which makes less good sense, unless the
intention is to stress that Paul continued to make his appeal, did not
withdraw it; but once the machinery was set in motion he would have
little choice about this.
For ἐπικαλεῖσθαι see 25.11; for τὸν Σεβαστόν, 25.21.
26. To decide to send Paul to Rome was one thing; to know what
to say in the accompanying report was another. For the duty of
sending a report see Digest 49.6.1: Post appellationem interpositam
litterae dandae sunt ab eo, a quo appellatum est, ad eum, qui de
appellatione cogniturus est, sive principem sive alium; quas litteras
dimissorias sive apostolos appellant. If Festus was convinced that
Paul was innocent he could have released him at once; he may
however have recognized that it was in Paul’s interests to be packed
safely off to Rome. ‘Die Sprache (ασφαλής, προάγω, ἐφ’ὑμῶν,
ἀνάκρισις) ist wieder juristisch-technisch’ (Conzelmann 137).
τω κυρίῳ, for the Emperor; cf. ὁ Σεβαστός in v.25. ‘ “ὁ κύριος”
ist hier der Kaiser. Ein langes Kapitel Religionsgeschichte und ein
langes Kapitel Kirchengeschichte klingt an’ (Bauemfeind 267).
1148 COMMENTARY ON ACTS

Bengel (480) is right with ‘Nuper orta erat haec domini appellation
but more detail is needed. ‘V. 26 enthält den ältesten uns bekannten
Beleg für die Anwendung des absolut gebrauchten Titels “Herr”
(griech. kyrios) auf den Kaiser’ (Roloff 350). Yet Luke could use the
word without explanation and assume that his reader would know
what was meant. For the use of dominus/κύριoς in the first century see
W. Foerster in TWNT4.1053-6; MM 365 (‘There is no evidence that
this title was applied to the Roman Emperor in the West before the
time of Domitian. Indeed it was specifically disclaimed by Augustus
and Tiberius as contrary to the Roman conception of the “Princi-
pate” ’); BA 933 (‘Seit Claudius... finden wir auch die röm. Kaiser in
steigendem Masse so bezeichnet—ganz vereinzelt schon früher’).
Among the most important references are Ovid, Fasti 2.143 (Tu
[Romulus] domini nomen, principis ille [Augustus] tenet); Suetonius,
Augustus 53 (Domini appelationem, ut maledictum et opprobrium,
semper exhorruit); Tiberius 27; Tacitus, Annals 2.87; POxy 1.37. 1.6;
Dittenberger, OGIS 2.606.1; Sylloge 2.814.30.
προάγειν, though it can be used generally, is a semi-technical term
for producing a prisoner in court; so e.g. Josephus, War 1.539 (τούς
γε μὴν υἱοὺς ού προήγαγεν εις τὴν δίκην); Ant. 16.393; Justin, 1
Apology 21.3—cited by BA 1406.
ὅπως. The pretext for producing the prisoner was to satisfy
Agrippa’s curiosity, but Festus does not conceal his intention of
using the occasion, and Agrippa’s knowledge of Judaism, to help
him in writing his report. This is at least as likely as the opposite
view, that the report was the pretext for the entertainment of Agrippa.
After this point we hear no more of a letter to the Emperor (Weiser
650). Luke knew that one was required but had not seen, and did not
need to invent, what Festus wrote.
σχῶ is changed into the more common ἔχω by P74 A Ε Ψ 81 614
945 1891 2495 al. The indirect question τί γράψω becomes τί
γράψαι in E <a>.
27. άλογον γάρ μοι δοκεῖ. No doubt it would seem equally
unreasonable to the Emperor. After μοι δοκεῖ the dative πέμποντι
would have been better that the accusative.
1. ἐπιτρέπεται, an aoristic present (BDR § 320, n. 2; Μ. 3.64; cf.
Acts 9.34): It is hereby permitted you to speak. An impersonal
passive, which gives a formal touch to the proceedings.
περί (P74 A C E 33 36 81 453 614 945 1739 1891 2495 al) and
υπέρ (Β ( Ψ ) <a>) probably do not differ greatly from each other: to
speak for yourself. But περί ought strictly to mean about yourself
that is, to give an account of your beliefs and behaviour. This may be
intended. ‘Paulus redet zwar “über sich selbst”, kommt dabei aber
am Ende auf Jesus Christus zu sprechen’ (Schneider 2.370).
ἐκτείνας. Cf. the use of κατασείειν at 12.17; 13.16; 21.40.
 61. FESTUS, AGRIPPA, AND PAUL. 25.23-26.32. 1149

ἐκτείνειν is often used for the stretching out of the hand for other
purposes (e.g. for healing) but it is not used in the sense of an orator’s
gesture elsewhere in the NT. LS 521 cites no example, nor does BA
495. BA mentions Quintilian 9.3.84ff. (an error for 11.3.84ff.—
Narratio magis prolatam manum, amictum recidentem, gestum dis-
tinctum ... postulabit) and Apuleius 2.21: porrigit dexteram, et ad
instar oratorum conformat articulum; duobusque infinis conclusis
digitis, caeteros eminentes porrigens ...
ἀπελόγειτο. See 19.33. ‘... although not formally on trial, the
word shows that the Apostle was defending himself’ (Knowling
500). This, at least, is the way Luke understands the scene. The tense
of the verb is an example of the imperfect (a sort of inceptive
imperfect) used for a word of speaking that introduces direct speech
(especially extended speech).
Here (cf. 25.24-26) the Syriac Harclean margin has, before
έκτείνας, an addition of its own: ‘tunc ipse Paulus, confidens et in
spiritu sancto consolatus, extendit manum’ (so Ropes, Begs. 3.233)
another example of the Western tendency to enliven the narrative.

2. περὶ πάντων ὧν ἐγκαλούμαι. Various constructions occur with

ἐγκαλεῖν; for the use of περί BA 434 compares PHib 96.22, περί ὧν
ἐνεκάλεσαν ἀλλήλοις. Cf. BDR § 178, n. 2.
’Ιουδαίων, anarthrous; it is customary with Greek orators not to
use the article with the names of opponents (BDR § 262.2, n. 5).
Delebecque (124) explains differently: ‘Paul ne dit pas “par les
Juifs”; de même au v. 3. Dans le premier cas Paul suggère que tous
les Juifs ne l’accusent pas; dans le second il s’incorpore au peuple
juif.’ Delebecque translates, par des Juifs, chez des Juifs. For ὑπό,
P74 945 1739 1891 pc have παρά: the accusation rose from the
Jews.
βασιλεύ Άγρίππα. It is usual for the vocative to stand earlier in
the sentence than it does here (BDR § 474. 6a, n. 10). ‘Magnam vim
habet compellatio in secunda persona, singulari praesertim, et pro-
prio nomine’ (Bengel 480).
ἥγημαι, perfect with present sense; this is the classical use (BDR §
341, n. 3), a literary touch (Μ. 1.148). Contrast Phil. 3.7, where Paul
uses the perfect with perfect meaning. It is of course arguable that for
an occasion such as this Paul would put on (or attempt to put on—
see below and the next verse) a higher style than he would naturally
use. Cf. Blass (264), who refers to Herodotus 1.126.6; Plato, Timaeus
19e.
After ἐμαυτὸν μακάριον, μέλλων (nominative) is harsh. It is
defensible, but would be less harsh if the order of the words were
changed: μέλλων ... ἥγημαι ἐμαυτόν μακάριον ... See Zerwick (§
15).
ἀπολογεῖσθαι: see v. 1.
1150 COMMENTARY ON ACTS

3. The attempt at literary style (if such it was) in v. 2 now breaks

down completely. The accusative and participle (γνώστην ὄντα σε)
has no verb to depend on. (P74) A C 33 (36) 614 945 1891 al syp
save the situation by adding ἐπιστάμενος after ζητημάτων, but this
can hardly (notwithstanding Dibelius 92) be Luke’s text. The
accusative γνώστην after σου (v. 2) is hard (Bauemfeind 269;
Moule, IB 37). BDR § 137.3, n. 3 suggest that the accusatives may be
dependent on ἥγημαι (v. 2); better is Page (245): ‘The acc. is
governed by the sense of “thinking” or “considering”, which is the
main idea of the sentence.’ This does not justify the grammar of the
defective sentence but explains how the sentence came to be
defective in the way it is.
‘μάλιστα usually precedes that which it emphasizes’ (Begs.
4.314). So it is here.
ἐθῶν τε και ζητημάτων. ἐθός is probably used here in the sense of
law; cf. 15.1. It could have been chosen as less offensive in a court
that knew only one (Roman) νόμος. The ζητήματα (cf. 25.19) are
the disputes which the Jews were known to have about the inter-
pretation of their laws. So essentially Calvin (2.269).
κατά ’Ιουδαίους. The use of κατά to suggest a possessive or
subjective genitive is ‘allgemein hellenistisch’ (BDR § 224.1, n. 3).
But the genitive equivalent will be different for the two nouns: the
customs (laws) of the Jews; the disputes current among them, in
which they engage.
δέομαι... ακοῦσαι. ‘Im Genitiv steht die angeredete Person bei
δεῖσθαι “bitten”; dazu tritt der blosse Inf. wenn der Gebetene Subj.
des Inf. ist’ (BDR § 409.5). Cf. Lk. 9.38. After δέομαι the genitive
σου is added by C <a> syp co?.
In vv. 2, 3 Paul is using the customary captatio benevolentiae; cf.
24.2f., 10. Beyer (148) misses the point with the observation that
Paul is relieved to speak to a Jew after making explanations to
ignorant Gentiles; he fared considerably better with Gentiles than
with Jews.

4. μὲν συν introduces the serious business of the defence.

Rather oddly, βίωσις appears to be, in use, a Jewish word. It is
used in Sirach, Prol. 14 (τῆς ἐννόμου βιώσεως), in Symmachus Ps.
38(39).6, and in a Jewish inscription of the first century AD (IG Rom.
4.655.12f., διά τε τὴν ἐνάρετον αυτών (βί]ωσιν). ΒΑ 283 surpris-
ingly misses the use in Symmachus and adds nothing more. LS 316
add a 6th century AD papyrus. That the word means manner of life is
clear.
ἐν τφ ἔθνει μου εν τε Τεροσολύμοις ... πάντες ’Ιουδαῖοι. Turner
(Insights 84f.) discusses the meaning and construction. Much turns
on the meaning here of τε. It is natural at first to suppose that ἔθνος
here means what it means on Paul’s lips at 24.17; 28.19, that is, the
 61. FESTUS, AGRIPPA, AND PAUL. 25.23-26.32. 1151

Jewish nation. Turner argues however that τε means and, not

including or actually. Thus life in my nation and life in Jerusalem are
distinguished. Paul ‘is referring not to his present position but to the
situation of childhood days before he went for his education to
Jerusalem. Before that, he would naturally have thought of “his
nation" as being the people of the home town of which he was so
proud, Tarsus, “no mean city’’ ’ (85). The most obvious difficulty in
this view is that it is hard to see how all the Jews (even allowing for
some rhetorical exaggeration) can have known Paul’s manner of life
in Tarsus—the Jews of Tarsus, perhaps even the Jews of Cilicia, but
in the present scene Paul must be including the Jerusalem Jews who
were accusing him. Turner excludes too readily meanings of τε other
than the simple and, see e.g. Acts 6.7, where the general increase in
the number of disciples includes the adherence to the faith of a large
number of priests. In the same way it is possible that in Jerusalem
should be included within the wider in my nation. This verse cannot
settle the question of the main centre of Paul’s education, discussed
at 22.3; in any case the question is bound to remain whether 22.3 and
26.4, whatever they mean, are correct. But it would be wrong to
conclude that this verse refers to Paul’s early youth in Tarsus as well
as to his education in Jerusalem. See Bruce (2.462); also Hanson
(237). If in fact the reference is to Jerusalem ἐξ ἀρχῆς is significant:
his life had been lived if not from birth at least from very early years
in Jerusalem. Begs. 4.315 takes ἔθνος to be contrasted with Jerusa-
lem, unless possibly it means province (i.e. Judaea). Phil. 3.5 shows
that Paul’s very early life was lived among Jews.
’Ιουδαῖοι, anarthrous, is read by P74 B C* Ε Ψ 33 81 323 614 945
1175 1739 2495 al. oἱ is added by A C2 <a>. See Delebecque on v.
2. See also Μ. 3.169 and Metzger (495).
ἴσασι. This seems to be in the NT the only absolutely certain (i.e.
unaffected by textual variants) example of the older forms of the
perfect (Μ. 2.221; see also BDR § 99.2, n. 2)—another example of
the literary Greek Luke puts into this speech. The third person plural
of οἶδα does not occur in the (13) Pauline letters; οἴδατε is common
(ἴστε imperative occurs at Eph. 5.5).
In order to explain the real issue Paul must ‘von der Geschichte
seines Lebens reden. Nur so trifft er den Kem der Sache’ (Beyer
149).

5. προ- in προγινώσκοντες and άνωθεν virtually reduplicate

each other; they have known me in advance, that is, before now; they
have known me from of old, from the beginning.
ἀκριβεστάτην is one of the three superlatives in –τατος found in
the NT (four if ἀπλούστατος is read with D in Mt. 10.16); see BDR
§ 60.1, n. 2. It is another mark of literary style in this passage. ND
1.37 notes an adverbial use of the superlative in an inscription from
1152 COMMENTARY ON ACTS

Sagalassos (JRS 66 (1976), 106-31. Josephus, Life 191, where the

Pharisees are said τών άλλων ακρίβεια διαφέρειν.
αἵρεσιν. See on 5.17, and FS Lohse, 96-110; also Fitzmyer
(Essays 276), referring to Josephus, Ant. 13.171; 20.199; Life 10,
12.
θρησκείας. The word occurs seldom in the NT, of the Christian
religion only at James 1.27 (contrast 1.26; Col. 2.18). Its most
common use suggests the cultic, ritual, and possibly formal side of
religion (see 4 Macc. 5.7); hence perhaps the choice of it here, where
Judaism and especially Pharisaism are set in an unfavourable light. In
addition to the dictionaries see Hatch (Essays 55-7); Ropes on James
1.27 (ICC, 182f.).
The aorist ἔζησα is constative, looking back on Paul’s pre-
Christian life and viewing it as a whole: the life I lived up to that
point was that of a Pharisee. On the Pharisees and Pharisaism see on
23.6. Cf. Phil. 3.5f. On the sense in which Paul as a Christian
continued to be a Pharisee see Schmithals (225).
6. καί νυν, in contrast with my former manner of life as a Pharisee;
yet also in continuity with it, for Paul continues to maintain that his
Christianity is true Judaism, and that he is maintaining the hope
which is the raison d’être of his people’s existence.
εις is a surprising preposition, not unnaturally ‘corrected’ by C <a>
to πρός. The omission of ἡμῶν by H L P 049 1241 <a> seems natural
in a Christian speech, but only superficially so: the promise was
spoken to the patriarchs, and Christians claimed that they were more
truly our ancestors than the Jews'. This is very characteristic of
Luke’s thought; Paul also speaks, for example, of Abraham as our
father (Rom. 4.1).
For ὑπό with an intransitive verb see LS 1873f. (s.v. A II 1).
ἕστηκα κρινόμενος, emphatically, and effectively, at the end of
the sentence. For this hope, which you might suppose would win
respect and even fame within Judaism, I am standing trial. This
pattern—description of the hope, followed by Paul’s position as a
man on trial—is repeated in the next verse. Strictly speaking Paul
was not now on trial, but it must have seemed like it, and it was
natural for Luke to describe it so.
7. εἰς ἥν will refer to ἐπαγγελία, not to ελπίς, since the Jews hope
to attain it. When Paul wishes to take up ἐλπίς with a relative he now
has to repeat the substantive, περί ἧς ἐλπίδος.
τό δωδεκάφυλον ἡμῶν, our twelve-tribe unit. For the use of the
neuter singular of an adjective as collective noun see BDR § 263.3,
n. 6. The adjective is used at Sibylline Oracles 2.171; 3.249,
the substantivized form at 1 Clement 55.6 (cf. 31.4 for
δωδεκάσκητρον). At Herodotus 5.66.2 there are the analogous
τετράφυλος and δεκάφυλος.
 61. FESTUS, AGRIPPA, AND PAUL. 25.23-26.32. 1153

ἐν ἐκτενείᾳ. The moral sense is mainly but not exclusively biblical

(Judith 4.9; 2 Macc. 14.38; 3 Macc. 6.41); but see Deissmann BS 262
and MM 198 for inscriptions and papyri. Cf. 1 Clement 33.1; 37.1;
also Cicero, Ad Atticum 10.17.1 (quam in me incredibilem
ἐκτενείαν!). Deissmann gives the meaning endurance so also BA
495 (Beharrlichkeit). But LS 521, zeal, assiduousness, ‘gush’,
‘empressement’ seem better.
νύκτα καὶ ἡμέραν, Blass (265) thinks of the tamid offering, but
the expression simply means continuously. With λατρεύειν, τω θεῷ
must be understood. In classical use ἐλπίζειν sometimes takes the
future infinitive; hence the reading of B pc, καταντήσειν; but the
aorist infinitive is also used, and καταντῆσαι is not wrong by
classical standards.
For the end of this verse see v. 6; the same emphasis is repeated. It
is concerning this Jewish hope that I am accused by Jews. There is of
course an obvious answer to this complaint. Paul is not being blamed
for maintaining the Jewish hope but for holding, falsely in Jewish
opinion, that it has been fulfilled by Jesus.
For the position of the vocative βασιλεύ cf. v. 2, and see BDR §
474.6a, n. 10.
P29 (= POxy 13.1597.1-5) has, using Grenfell’s conjectures
(Begs. 3.235): τό δωδεκ[άφυλον ἡμῶν ἐν ἐκτε]νία νύκτ[α καί
ἡμέραν λατρεύει ἐν(?)] ἐλπίδι κ[αταντήσαι. περί ἧς νῦν(?)]
ένκαλού[μαι ὐπό ’Ιουδαίων εἰ(?)] ὁ ΘC νεκρ[ούς έγείρει]
On this Clark (382) writes: ‘The editors [Grenfell and Hunt]
remark that “the omission of a line containing τι... υμιν is an easy
hypothesis. I may point out that the words would form an excellent
στίχος).’ We must regard this as an example of the freedom—and
carelessness—with which Western editors handled the text. ‘This
papyrus and PMich 1571 raise the question whether the text of Acts
in the third century did not vary from any extant authority even more
than we had supposed. These two papyri are the oldest extant
manuscripts of Acts, they have the “wildest” Western text known,
yet they come not from Africa or Edessa but from Egypt’ (Begs.
4.316).

8. Black (AA 59) notes the asyndeton of this verse, but it is not to
be ascribed to Semitic influence; there is no grammatical connection
with what precedes but there is a connection of thought. The thought
of resurrection arises out of the reference to the hope of Israel in v. 7,
but it is not in fact the most natural expression of the hope; the hope
of the twelve tribes was that they would live unitedly in peace and
prosperity in the land promised them. Eb. Nestle suggested that this
verse is misplaced and should stand between v. 22 and v. 23; cf.
Stählin 306, 307. who thinks that v. 23 should follow v. 8 either at
this point or at the end of the speech. The suggestion is attractive
1154 COMMENTARY ON ACTS

linguistically (the word εἰ in v. 23 would follow acceptably after εἰ in

this verse) but is not necessary. For Luke (and Paul) the hope was
concentrated and fulfilled in resurrection, especially in the resurrec-
tion of Jesus; the transition would not be convincing to a Jew, but to
Luke and Paul it was. For the connection with what follows see on
the next verse. It is of course true (Roloff 351) that the Christian hope
of resurrection has a different basis from the Pharisaic.
For ἄπιστον κρίνεται cf. Josephus, Ant. 18.76, χρὴ άπιστα αυτά
κρίνειν; for παρά, ‘nach dem Urteil jmds.’, BDR § 238.2, n. 3. Cf.
Bengel (481): incredibilia veteres dixere fabulas poëticas ... sic
incredibile duxit resurrectionem Festus.
ὑμῖν is surprising here, especially after the singular βασιλεύ in v.
7. Luke probably thinks of the Christian on trial as addressing Jewish
hearers, who might be expected to share Paul’s belief in resurrection.
If pressed he might have replied that Agrippa was well acquainted
with Judaism (v. 3) and that Paul could claim that he believed the
prophets (v. 27). For the dative cf. Mk 10.27 and parallels.
For the use of εἰ where ὅτι might have been expected see H. J.
Cadbury (JBL 48 (1929), 421f.). There may be a little more to say.
Thus Delebecque (125): ‘Dans le meilleur grec beaucoup de com-
plétives peuvent être introduites par εἰ signifiant la condition, aussi
bien que par ὅτι signifiant la cause’; Zerwick (§ 404): εἰ is ‘fere
aequivalens ὅτι post verba quae exprimunt animi commotionem
(admirationem, indignationem), e.g. Mc 15.44 ... A 26.8 ‘Quid
incredibile iudicatur εἰ ό Θεός νεκρούς εγείρει: Deum mortuos
resuscitare.’ Haenchen (653) translates ‘wenn tatsächlich’.

9. μὲν οὖν (cf. ν. 4) takes the speech a stage further. Page (246f.)
says, rightly, that it resumes the narrative from v. 5 (vv. 6-8 being a
parenthesis). He notes that others take ἐγὼ μ. o. as answering the
question of v. 8, but answers convincingly, (1) such a use of μὲν σὖν
requires justification, (2) the words do not answer the question, and
(3) the question (of v. 8) is rhetorical and needs no answer. Paul can
assert the theme of resurrection in its Christian form (i.e., the
beginning of the general resurrection, Christ having been raised as
the firstfruits of all who sleep) for two reasons, both of which he is
able to cover by means of one narrative, on which he now embarks:
(a) he has himself seen the crucified Jesus alive; (b) this event has
transformed him from persecutor to apostle. The story of his
conversion (cf. 9.1-19; 22.4-21) opens with an account of his
enthusiastic resistance to Christianity.
ἔδοξα ἐμαυτφ is a lapse, though not a serious one, from classical
usage, which would have had μοι (BDR § 283.1, n. 3; Μ. 3.42). See
however Aristophanes, Wasps 1265 (1257), πολλάκις δὴ ’δοξ’
ἐμαυτῷ δεξιός πεφυκέναι; Demosthenes 18.255 (312), ὡς ἐμαυτῷ
δοκῶ (there is a misprint in Blass 266 and BA 406).
 61. FESTUS, AGRIPPA, AND PAUL. 25.23-26.32. 1155

πρὸς τό όνομα. With εναντία the dative would be more usual; e.g.
Demosthenes 18.213 (299), ἐναντί’ ἐπράξατε Θηβαίοις. For the
significance of the name in Acts see on 3.6; positively, Christian
things are done in the name of Jesus; correspondingly, resistance is
offered to his name; ‘erst hier wird also wieder (vgl. 4.17f.; 5.28,
40f.; 9.15f.) ganz deutlich, dass es der Name Jesu “des Nazoräers”
... ist, an dem die Wege der Juden und Christen total auseinanderge-
hen’ (Stählin 307). ‘Der “Name” Jesu umschreibt die Jesus zuge-
hörige Sphäre, d.h. nicht nur ihn selbst, sondern auch die an ihn
Glaubenden’ (Roloff 352).
For Ναζωραίος see on 2.22, also Rüger in the note on 22.8.

10. ὃ καί ἐποίησα ἐν Ίεροσολύμοις. See 8.3.

καί πολλούς τε. It is impossible to see any purpose that might be
served by τε in this sentence; the word has become a mannerism of
Luke’s (Moule, IB 197). For τε, 36 453 pc have δέ; Β Ψ <a> omit—
understandably.
τὴν ἐξουοίαν ... λαβών. See 9.1f., where the authority is
expressed in letters and is related to Paul’s journey to Damascus.
ἀναιρεῖν is a Lucan word (twice in Lk.; 19 times in Acts; 3 times
in the rest of the NT). While the witnesses were actually being killed
(ἀναιρουμένων, present participle) voting, one way or the other,
would be too late, but one knows what Luke means. Zerwick (§ 274)
comments, ‘cum ageretur de eis morte plectendis, detuli sententiam’.
ψήφος is a pebble, here the pebble used (at Athens and elsewhere)in
voting; hence the vote itself, ψῆφον φέρειν is to cast one's vote;
καταφέρειν here means to cast it unfavourably, that is, to vote for the
death penalty against Christians. Jeremias (Jerusalem 255, n. 34)
observes that this statement proves (if we accept its accuracy) that
Saul was an ‘ordained scribe’. Others express some caution about
this. ‘Lk würde diesen v. nicht so geprägt haben, wenn er nicht
gewusst hätte, dass Paulus Rabbi im eigentlichen Sinne gewesen ist,
weitere Folgerungen aber sind kaum statthaft’ (Bauemfeind 269).
Bruce (1.443) remarks that the statement ‘cannot be said to prove'
anything since the phrase may be used officially or unofficially.
Other inferences, interesting if not wholly convincing, are that the
Council followed correct legal procedures in acting against Chris-
tians (Stählin 307), and that the Sanhedrin had the power to inflict the
death penalty; cf. 25.16, and Jn 18.31.
The verse seems to imply that a number of Christians were put to
death. Acts has mentioned Stephen and James (with the death of the
latter Paul apparently had nothing to do); no more. Doubt has been
cast on the implication. 'On the whole, therefore, it seems best to
take Paul’s statement as being somewhat rhetorical’ (Marshall 393).
This may be true, but the grounds for such a negative conclusion
should be examined. ‘Hier wird im Gegensatz zu 8.1 eine Reihe von
1156 COMMENTARY ON ACTS

Hinrichtungen von Christen behauptet’ (Haenchen 654). But 8.1

though it names only one martyr does not say that only one died. Cf.
Jn 16.2; Mt. 24.9; Mk 13.12. Luke was writing many years after
Saul’s work as a persecutor, but he probably represents at least folk
memory of that period.
11. κατὰ πάσας τὰς συναγωγάς. There is in chs. 8, 9, 22 no
reference to synagogues in Jerusalem; at 9.2 we have synagogues in
Damascus. It is implied that Christians still frequented synagogues;
that is, they wished to continue to be Jews though now they had
become Christians. This was true also (at a later date) of Paul
himself, who for this reason received five synagogue beatings (2 Cor.
11.24). Such punishment might be implied by τιμωρῶν, to exact
vengeance from, hence to punish (more frequently in the middle);
e.g. Josephus, Ant. 2.107 (τιμωρούντος αυτούς τού θεού); Sopho-
cles, Oedipus Rex 107.
The imperfect ἠνάγκαζον is often (e.g. Μ. 1.128f.; Turner,
Insights 86f.; Moule, IB 9 (‘may be’)) taken to be conative. BDR §
326, n. 1 are probably right in taking it to be both conative and, like
ἐδίωκον, which cannot be conative, repetitive. Cf. Blass (266). We
have no right to suppose that Saul was never successful in compel-
ling Christians to blaspheme—presumably to utter what would be
blasphemy from a specifically Christian point of view, perhaps
cursing Christ (Roloff 352). Cf 1 Cor. 12.3 (ανάθεμα 'Ιησού); Pliny,
Epistles 10.96.5 (maledicere Christo). Josephus, War 2.152, records
that the Essenes would not yield to torture intended ἵν’ ἢ
βλασφημήσωσιν τόν νομοθετήν ἢ φάγωσίν τι ασυνηθών.
περισσῶς τε ἐμμαινόμενος. There is no difficulty here in τε (=
and); contrast v. 10. περισσῶς according to BDR § 60.3, n. 5 is no
more than ‘sehr', but there is no reason why the extraordinarily,
exceedingly of LS 1387 should be avoided, ἐμμαίνεσθαι is to be
furiously (madly) angry with someone; so Josephus, Ant. 17.174,
ἐμμαινομένου πᾶσι τού βασιλέως, where King Herod’s rage led him
to plot a frightful massacre. Paul had in the past been mad (with
rage); he was sane now (v. 25).
τὰς έξω πόλεις. 9.2; 22.5 mention only Damascus.
12. ἐν οἷς: Moule (IB 131, 197) does not decide between and so
and in the course of which activity. The former is perhaps better. Cf.
9.1; 22.6.
On ἐξουσίας καί επιτροπής τής τών αρχιερέων Μ. 3.218 com-
ments that the last three words are (or give the impression of being)
a kind of afterthought; cf. 1.12 (ἀπὸ όρους τού καλουμένου
έλαιώνος—article with an attribute after an anarthrous noun). The
singular article probably refers to both ἐξουσίας and ἐπιτροπής;
certainly both would be derived from the chief priests as the source
of authority in Jerusalem.
 61. FESTUS, AGRIPPA, AND PAUL. 25.23-26.32. 1157

τής is omitted by P74vid A E 36 1704 pc; παρά is inserted before

τών αρχιερέων by C Ψ <a>.
According to Kosmala (340) Paul went to Damascus to persecute
the people of the (Qumran) Way who had migrated thither; see on
9.2.
On this account of Paul’s conversion (vv. 12-18) see Wilson
(Gentiles 161-7). Details of the earlier accounts have been sup-
pressed so as to make room for more details (vv. 16-18) about Paul’s
vocation (Hanson 238).

13. ἡμέρας μέσης. Cf. 22.6, περί μεσημβρίαν. There is no

corresponding note of time in ch. 9. The use of the genitive is not
classical (BDR § 186.3; 270.3, n. 5; Μ. 3.235, noting that there are
papyrus parallels). ‘Elegantius ἡμ. μεσοῦσα’ (Blass 267), but cf.
Aeschines, Epistle 1 ... περὴ μέσην ἡμέραν κατήχθημεν εις
Ναρησσὸν τήν Κείων. Cf. 16.25; 27.27, both with reference to
night.
κατά τήν οδόν. There is no precise parallel in ch. 9 or in ch. 22,
but both narratives note that Paul is travelling—presumably on the
road.
βασιλεύ. For the position of the vocative see on v. 7.
οὐρανόθεν. Both 9.3 and 22.6 have ἐκ τοῦ οὐρανοῦ. Cf. 14.17.
The word is omitted by P74.
υπέρ ... ήλιου. In 9.3 simply φῶς, in 22.6 φῶς ικανόν. For υπέρ
with the accusative as a form of comparison see BDR § 230, n. 3.
περιλάμψαν. In 9.3 περιήστραψεν, in 22.6 περιαστράψαν.
καί τούς ... πορευομένους. In the earlier narratives there is no
reference at this point to Paul’s fellow travellers.

14. πάντων ... εἰς τὴν γῆν. At 9.4; 22.7 only Paul is said to fall to
the ground. It is not however correct to say, ‘In 9.4 fällt nur Paulus
nieder’ (Haenchen 655). There is nothing in the other passages to
suggest that others did not fall.
ἤκουσα φωνήν. 9.4, ἤκουσεν φωνήν; 22.7, ἤκουσα φωνής. On
the case taken by ἀκούειν see 1.451f. and BDR §§ 173.2, n. 5;
416.1b, n. 5. In the present account the question whether the
remainder of the company heard the voice is not raised (though it is
assumed by the addition between γῆν and ἤκουσα of διά τόν φόβον
ἐγὼ μόνος, in 614 2147 pc (gig syhmg sa bomss)).
The voice is described as λέγουσαν πρός με in P74 A B C (E)
048 096 36 81 945 (1175) 1739 1891 al; this is certainly the true
reading, but H 23 al e vg have λαλοῦσαν πρός με, and Ψ <a> (gig)
have the conflate λαλοῦσαν πρός με καὶ λέγουσαν.
τῇ Έβραῖδι διαλέκτῳ: not in ch. 9 or in ch. 22. ‘The introduction
of “in the Hebrew tongue” as an apology for the barbarous name
“Saul” is more likely, to come from Luke, though it is conceivable
1158 COMMENTARY ON ACTS

that St. Paul would apologize in this way’ (Knox, Hell. El. 29).
Aramaic is probably the language intended; see 21.40.
Σ., Σ., τί με διώκεις; See 9.4; 22.7.
σκληρόν σοι πρὸς κέντρα λακτίζειν. There is no parallel in ch. 9
or in ch. 22 (in the correct text). The proverb is a Greek one (from
Pindar and the Tragedians onward) and therefore unlikely to have
presented itself to Paul’s conscious mind at the time of his conver-
sion; though Knox (ibidem) notes Ps. Sol. 16.4 (ἔνυξέν με ὡς
κέντρον ἵππου ἐπί τήν γρηγόρησιν αὐτοῦ) to show that the proverb
‘may have been acclimatized in Judaism, and such a proverb might
well have found its way into a collection of proverbs available for
Jewish students of Greek’. Bultmann rightly says that the phrase does
not refer to an inner struggle, but rather is a widespread proverbial
expression that means that man cannot withstand the divine (E&F
134). It is conflict against the gods that is in mind in the tragic poets;
here the proverb takes up the thought that Paul is resisting—
persecuting—Jesus by hindering his work. See L. Schmid in TWNT
3.665-7; also Dibelius (188-91). Commentators and others cite
many parallels; among the most important are the following:
Pindar, Pythians 2.94-96: ποτὶ κέντρον δέ τοι λακτίζεμεν
τελέθει ὀλισθηρός οἶμος.
Euripides, Bacchae 795: πρὸς κέντρα λακτίζοιμι θνητός ὢν θεῳ.
The latter is probably the most important passage. BDR § 487.1, n. 2
point out the metre, but think that there is no literary dependence. See
further below; also Iphigeneia in Tauris 1395f.; Aeschylus, Aga-
memnon 1624, πρὸς κέντρα μή λάκτιζε. In Latin see Terence,
Phormio 1.2.27f.; Plautus, Truculentes 4.741.
Jewish material is hard to find. See Ps. Sol.16.4 (above); Philo, De
Decalogo 87. Bengel (481) says, ‘Syriacum adagium notat Light-
foot' but he seems to have taken a translation into ‘Syr’ as an
illustrative quotation!
Some think that Luke knew the Bacchae of Euripides, citing in
addition to this passage 5.39 and 16.26. See Bauemfeind (269) and
Roloff (352); also A. Vögeli, ‘Lukas und Euripides’, ThZ 9 (1953),
415ff.

15. This verse agrees substantially with 9.5; 22.8. ὁ Ναζωραίος is

added by 048 6 104 614 1175 pc gig vgmss syph**, assimilating to
22.8.

16. Only in ch. 26 (not in ch. 9 or ch. 22) is the apostolic

commission given at the time of the conversion; Stählin 310 notes its
resemblance to resurrection narratives in which a commission is
given. The question of the provenance of these stories is acutely
raised. It cannot be said that the version in this chapter is due solely
to abbreviation, for it would have required only a word or two to
 61. FESTUS, AGRIPPA, AND PAUL. 25.23-26.32. 1159

introduce the commission with ‘Later, in the Temple’ or some such

prefix. Either Luke is deliberately introducing variation, so as to
achieve the emphasis that comes from repetition and at the same time
avoid the risk of losing his reader, or he is using different traditions.
The former alternative is more probable and is supported by a further
motivation supplied by Blass (267); ‘In hac autem or[atione] tota
persona Ananiae sublata est, quippe quae non esset apta apud hos
auditores; unde sequebatur ut mandata omnia ut ipsius lesu effate
inducerentur.’
αλλά, as at 9.6. ‘Sometimes before a command [αλλά] is not so
much adversative as consecutive, and is best translated as an
interjection, Well!’ (Μ. 3.330). There is however an adversative
element in the word here. T am Jesus, whom you are persecuting.
But all this is to change now. Get up and..
ἀνάστηθι καὶ στῆθι is a very unpleasing repetition, justified only
by the fact that the second verb is virtually a quotation of Ezek.
2.1-3 (στῆθι ἐπί τοὺς πόδας σου ... ἤκουον αυτού ...
ἐξαποστέλλω ἐγώ σε ...). This is one of a number of OT passages
alluded to here; Hanson (239) collects in addition Jer. 1.7; Isa. 35.5;
42.7, 16; 1 Chron. 16.35.
ὤφθην. Cf. 1 Cor. 15.8, ὤφθη κἀμοί. But note also ὀφθήσομαι at
the end of the verse; the word is not confined to resurrection
appearances.
προχειρίσασθαι; cf. 3.20. The present verse shows that no
temporal significance is to be attached to the προ- in προχειρίζειν;
the appearance is for the purpose of appointing. It is not said here that
Paul is appointed to be an apostle, and in view of the almost complete
silence of Acts regarding Paul’s apostleship (see 14.4, 14 and the
notes; also Introduction, p. lxxxix) this can hardly be accidental. Paul
is to be a υπηρέτης; cf. 13.5, also Lk. 1.2, ὑπηρέται τού λόγου. Paul
is to be such a υπηρέτης, and a witness, of what he has seen and will
see. Here there may be a contradiction with 13.31f., where the
witnesses of the resurrection, who had followed Jesus from Galilee to
Jerusalem, appear to be distinguished from Paul and his colleagues
who preach the Gospel.
Paul’s testimony will not be confined to a resurrection appearance
(this seems to be the primary sense of ὧν τε εἶδές με—με is omitted
by P74 A C2 E Ψ 096 <a> latt bo, perhaps rightly); it extends to ὧν
ὀφθήσομαι σοι. This may refer to visions such as 18.9; 23.11. In the
epistles Paul displays considerable reluctance to speak about visions
(2 Cor. 12.1-5; 2 Corinthians 34, 250, 305-13); certainly the
epistles contain few references to them. It might be possible to assert
that, in a sense, the Lord appeared to Paul in his intellectual
apprehension of the Gospel, but it is unlikely that this is what Luke
had in mind.
ὧν τε ... ὧν τε. This is the only example of co-ordinated τε ... τε
1160 COMMENTARY ON ACTS

in the NT. Zerwick (§ 466) (‘in elato stilo oratorio’) notes that it
appears where Paul is addressing Agrippa—another mark of careful
literary composition. See also BDR § 444.1, n. 1. Dibelius however
writes, ‘Utterly impossible is the text of 26.16 ... The heavenly voice
promises Paul not that Christ shall again appear to him but that much
shall be shown him. Obviously the clause was corrupted through the
influence of the preceding ὤφθη.’ The sentence is unusual rather
than impossible and corrupt. Page (248) is better: ‘ὧν ὀφθ. =
ἐκείνων ἃ ὀφθ. where ἅ is acc plural, such a use being very
common with the neuter plural of pronouns even after intransitive
verbs, and ἃ ὁρῶμαί σοι = exactly “the visions in which I am seen
by you”. Cf. Sophocles, Oedipus Tyrannus 788 ὧν μὲν ἱκόμην
ἄτιμον ἐξέπεμψεν. The passive form of the phrase is due to a desire
to bring out the agency of God.'
Two of Weiser’s comments should be noted. ‘Inhalt und For-
mulierungen sind an atl. Berufungs- und Sendungstexten sowie an
jüdisch-hellenistischer und urchristlicher Missionsterminologie ori-
entiert: “die Augen öffnen’’ (vgl. Jes. 42.7; 61.1); “von der
Finsternis zum Licht’’ vgl Test. der Zwölf Patriarchen 19.1 [Testa-
ment of Levi 19.1 is intended]; Josef und Asenet 8.9; Eph. 5.8; Kol.
1.12f.; 1 Pet. 2.9); “sich bekehren zu Gott”...’ (653). ‘Umkehr
versteht Lukas hier vor allem als das Zur-Glaubenseinsicht-Kom-
men, das durch die Verkündigung ermöglicht wird und das entspre-
chende Verhaltungsweisen zur Folge haben soll’ (654). Another
parallel to note is 1QH 11.10-14.
Schille (451) sees in this verse an ordination formula.

17. ἐξαιρούμενος. The exalted style of some of the earlier verses

(see the notes) might have suggested the future rather than the
present participle, but in fact the present participle takes its time
reference from ὀφθήσομαι in v. 16. Page (248) insists that the word
must have its classical meaning of choosing rather than delivering
(which he agrees is correct at 7.10, 34; 12.11; 23.27). Bruce (1.445)
disagrees, rightly. The determining factor is the probable allusion to
the call of Jeremiah (Jer. 1.8,... μετὰ σοῦ ἐγώ εἰμι τού ἐξαιρεῖσθαί
σε). See further Le Déaut (319) on 13.47 for the parallel between
Paul and Moses, also between Paul and the Servant of the Lord. Paul
too is to be a light of the Gentiles. This is brought out more strongly
in v. 18.
εἰς οὓς ἐγὼ αποστέλλω σε. See 22.21; there this statement
provoked the crowd to frenzy. Naturally there is no such reaction
from Agrippa and Festus. The masculine singular λαού (the People,
Israel) and the neuter plural ἐθνῶν are taken up—ad sensum and
quite naturally—in the masculine plural οὓς, which refers to the
multiplicity of persons concerned. It is however quite possible that
οὓς refers to έθνη only; it was important to make the point that the
 61. FESTUS, AGRIPPA, AND PAUL. 25.23-26.32. 1161

activity that so incensed the Jews was undertaken in obedience to a

divine command.
αποστέλλω, present, I am here and now sending you, the reading
of A B E 048vid makes the best sense and is almost certainly
<a>
correct. Ψ 096 6 81 104 614 945 1241 2495 al co have ἀποστελώ;
P74 C have ἐξαποστέλλω; 36 323 453 1175 1739 1891 2464 al have
έξαποστελώ.

18. This is an important verse in that it shows how Paul (according

to Luke) understood his mission to the Gentiles (also apparently to
the Jews to judge from v. 20; see on that verse as well as the close of
ν. 17). The language used is based on the OT (see the passages
mentioned in the note on v. 16), and had probably already come into
existence as ‘traditionelle Bekehrungsterminologie’ (Schweizer,
Beiträge 187 η. 19; also 116 η. 10; cf. Kosmala, 172 n. 28). Similar
language is found at Qumran; see especially 1QS 11.7f. (note
The image of light and darkness, and Satan and Belial,
also occurs in the Qumran literature; see however Braun, ThR 29
(1963), 175.
The language is vividly pictorial; in a sense mythological. ‘From
darkness to light’ is perhaps no more than metaphorical, and it is easy
to specify contrasts in the realms of morals and understanding which
conversion signifies, though it must be admitted that Acts provides
little by way of example. Paul is the only ‘bad man’ (in the sense of
being a persecutor, not otherwise) in Acts who becomes a good one,
and there is little in the book that suggests a new intellectual
apprehension of the meaning of life. ‘From the power of Satan to
God’ implies that men are held captive by Satan, and by the Gospel
are liberated so as to return to God their creator.
ἄφεσιν αμαρτιών: see on 2.38.
τούς ηγιασμένους are more frequently in Paul the άγιοι, but the
perfect participle passive is also used (1 Cor. 1.2; in Acts see 20.32,
with the note, including the reference to Wilcox 35-7). For κλήρος
in the sense of (allotted) portion see Acts 1.17; 8.21; Col. 1.12. There
is nothing to indicate whether the portion is among the members of
the church in the world, or in the age to come; both are intended. The
inheritance is (and will be) received by faith in Christ. As at other
points in this verse, so here no explanation is given of the relation
between faith and the gift received, or indeed between faith and the
action of God which is clearly assumed in the opening of men’s eyes,
and the act of turning them from darkness to light. Cf. 16.14, and see
Stauffer, Theologie 163 (‘Gottes übermächtiger Wille ist ein
befreiender Wille’).
The structure of the sentence is given by three infinitives, the
second and third introduced by τοῦ: ἀνοῖξαι... τοῦ ἐπιστρέψαι...
τού λαβείν. The first is clearly an infinitive of purpose, dependent on
1162 COMMENTARY ON ACTS

αποστέλλω (v.17). The second is probably dependent on the first (to

open their eyes so that they may turn ...). It is not clear whether file
third is dependent on the second or, in parallel with the second, is
dependent on the first (to turn ... so that they may receive ..., or to
open their eyes ... so that they may receive ...). The absence of any
connecting particle before τοῦ λαβεῖν suggests but by no means
proves the former alternative. It would however be mistaken to read
out of this verse a rigid sequence of elements in the process of
conversion. Luke is assembling a number of vivid images almost any
one of which would, if developed, stand for the whole. Men are
unable to see: this inability may be due to inward blindness or to
exterior darkness. However it is thought of, the situation is reversed.
Light suggests God and darkness Satan; ability to see and inability to
see suggest life under the authority of God or of Satan. Turning from
Satan to God involves pardon for the old subservience to Satan.
Turning to God means joining the ranks of those who are devoted to
him. The piling up of images constitutes an impressive, if not
perfectly clear, climax. Cf. Isa. 42.7,16; Eph. 6.12; James 1.17;
especially Col. 1.12-14. See also 1 Clement 59.2; Wilcox (73f.).
thinks of a liturgical background.
Calvin (2.278) insists that πίστει is to be taken not with
ἡγιασμένοις but with the whole series of clauses. It is probably
correct that Luke adds the reference to faith with the intention that it
shall cover all that he is saying. In v. 18b with v. 20b is ‘die Summe
der biblischen, jüdisch-christlichen Verkündigung, wie Lukas sie
versteht’ (Schmithals 227).
τυφλῶν for αὐτῶν (Ε 096 vgmss) is a not very thoughtful attempt to
improve the sentence.

19. An example of Lucan litotes; BDR § 495.2, n. 9 lists as

examples 12.18; 15.2; 19.11, 23, 24; 20.12; 21.39; 26.19, 26; 27.20;
28.2. The impression given here is of modesty on Paul’s part, but in
fact Luke seems use this mode of speech as a form of emphasis. Paul
was wholeheartedly obedient.
ὀπτασία is a Lucan word (Lk. 1.32; 24.23); also 2 Cor. 12.1. This
Pauline passage is enough to show that Paul was not incapable of
speaking of visions that he had received. See also (for the converting
and appointing vision) Gal. 1.15f. οὐκ ... απειθής itself implies that
the vision included a command. In the present passage this reflects
the ἀποστέλλω of v. 17.
ουράνιος here is an adjective of two terminations; for use
elsewhere see BA 1201.
Haenchen (656): ‘Die “himmlische Erscheinung’’ meint hier das
himmlische Wesen, das erschienen ist... Es war für Paulus unmö-
glich ..., dem himmlischen Befehl zu widerstehen. Damit ist die
christliche Mission gerechtfertigt.’
 61. FESTUS, AGRIPPA, AND PAUL. 25.23-26.32. 1163

20. In obedience to the divine commission Paul has carried out

what is intended to be understood as (at least in outline and in
principle) a universal preaching mission. The description of the
scope of the mission and of the content of its message is however
expressed in a sentence of dubious grammar. The opening phrase is
probably intended to refer to Jews; in itself, τοῖς ἐν Δαμασκῷ might
include Gentiles (a majority of the city’s population; see I.447), but
(τοῖς ἐν) 'Iεροσολύμοις must be Jews, and a contrast is needed with
τοῖς ἔθνεσιν later in the verse. For Paul’s preaching in Damascus (in
the synagogues) and in Jerusalem see 9.20-23, 28-30. The first τε
would stand better after Δαμασκῷ; it is omitted by the majority of
later MSS. See Moule (IB 169) and cf. v. 22. There follows a phrase
in the accusative, πᾶσάν τε τὴν χώραν τῆς Ίουδαίας. τε links this
with τοῖς ἔθνεσιν; one expects the dative case, and it is by no means
clear why χώραν should be in the accusative. Ε Ψ <a> lat prefix εἰς,
which is some improvement (and probably secondary). Ropes (Begs,
3.237) writes, ‘With so firmly attested a text the theory of a Semitism
suggests itself, in view of the strikingly Semitic cast and grammatical
difficulties of vv. 16-18. Cf. Deut. 1.19, ἐπορεύθημεν πάσαν τήν
ἔρημον τήν μεγάλην και τήν φοβεράν.’ But the hypothesis of a
Semitism requires much clearer support than this. Dibelius (92)
suggests that ‘the dropping-out of εις (after 'Iεροσολύμοις, and thus
haplographically identified with the ending –οις) is more credible
(see the Antiochian text)’ and this suggestion is taken up in BDR §
161.1, n. 1. See also the text of P29, quoted below. There is probably
no better hypothesis available (though it is in the nature of things
beyond proof) that Luke began to write πάσαν τήν χώραν τής
Ίουδαίας καὶ τά ἔθνη εὐηγγελισάμην, and changed course part way
through the sentence.
A difficulty of a different kind is that according to Gal. 1.22 Paul
remained personally unknown to the churches of Judaea (ταῖς
ἐκκλησίαις τῆς Ίουδαίας). At Rom. 15.19 however he claims to
have preached the Gospel from Jerusalem round as far as Illyricum.
Paul’s preaching is described by the word ἀπαγγέλλειν (16 times in
Acts). His hearers are to repent (cf. 2.38) and turn to God (cf. 3.19),
and to do works worthy of repentance, that is, they are to prove the
sincerity of their repentance by amended; lives. The accusative
πράσσοντας follows upon the dative τοῖς ἔθνεσιν; πράσσουσιν
would have been possible, but a shift of case of this kind is not
unknown (see Zerwick § 394 and cf. 15.22). The persons concerned
are addressed (which suggests the dative) but they are also the
subject of the infinitive (which suggests the accusative). For the
requirement cf. Lk. 3.8. The ‘whole phrase seems selected to show
that Paul’s preaching was not other than a Jewish missionary would
have used in attempting to convert the heathen (cf. 3.19; 9.35; 14.15;
15.19)’ (Begs. 4.320).
1164 COMMENTARY ON ACTS

The second of the two fragments of P29 (on which see on v. 7)

occurs at this point. It is said by Hanson (239) to smooth over the
difficulty mentioned, but it is probably too fragmentary to have this
or any other effect. The text, as reconstituted by Grenfell, runs:
[ἀπειθής τῇ οὐρανίω οπτασία ἀ]λλὰ τοις ἐ[ν Δαμασκῷ
πρῶτον τε καί Ίερο]σολύμοις κα[ί τή Ίουδαίᾳ καί τοῖς
ἔθνεσιν] ἐκήρυξα [... μετανοεῖν καί ἐπιστρέφειν ἐ]πὶ τόν
θεόν, [... άξια τής μετάνοιας ἔργα πρ]άοσοντας.

It will be noted that the dative τή Ίουδαία (which goes much better
than the familiar text with the dative τοῖς ἔθνεσιν) is an editorial
conjecture; the space available however would, it seems, not permit
πάσαν τε τήν χώραν τής Ίουδαίας. See Ropes (Begs. 3.235, 237).

21. ένεκα τούτων, ένεκα (over against ἕνεκεν, εἵνεκεν) is the

Attic form, and is ‘regarded by Blass [but this is not mentioned in
Blass’s commentary, or in BDR] as in keeping with a speech in the
presence of royalty’ (Μ. 2.67). Cf. however 19.32 and 28.20 (v.1.);
there is little weight in this observation. The speech is in any case an
odd mixture of style and solecism, τούτων, the things on account of
which the Jews seized Paul amount in fact to one thing, his mission
to the Gentiles which had the effect of placing Gentiles on the same
level as Israelites as potential heirs of salvation. Perhaps one should
add, as a second (though closely related) ground of attack, his
implicit claim that in doing so he was a better representative of
Judaism than his opponents, τούτου (singular) is read by P74 pc.
συλλαβόμενοι (Ρ74 Ε Ψ 33 36 81 614 945 1175 1739 1891 2495
al latt add ὄντα; A B 048 <a> do not): see 21.30f.
ἐπειρῶντο διαχειρίσασθαι. The imperfect means that they were
attempting, when prevented by Roman intervention, πειρᾶσθαι with
the infinitive occurs here only in the NT; see BDR § 392.1a, n. 2. For
διαχειρίζεσθαι see 5.30; it is used in the NT in these two places
only.
This verse is a sufficiently accurate summary of what is described
at length in chs. 21, 22.

22. Paul now winds up the speech, in which he has described his
pre-christian life in Judaism, his conversion, his call to evangelize
the Gentiles, and the attack on him which led to his arrest and thus to
his present appearance before Festus and Agrippa, by summing up
both the religious basis and the theological structure of his Christian
life and belief, οὖν brings him back to the main theme (BDR § 451.1,
n. 2).
επικουρία occurs here only in the NT; the unusual word may have
been chosen on account of the distinction ascribed by Bengel (482)
to Ammonius [107]: βοηθεῖ μὲν ό συνών. ἐπικουρεῖ δέ ό έξωθεν
 61. FESTUS, AGRIPPA, AND PAUL. 25.23-26.32. 1165

εις βοήθειαν ἥκων. Paul’s help comes not from within himself but
ἀπὸ τοῦ θεοῦ. επικουρία is used by medical writers but also in
military contexts; ND 3.67f. quotes an inscription from Cyrene (AD
154), re-edited by J. Μ. Reynolds (JRS 68 (1978), 111-21) contain-
ing the words κα]θεστῶτες τὴν ἐπικουρίαν παρά τών
Έλλή[νων?..., rallying help from the Greeks. τυγχάνειν occurs once
in Lk., 5 times in Acts, 8 times in the rest of the NT; on the whole a
Lucan characteristic.
ἕστηκα, perfect in form, is commonly used with a present
meaning, I stand, but in this context the perfect element that
underlies the present (I have taken my stand and here I am) is clear:
Paul has been standing as he now does ἄχρι τής ἡμέρας ταύτης. He
stands as one who bears witness to all, of whatever rank.
The content of his testimony is, according to this verse, simply the
message of the OT, reinforced by the affirmation that what the OT
foretold has now been fulfilled. For this frequent insistence that
Paul’s Christian message is simply what the OT contains (and
therefore ought not to offend Jews but to be accepted by them) see
e.g. 24.14; 28.23.
Moule (IB 169) notes a second (cf. v. 20) example of the
misplacing of τε. One would have expected οἵ τε προφῆται... καὶ
Μωυσῆς. The result is that though τε suggests that some second
member of the sentence is to come, καί Μωυσῆς gives the impres-
sion of being an afterthought, as if Luke intended at first to refer
simply to the predictive element in the OT and then reflected that the
Gospel is the fulfilment not only of the prophets but of the Law too.
For a similar addition cf. Jn 1.45. ‘Dass hier Mose an zweiter Stelle
genannt wird, hängt wohl damit zusammen, dass er die künftige
Totenauferstehung lediglich “angedeutet” hat’ (Schneider 2.376).
εκτός, meaning except, is unclassical (BDR § 216.2, n. 7); but
ούδέν (rather than μηδέν) with a participle is classical (BDR § 430.2,
n. 4). The speech retains its mixed character to the end.
For Eb. Nestle’s conjecture that v. 8 should be transferred to
follow this verse and precede v. 23 see on v. 8.

23. The sentence follows upon μαρτυρόμενος (v. 22); it could

follow neatly upon εἰ ὁ θεός νεκροὺς ἐγείρει (ν. 8), but this is not
sufficient ground for a transposition (see above).
Dodd (AS 17) writes that ‘the main heads of Paul’s discussions ...
are introduced by the particle εἰ used interrogatively: “whether the
Messiah is a suffering Messiah ...” This interrogative form is
evidently appropriate to a method of teaching which is described by
the verb διαλέγεσθαι, “to discuss”. We conclude that in addition to
what is called “preaching” (κηρύσσειν), early Christian mission-
aries also employed the method of discussion, in which certain
questions were propounded—questions arising unavoidably out of
1166 COMMENTARY ON ACTS

the kerygma—and answers sought by a study of the Old Testament.’

It is to be observed that διαλέγεσθαι is not used in this context (it is
used at 17.2, which Dodd brings into his argument). Neither this
suggestion, nor that of Rendel Harris (Testimonies, Part I, 1916;
19f.), that we have before us the headings of sections in a Testimony
Book (‘The words are headlines of Testimonies, awkwardly incorpo-
rated in the text, and are betrayed as such by the previous reference to
the prophets and Moses, who are to answer the questions’) is called
for. εἰ may bear the meaning that after a verb of feeling; see BA, s.v.
II, 442, with reference to Kühner-Gert 2.369.8 [better, II.2.369.8];
also Begs. 4.321, ‘εἰ is best rendered by “that”, but there is in εἰ a
stronger implication that the proposition which follows is denied and
must be argued out than would be made by the simple ὅτι.’ Similarly
Blass (269).
παθητός, liable to suffering (e.g. Plutarch, Numa 8 (65), Pelopi-
das 16 (268) ) occurs nowhere else in the NT. It was taken up in the
second century (Ignatius, Ephesians 7.2; Polycarp 3.2; frequently in
Justin, e.g. Trypho 34, 36) not least in controversy with Jews.
πρώτος seems to qualify the (unexpressed, though ό Χριστός is
clearly implied) subject of μέλλει. He, the Messiah, will proclaim
light (in this context generally, an image of salvation) to both the
people (λαός, the Jews) and the Gentiles. He will be the first to do
this, and he will do it ἐξ ἀναστάσεως νεκρών. Turner (Μ. 3.260) and
Moule (IB 73) are undecided whether to regard ἐξ as instrumental or
as (quasi-)local (Moule: ‘if he was destined to be the first (to come
from) a rising from the dead and announce...’). We may render on
the basis of (cf. Paul’s use of ἐκ πίστεως, ἐξ ἔργων); this is a sort of
instrumental use which may also be said to carry the local with it. See
however Page (250): ‘... that he first by a resurrection from the dead
...’; and Bruce (1.447): ‘the first that should rise from the dead’.
The Messiah, who suffers and is raised from death, proclaims light
to both Jews and Gentiles. Thus the three points are made which,
though Paul insists that they are already proclaimed by the prophets
and Moses, incense the Jews. It is asserted that these propositions are
already contained in the OT, but no passages are cited in support of
the assertion. In controversy between Jews and Christians such
passages must have been quoted and their interpretation disputed
(see the note on v. 24), but this is a debate into which Luke never
enters (though a passage such as 17.11 shows that he was aware of
it).
24. ἀπολογουμένου. For this word see 19.33; 26.1f. Strictly
speaking, Paul was not now making a defence; he was not on trial,
since his case had been referred to Rome. He knew however that he
was accused by the Jews (v. 2) and that Festus, with the assistance of
Agrippa, must have been preparing a dossier that would present the
 61. FESTUS, AGRIPPA, AND PAUL. 25.23-26.32. 1167

case before the imperial court. A good impression made upon Festus
would be to his advantage, and to the advantage of his cause.
Festus interrupts at the point at which earlier the Jews had
interrupted, but for a different reason, μεγάλη is predicative: Fes-
tus—loud was his voice—said ... μαίνη: the story of a crucified and
risen Messiah is nonsense, (a) because a king would not proceed by
the way of suffering and death, and (b) because dead men do not rise
up (see on 17.32).
Knox (Hell. El. 29) writes, ‘Festus suggests, whether seriously or
not, that St Paul’s studies have driven him mad, implying that he has
been quoting from a number of “writings” ... But Luke has omitted
the testimonies which alone would explain Festus’ interruption.’
Knox is presumably referring to the word γράμματα; for this
however the rendering learning is quite possible; BA 330 cite
Xenophon, Cyropaedia 1.2.6; Plato, Apology 26d; PLond 43.2.γ;
Dan. 1.4; Epistle of Aristeas 121; Testament of Reuben 4.1; of Levi
13.2. A adds (after γράμματα) ἐπίστασθαι—to know many writings.
On the position of σε, cf. BDR § 473.1: ‘Auch im NT gilt die alte
Regel, dass unbetonte (enklitische) Pronomina udgl dem Satzanfang
möglichst nahe gerückt werden, wodurch oft Zusammengehöriges
getrennt wird.’ Bruce (1.448): ‘The effect is an added emphasis both
on σε and τά πολλά γράμματα.’
For the impression of madness cf. 1 Cor. 14.23 (μαίνεσθε); but it
is learned, not enthusiastic, madness of which Festus accuses Paul.
Cf. ἐμμαινόμενοι in v. 11 (with the note). Thus Bruce (1.448), ‘The
remark was not offensive; both μαίνομαι and μανία are cognate with
μάντις, “seer”, “inspired person” ’; he refers to Plato, Phaedrus
245a. Similarly Page (250) speaks of ‘the philosophic “madman”,
and quotes Phaedrus 249d. StrB 2.770 quote Targum Yerushalmi I
on Num. 22.5: Balaam had become ‘irrsinnig’ because of the
greatness of his learning. This rather flattering interpretation of
Paul’s ‘philosophic’ madness is not suitable to the present context. In
the next verse Paul firmly rebuts the suggestion of madness; he is not
what Festus says that he is. This also is against Schille’s view (453)
that for Luke the charge is a way of commending Paul’s ‘Schrift-
gelehrsamkeit’.
For εις μανίαν περιτρέπει cf. Lucian, Abdicatus 30, ές μανίαν
περιέτρεψε (in a discussion of the various causes that lead to μανία).
MM 388 quote an unedited Tebtunis papyrus: φαίνη εις μανίαν
έμπεπτωκέν[α]ι. See also Longinus, De Sublimitate 8.4; Sibylline
Oracles 1.171f. POxy 1.33. Col. 4.9-15 (... νὴ τὴν σὴν τύχην ούτε
μαίνομαι οὕτε ἀπονενόημαι ...); Pliny, Epistles 10.96.4 (fuerunt
alii similis amentiae—of the Christians’ inflexibilis obstinatio).

25. Paul rebuts the charge sharply, οὐ μαίνομαι, but with due
courtesy: κράτιστε Φῆστε. Cf. Lk. Γ.3; Acts 23.26; 24.3. κράτιστος
1168 COMMENTARY ON ACTS

is equivalent to the Latin egregius, and is fittingly used of the prefect,

as a knight.
αλλά, on the contrary.
ἀποφθέγγομαι. Cf. 2.4, 14, where the word is often said to
suggest inspired, even ecstatic, speech. This may be true in those
verses but is hardly true here, where Paul insists that he is speaking
words αληθείας καί σωφροσύνης. The first of these terms insists
that what Paul says (and at the same time its use suggests the
contents of the Gospel) is in accordance with the fact; the second
claims that his treatment of the facts is in accordance with sober
reason. Betz (206, n. 5) notes a number of parallels in Lucian;
Icaromenippus 30; De Mercede Conductis 36; De Saltatione 12;
Timon 55; Bis Accusates 17; Piscator 16, 41; Somnium 10; Pro
Imaginibus 11, 20. For a contrast with σωφρονεῖν, Abdicates 1; De
Syria Dea 22. For this contrast see also Xenophon, Memorabilia
1.1.16; also POxy 1.33 (quoted on v. 24; Schille 453); Plato,
Protagoras 323b; also Lk. 8.35 (Mk 5.15) (Pesch 2.279). These
passages illustrate and confirm Roloff’s statement (355) about ‘das
in der hellenistischen Literatur geläufige Gegensatzpaar: Wahnsinn
(d.h. Masslosigkeit und Unkontrolliertheit des Denkens)—Vernunft
(d.h. besonnen-realitätsbezogenes Denken).’
In αληθείας κ. σωφρ. <b>ήμ., ‘the first gen. is objective, the second
subjective’ (Page 250). This is correct: the words set forth truth, and
they are controlled by sober judgement. The remark by Delebecque
(127), ‘Luc donne ici au mot αλήθεια le sens qu’il a chez Paul, ou il
est constant’, is not quite so convincing. Still less convincing is
Lüdemann’s judgement that σωφροσύνη is a thoroughly un-Pauline
term. He overlooks Rom. 12.3; 2 Cor. 5.13; not to mention Paul’s
treatment of glossolalia in Corinth.

26. The king knows about these things. It might at first seem that
these things were the matters which (according to Paul) were written
in the prophets and Moses, that is, that the purpose of Israel was to be
fulfilled in the death and resurrection of the Messiah and the
extension through him of the light of God’s truth and blessing
beyond Israel to the Gentile world. The reference however at the end
of the verse to something done suggests rather the claim that the king
was aware of the Christian event, that is, the ministry, death, and
resurrection of Jesus. Agrippa II (see on 25.13) was born in AD
27/28, and therefore certainly was not aware of the story of Jesus at
first hand; moreover, he lived at Claudius’ court till 50 or later, when
he was appointed king of Chalcis. He did, however, show consider-
able concern over Jewish affairs, doing his best to prevent rupture
and war with Rome (see e.g. his long speech dissuading the Jews
from war, Josephus, War 2.344-407); there is no reason why he
should not have heard of the origins of Christianity, especially if
 61. FESTUS, AGRIPPA, AND PAUL. 25.23-26.32. 1169

Christians were felt to be a disturbing factor in Jewish life. So, more

or less, Chrysostom, Homily 52.4; Ενταύθα περὶ τού σταυρού λέγει
τούτο, περί τῆς ἀναστάσεως, και ὅτι πανταχοῦ τῆς οικουμένης
γέγονε τὸ δὸγμα.
παρρησιαζόμενος λαλώ. Cf. ν. 2. παρρησιάζεσθαι is an Acts
word: 9.27, 28; 13.46; 14.3; 18.26; 19.8; elsewhere in the NT only
Eph. 6.20; 1 Thess. 2.2. Paul can speak openly and freely.
The transitive use of λανθάνειν (object, αὐτὸν) is classical: BDR
§ 149.1, n. 2. Begs. 4.322 notes the present tense: none of these
things is escaping him. For the litotes cf. v. 19. The form οὐθέν is
much less common in the NT than that with δ; see on 15.9. Moule
(IB 167f.): ‘οὐ belongs not with πείθομαι but, as a double negative,
with λανθάνειν.’
καί is omitted by B 104 1175 pc h vgmss. τι is read by P74 A E syp;
it is omitted by Β Ψ 36 614 1175 2495 pc syh and replaced by ἐγώ in
945 1739 1891 pc. ούθέν is omitted by P74 A E 33 81 al.
ού γάρ έστιν ... πεπραγμένον. Moule (IB 19) raises the question
why the periphrastic perfect should be used here but makes no
attempt to answer it. We must probably be content to say only that
Luke preferred it to πέπρακται.
‘Done in a comer’ has become an English idiom, but it was not
used in antiquity. Delebecque (127): ‘L’expression, qui n’est pas
courante, semble prise à Platon, Gorgias 485d.’ This passage runs:
... μετά μειρακίων ἐν γωνίᾳ τριών ἣ τεττάρων ψιθυρίζοντα. We
may add Epictetus 2.12.17: τὸν γὰρ ποιοῦντα αὐτὸ οὐκ ἐν γωνία
δηλονότι δεήσει ποιεῖν, αλλά .. .Bruce 1.449 adds Terence, Adelphi
4.9.10f. (or 5.2.10f.): ... interea in angulum aliquo abeam.
27. Agrippa, like some other members of his family, though in fact
a Gentile, could on occasion represent himself as a Jew in spirit, and
had certain rights in the Temple and in the appointment of the high
priest. Paul appeals to his knowledge of and belief in the OT—
knowledge and belief which in some contexts he might have
claimed, whether he had them or not. Reference to his belief in
the prophets points back to τούτων in v. 26, and, contrary to the
suggestion made on v. 26, might suggest that these things are the
prophetic notions of vv. 22f.—or rather, that there is no great
difference between the two interpretations of τούτων. What the
prophets foretold is what has happened in the story of Jesus. Belief in
the prophets is (for Paul—and Luke) not an end in itself but a step on
the way to belief in Christ.
28. Agrippa’s words to Paul are perhaps the most disputed,
as regards their construction and meaning, in Acts. The text
most widely accepted (and printed in NA26) is: ἐν ὀλίγω με πείθεις
Χριστιανόν ποιῆσαι. For πείθεις, A reads πείθη. This is often
ascribed to accidental error: HI is not unlike EIC. But the two
1170 COMMENTARY ON ACTS

syllables are not really alike, and πείθη would take up πείθομαι in v.
26: You are persuaded that I know all about Christianity; perhaps
you are persuaded that I believe it too. Blass (270): ‘tutissimum est A
sequi’. But the connection with v. 26 could cut both ways; πείθεις
might have been changed into πείθη in order to make it closer. And it
is perhaps not very safe to follow A alone. For Χριστιανόν, has
χρηστιανόν; this is a purely orthographic variant. For ποιῆσαι, Ε Ψ
latt sy, with Cyril of Jerusalem, have γενέσθαι: You are
<a>
persuading (seeking to persuade) me to become a Christian. This
makes good sense; it evades a difficulty however and the evidence is
on the whole late, so that it would be unwise to adopt this reading.
Two interrelated questions remain: the meaning of ἐν ὀλίγῳ and of
ποιήσαι.
With όλίγῳ may be understood χρόνῳ or λόγῳ. Calvin (2.283)
says that the phrase may refer to time or degree, but degree is not
easy to support. Μ. 3.262 similarly: either in a short time or by a
short argument. It does not seem possible to take the phrase in the
sense all but a little, that is, almost, though ἐς (εἱς) ὀλ. and παρ’ ὀλ.
have this meaning, or something like it (e.g. Thucydides 4.129.5, ές
ὀλίγον ἀφίκετο πᾶν τό στράτευμα ... νικηθῆναι. There is no great
difference between time and argument: a brief argument would
occupy a short time, a short time would not permit a long argument.
Probably the decisive consideration—or the consideration that
comes nearest to being decisive—is the recurrence of ἐν ὀλίγω
together with ἐν μεγάλῳ (for a variant see on v. 29) in the next verse.
μεγάλῳ would hardly be used of time. 'It seems best to understand
πόνῳ with ὀλίγῳ, as this noun could fitly stand with both μεγάλῳ
and ὀλίγῳ = with little trouble, with little cost’ (Knowling 513). The
same opinion is held by Page (251): ‘ἐν ὀλίγω is clearly = “with
little (trouble, effort)”, ἐν being instrumental, its sense being
determined by the use of the phrase in Paul’s reply ἐν ὀλ. καί ἐν
μεγάλῳ = “with little or with great (trouble)”. It cannot = ἐν ὀλίγῳ
χρόνφ “quickly” for ἐν μεγάλῳ ... could not mean “in much
time”.’ Similarly Delebecque (128). Cf. Thucydides 2.84.2: αἱ νῆες
ἐν ὀλίγῳ ἤδη ουσαι... See however on ν. 29.
The second question is the meaning of Χριστιανόν ποιῆσαι. There
are two main possibilities, (a) A. Fridrichsen (Coniectanea Neotesta-
mentica 3 (1939), 14f.) drew attention to the parallel in Xenophon,
Memorabilia 1.2.49: πείθων μὲν τοὺς συνόντας ἑαυτῷ
σοφωτέρους ποιεῖν τών πατέρων, persuading his companions that
he was making them wiser than their fathers. From this is deduced
and illustrated the rule ‘nec subiectum nec obiectum infinitivi
exprimi debet si coincidit cum subiecto obiectove verbi principalis’
(Zerwick § 395). Zerwick and Turner (Μ. 3.147; at greater length in
Insights 97-100) take this as justifying the rendering mihi vis
persuadere (te) brevi tempore (me) fecisse Christianum (Zerwick),
 61. FESTUS, AGRIPPA, AND PAUL. 25.23-26.32. 1171

You seek to convince me that you have made me in a moment a

Christian (Turner). This has the great advantage of providing a better
connection with v. 29, where Paul seeks that Agrippa and his other
hearers should become (γενέσθαι) Christians, that is, he desires to
make them Christians. On the other hand, You want to persuade me
that you have made me a Christian, hardly makes sense, even if we
allow for a good deal of irony. One who has been made a Christian
does not need to be persuaded that he has been made a Christian, (b)
The main alternative is based on the use of ποιεῖν in 3 Kdms 20.7
(LXX) where σὺ νῦν οὕτως ποιείς βασιλέα ἐπὶ ’Ισραήλ renders the
Hebrew (1 Kings 21.7) and appears
to mean ‘play the king’ (cf. Latin (Christianum) agere; BDR § 5.4, n.
23), so that the meaning of the present passage would be, You are
persuading me to play the Christian—that is, to act a part for your
convenience (by getting me to confess belief in the prophets). This is
an attractive suggestion, but one over-literal rendering of a Hebrew
verb is not sufficient to establish a Greek idiom. It is ‘insufficiently
documented’ (Begs. 4.323). Haenchen adds references to Johannes
Climacus and Johannes Malalas; these are too late to give much
support, but the Latin parallel, Christianum agere (noted already by
P. W. Schmiedel in EBib 754) may be worth more than he allows.
Opinions differ. Ropes (Begs. 3.239): ‘The reading ... of B ...,
although difficult, yields an intelligible sense (“play the Christian’’)
and must be accepted’. But Clark lii: ‘That ποιώ Χριστιανόν can
mean “play the Christian” is to me incredible. The reading of A
πείθη = ‘ ‘thou thinkest (cf. οὐ πείθομαι in v. 26) gives a fair sense.
The variant γενέσθαι ... for ποιῆσαι looks like a conjecture.’
Bauemfeind (270) thinks that what we have in the text is a
combination of two thoughts, In Wenigem überredest du mich, and
In Wenigem machst du mich zum Christen. Schneider (2.378),
mentioning both possibilities, makes the important additional point,
‘Auf jeden Fall nimmt Paulus selbst die Aussage des Königs ernst,
wie der folgende Vers deutlich macht.’ It has usually been assumed
that the answer is one of irony on the official side (‘ “Christian” in
the mouth of Agrippa can only be interpreted as a sneer’—Begs.
4.322), serious on Paul’s, but this may not be true, and we have to
ask what impression Luke wished to convey rather than, in the first
instance, what actually happened. Weiser (655) speaks of Agrippa’s
‘zwischen Emst und Ironie schwebende Antwort’. He is not speaking
of ‘playing the Christian’ but of being made a Christian; this is
confirmed by Paul’s answer. It may well be that Bengel (483) rightly
caught the impression that Luke wished to suggest: 'Occurrit ergo
hic, Festus, sine Christo: Paulus, Christianissimus: Agrippas, in
bivio, cum optimo impulsu.’

29. εὐξαίμην ἄν. ‘Der Potentialis (Opt. mit ἄν) zur Bezeichnung
1172 COMMENTARY ON ACTS

des lediglich Gedachten ist in der Volssprache ganz abhanden

gekommen, im NT nur noch selten und nur in Lk und Apg’ (BDR §
385.1); another example of the good style used (intermittently) by
Paul on this occasion. Cf. Acts 8.31; 17.18. See Μ. 3.123. Moule (IB
151) points out that the construction is logically that of a conditional
sentence with the protasis omitted: if only it were possible, I...
καὶ ἐν ὀλίγω καὶ ἐν μεγάλῳ. See the discussion of ἐν ὀλίγω in ν.
28. If it was right there to expect the same noun supplement with
ὀλίγῳ and with μεγάλῳ, it will be necessary here to take them as
with little trouble, with great trouble. The difficulty here is that the
two phrases are joined not, as might be expected, with ἤ (or) but by
καί (and). This is surprising but not in fact impossible. See BDR §
442.9a (καί altemativum) n. 27 (examples including Acts 10.14,
κοινὸν καὶ ἀκάθαρτον, common or unclean). Cf. also Plato, Apol-
ogy 23a, where as in our verse the καί is corrective. ‘With little
argument, aye, and with much, if needed ...’. The alternatives to this
will be either Begs. 4.322 (‘ “in small and great”—meaning “alto-
gether,” “wholly” ’) or the sense of v. 22, μικρῷ τε καί μεγάλῳ:
Paul’s desire to see men become Christians applies to the least and to
the greatest, to the king himself. If either of these ways of taking the
phrase is accepted the question of ἐν ὀλίγω in v. 28 will be reopened.
On the whole see P. Harlé, ‘Un “Private-Joke” de Paul’, NTS 24
(1978), 527-33.
σήμερον may be taken either with the preceding participle or the
infinitive that follows. There is no way in which this question can be
decided with confidence.
τοιούτους ὁποῖος is rightly translated in the Vulgate, tales qualis:
of such a kind (of person) as I am. Paul wishes for all his hearers the
election, the call, and the commission that he himself has; he does
not wish them his bonds. For παρεκτός cf. ἐκτός in v. 22.
‘Die Replik des Paulus ist keineswegs von geringerer rhetorische
Eleganz’ (Roloff 355).
For εὐξαίμην, A H L Ρ 049 81 326 1241 al have εύξάμην.
μεγάλφ is read by Ρ74 Α Β Ψ 33 81 945* 1739* pc latt; instead
<a> sy have πολλῷ. For the significance of this see on v. 28. It makes
χρόνῳ a possible supplement.

30. ἀνέστη τε. A few MSS (33 945 1739 1891 2495 al have δέ
instead of τε; there is a major variant, probably of Western origin, in
(614) <a> h (syh**) sa, καὶ ταῦτα εἰπόντος αὐτοῦ ἀνέστη. This gives
a more explicit connection with what precedes and is almost
certainly secondary.
ὁ βασιλεὐς καί ὁ ἡγεμών. The repeated article makes it clear that
two persons are intended.
συγκαθήμενοι: the verb is used specifically of those who sit with
others, or another, as assessors in court. So for example Xenophon,
 61. FESTUS, AGRIPPA, AND PAUL. 25.23-26.32. 1173

Hellenica 2.4.23, οἱ μὲν τριάκοντα ... ξυνεκάθηντο ἐν τῷ

ξυνεδρίῳ. ΜΜ 608 have a papyrus example.
‘Hier [in this verse and the two following] hat die Apologetik
erzählerisch ihren Höhepunkt erreicht’ (Lüdemann 265). This is a
fair statement; Paul is explicitly pronounced innocent by the highest
authorities he has yet encountered. How Luke could be aware of
what the king and governor said to each other is not explained.

31. ἀναχωρήσαντες, from the (possibly) public courtroom (25.23,

the ἀκροατήριον) into a retiring room. Aorist participle: the con-
versation took place after they had retired.
πράσσει: BDR § 322.3, n. 3 rightly say that the present tense is
‘auf des Paulus ganze Lebensweise und besonders sein Christentum
bezogen’.
ὅτι is omitted by P74 pc. τι is omitted by B <a> it sy ; this improves
the Greek. With ούδέν, τι is not required; it is surprising that it is
retained (quid) in the Vulgate.
Hemer (349) commenting on the lack of any source for the private
conversation between Festus and Agrippa, adds ‘presumably a
verdict in some such terms must in any case have been made known
to Paul after the consilium of the judges’. But Festus and Agrippa
were not acting as judges; Festus was gathering information for his
dossier and Agrippa was indulging a curious whim. Of course, if
kindly disposed they may have passed on to Paul the essence of the
report that would be sent to the higher court.

32. Agrippa (whom Luke may possibly—though inaccurately—

have been thinking of as a representative Jew, matching Festus, the
representative Roman—see Roloff 348) had been consulted about
Paul’s case (25.26) and now gives his opinion, which, like v. 31 and
other passages to the same effect, must have seemed to Luke and his
readers very quotable. The appeal has been made and must go
forward; otherwise Paul could have been set free. It seems, however,
that continuing custody with a free voyage to Rome will have suited
Paul very well. At least he was protected from Jewish assassins. This
may have been in mind from early in the proceedings. ‘This does not
mean that in strict law the governor could not pronounce an acquittal
after the act of appeal. It is not a question of law, but of the relation
between the emperor and his subordinates, and of the element of
non-constitutional power which the Romans called auctoritas,
“prestige”, on which the supremacy of the Princeps so largely
depended. No sensible man with hopes of promotion would dream of
short-circuiting the appeal to Caesar unless he had specific authority
to do so ... Since the charges were extra ordinem in large part, the
appeal was automatically valid. Festus was naturally only too glad,
politically, to rid himself of the prisoner. To have acquitted him
1174 COMMENTARY ON ACTS

despite the appeal would have been to offend both the emperor and
the province’ (Sherwin-White 65).
The apodosis (ἀπολελύσθαι ἐδύνατο) lacks ἄν (Μ. 3.91f.). BDR
§ 358.1, n. 2 point out that the imperfect without ἄν is used to
express necessity or possibility where the thing in question did not
happen. Similarly Page (251): ‘ἄν is often omitted with simple verbs
such as ἔδει (24.19), ἔχρην etc. So in Latin poterai, debebat, instead
of posset, deberet.' Cf. Wisdom 11.20. With the pluperfect
ἐπεκέκλητο Μ. 3.91 compares μεμενήκεισαν in 1 Jn 2.19, but this is
in the apodosis. The pluperfect is ‘significant as implying that this act
of Paul had placed him in an irrevocable position’ (Hemer 132).
326 2464 omit Άγρίππας... οὖτος, probably by homoeoteleuton
(see ν. 31). P74 has after these words only ἀπολελύσθαι ἐδύνατο;
this may be intended abbreviation achieved by combining two
favourable statements. At the end of the verse 97 pc h w (syp syhmg)
add καὶ οὕτως ἔκρινεν αὐτόν ὁ ἡγεμὼν ἀναπέμψαι Καίσαρι. This
decision had been taken in principle as far back as 25.12; but ἔκρινεν
here may refer to a specific decision to send Paul on a specific ship
under a specific guard. It is to be noted that h syp syhmg omit the first
nine words of 27.1.
 XVI
PAUL REACHES ROME
(27.1-28.28)

62. THE SEA VOYAGE 27.1-44

(1) When it was decided that we should set sail for Italy they handed over
Paul and some other prisoners to a centurion called Julius, of the Augustan
Cohort. (2) We embarked on a ship of Adramyttium which was about to sail
to places on the coast of Asia and put out to sea; the Macedonian
Aristarchus, of Thessalonica, was with us. (3) The next day we put in at
Sidon and Julius treated Paul in a kindly way, allowing him to go to his
friends to receive care. (4) Thence we put out to sea and sailed under the lee
of Cyprus because the winds were contrary. (5) We sailed across the open
sea off Cilicia and Pamphylia1 and put in at Myra in Lycia. (6) There the
centurion found an Alexandrian ship sailing to Italy and embarked us on it.
(7) Sailing slowly for a number of days and barely getting as far as Cnidus,
since the wind did not permit us to approach we sailed for refuge under the
lee of Crete off Salmone. (8) Coasting along it with difficulty we came to a
place called Fair Havens, to which the town of Lasaea is near.
(9) Since a considerable time had now elapsed and sailing was already
risky because even the Fast was now already past Paul offered advice, (10)
saying to them, ‘Men, I see that the voyage is going to be attended by
damage and much loss not only of the cargo and the ship but also of our
lives.’ (11) But the centurion was persuaded by the captain and the owner2
rather than by Paul’s words. (12) Since the harbour was not suitable for
wintering the majority formed the plan of putting out from there and getting
if they could to Phoenix, a harbour of Crete looking into the south west and
north west winds, and wintering there.
(13) When a gentle south wind sprang up they supposed that they had
achieved their purpose; they set out and coasted along Crete as close in shore
as possible. (14) But before long a tempestuous wind called Euraquilo flung
itself down from3 the island. (15) The boat was seized by the wind and was
unable to head into it, so we gave way to it and ran before it. (16) We ran
under the lee of a certain small island called Cauda and were scarcely able to
gain full control of the dinghy. (17) They hauled it up and4 made use of
auxiliary devices, frapping the ship, and fearing lest they should be cast upon

1NJB, taking a fortnight to reach.

2Or, manager.
3Greek, from it; or, upon it (the ship).
4NJB, used it to undergird.
1175
1176 COMMENTARY ON ACTS

the Syrtis they dropped the sea anchor, and so drifted. (18) Since we were
severely storm tossed on the next day they5 jettisoned [some of the cargo].
(19) On the third day they threw out with their own hands the ship’s gear.
(20) For many days neither sun nor stars were to be seen, no small storm was
upon us, and finally all hope of our being saved was disappearing.
(21) 6Since many were going without food Paul then stood in their midst
and said, ‘Men, you ought to have listened to me so as not to have left Crete
and incur this damage and loss. (22) And now I advise you to take heart, for
there will be no loss of life of any of you but only of the ship. (23) For this
night there stood by me an angel of the God whose I am and whom I serve,
(24) saying, “Fear not, Paul, you must stand before Caesar, and see, God has
granted you all those who are sailing with you.” (25) So take heart, men; for
I believe God that it will be just as it has been told me. (26) But we have to
be cast on a certain island.’
(27) When the fourteenth night came and we were7 being tossed about in
Adria, in the middle of the night the sailors thought that land was approaching
them. (28) They took soundings, and found twenty fathoms; they moved the
boat on a little, sounded again, and found fifteen fathoms. (29) Fearing lest we
should fall upon rocky places they threw out four anchors from the stem and
wished it were day. (30) The sailors were seeking to escape from the ship and
let down the dinghy into the sea, under the pretence that they were going to let
out anchors from the bow. (31) Paul said to the centurion and the soldiers,
‘Unless these men stay in the ship you cannot be saved.’ (32) Then the soldiers
cut the ropes of the dinghy and let it fall away.
(33) When it was nearly day Paul exhorted them all to partake of food,
saying, ‘Today8 you are looking to the fourteenth day and are continuing
without food, having taken nothing. (34) Therefore I urge you to partake of
food. For9 this is for your welfare,10 for not a hair will be lost from the head
of any of you.’ (35) When he had said this he took a loaf and gave thanks to
God before them all; he broke it and began to eat. (36) They all took heart
and themselves also partook of food. (37) We were in all 276 souls in the
ship. (38) When they were satisfied with food they set about lightening the
ship, casting the11 food into the sea.
(39) When day broke they did not recognize the land but perceived a bay
with a beach on which they planned if possible to run the ship. (40) They
detached the anchors and let them slip into the sea, at the same time
loosening the fastenings of the rudders and raising the foresail to the breeze
they held on to the beach. (41) They12 ran upon a shoal and ran the ship
aground. The bow became fixed and remained firm but the stem began to
break up under the violence of the waves. (42) It was the plan of the soldiers

5Greek, made a jettisoning.

6Greek, literally: Since there was much foodlessness.
7RSV, drifting across the sea of; NJB, being driven one way and another.
8RSV, you have continued in suspense; NEB, you have lived m suspense; NJB, you
have been in suspense.
9RSV, it will give you strength; NEB, your lives depend on it; NJB, your safety
depends on it.
10Welfare; elsewhere the Greek word might be translated salvation.
11RSV, wheat; NJB, com.
12NEB, found themselves caught between cross-currents; NJB, the cross-currents
carried them into a shoal.
 62. THE SEA VOYAGE. 27.1-44. 1177

to kill the prisoners, lest any should swim off and escape. (43) But the
centurion, wishing to get Paul safely through, put a stop to their intention and
commanded those who were able to swim to throw themselves overboard
and come first to land (44) and [then]13 the rest, some on planks of wood and
some14 on some of those who came from the ship. And in this way it came
about that all got safely to land.

Bibliography
A. Acworth, JTS 24 (1973), 190-3.
C. K. Barrett, FS A. T. Hanson, 51-64.
J. Behm, TWNT 4.928-32.
Μ. E. Boismard and A. Lamouille, EThL 63 (1987), 48-58.
H. J. Cadbury, JBL 48 (1929), 419f.
L. Casson, Ships and Seamanship in the Ancient World (1971).
D. J. Clark, BT26 (1975), 144-6.
J. Μ. Gilchrist, JSNT 61 (1996), 29-51.
E. Haenchen, FS Bultmann (1964), 235-54.
R. P. C. Hanson, TU 102 = StEv IV 315-18.
C. J. Hemer, Tyndale Bulletin 36 (1985), 79-109.
N. Heutger, BZ 28 (1984), 86-8.
R. A. Kraft, JBL 94 (1975), 256-65.
D. Ladouceur, HThR 73 (1980), 435-49.
G. B. Miles and G. Trompf, HThR 69 (1976), 259-67.
Eb. Nestle, ZNW 8 (1907), 75, 76.
Μ. Oberwies, NovT 30 (1988), 169-83.
R. Μ. Ogilvie, JTS 9 (1958), 308-14.
P. Pokorny, ZNW 64 (1973), 233-44.
S. Μ. Praeder, CBQ 46 (1984), 683-706.
B. Reicke, ThZ4 (1948), 401-10.
V. K. Robins, Biblical Research (1975), 1-14.
J. Rougé, VigCh 14 (1960), 193-203.
J. Rougé, Recherches sur l'organisation du commerce maritime en Médi-
terranée sous l'empire romain (1966).
J. Smith, The Voyage and Shipwreck of St Paul (1880/1978).
A. Suhl, ZThK 88 (1991), 1-28.
G. W. Trompf, in C. H. Talbert, Luke-Acts, 225 –39.
J. Wehnert, ZThK 87 (1990), 687-99; 88 (1991), 169-80.
W. J. Woodhouse, Encyclopaedia Biblica, 3690-92.
W. P. Workman, ExpT 11 (1900), 316-19.
13Then is not in the Greek.
14RSV, on pieces of the ship; NEB, on parts of the ship; NJB, on pieces of
wreckage.
1178 COMMENTARY ON ACTS

Commentary

There can be no doubt that whoever was responsible for the

substance of this chapter was familiar with the sea and with
seafaring, in particular with conditions and places in the Medi-
terranean and the Adriatic. On this subject J. Smith, The Voyage and
Shipwreck of St Paul (Grand Rapids 1978, reprint of the 1880
edition), is still a classical work. It might almost be said that the
writer’s knowledge of the sea and of sailors is too good for us; he
uses what are plainly technical terms, some of which are otherwise
unknown so that we can only guess at their meaning (see the detailed
notes below). The question that arises is whether all these details, and
the events described, belong to a journey actually made by Paul on
his way from Caesarea to Rome. The outline of the voyage was taken
from life; were the references to Paul, to the things said and done by
him on the voyage, an integral part of the story from the beginning,
or were they inserted into a narrative of a voyage and shipwreck that
did not originally contain them?
The outline of the story is cast in the first person plural (vv. 1,2, 3,
4, 5, 6, 7, 8, 15, 16, 18, 20, (26), 26, 29, 37). That there are verses
where Paul is subject or object and is referred to in the third person
singular is no proof that either set of verses was inserted. It would be
not only untrue but ridiculous to write in vv. 9f., We spoke the
following words. If anyone spoke them it was Paul and Paul alone.
The same sort of observation covers the use of the third person
plural. Thus in v. 12 the majority who ἔθεντο βουλήν ... did not
include Paul or Paul’s companions—hence they is used; v. 13
follows upon this—they must include Paul and any who had agreed
with him for they were still on the ship and were coasting along by
Crete, but παρελέγοντο is used with those in mind who had chosen
that course, whereas in v. 15 the first person plural ἐφερόμεθα
applies to all who were on board, none of whom wished to be driven
by a hurricane. Similar considerations apply throughout. It has to be
asked whether the Pauline interventions are credible as parts of the
story as a whole.
Weiser (659) counts as Pauline insertions vv. 3b, 9-11,21-26,31,
33-36, 43a, and considers that these passages show that ‘der Text
nicht ein historisch getreuer Geschichtsbericht ist’ ; similarly Haen-
chen (678-80), and many others. ‘Die eingeschobenen Szenen
entsprechen genau dem lukanischen Paulusbild. Paulus steht immer
im Mittelpunkt. Er ist nie um Rat verlegen ... Lukas ahnte nicht...,
dass Paulus am Leben verzweifeln konnte (2 Kor 1.8) und gerade so
das Wunder des Gottes erfuhr, der die Toten erweckt (2 Kor 1.9f.). Er
kennt nur den starken, unerschütterlichen und von Triumph zu
Triumph schreitenden Liebling Gottes’ (Haenchen 680). There is
much truth in this as a general observation, though Paul’s triumphal
 62. THE SEA VOYAGE. 27.1-44. 1179

progress (if such it may be called) from Jerusalem to Rome is set

within a framework of mob violence, imprisonment, attempted
assassination, mocking, and a narrow escape from death by drown-
ing. But there is also a good deal that is thoughtless and out of touch
with reality. Of course Paul ‘steht immer im Mittelpunkt’; Luke’s
history has become a biography of Paul and the subject of a
biography will not surprisingly occupy the centre of the stage. One
could imagine a biography of the centurion Julius, in which Julius, as
responsible for the whole voyage, final authority on matters of route,
protector of all his charges, stood at the centre; and it would be
impossible to find fault with such a biography. Luke admired Paul;
this may have been on his part an error of judgement, but it is hardly
surprising that it determined his presentation of his material. If he
was not mistaken in regarding Paul as an outstanding person this too
is a fact that must be taken into account in estimating the historicity
of Paul’s interventions in the story. ‘Die Unwirklichkeit der Szene
sieht man am leichtesten bei V. 21-26. Paulus hält auf stampfendem
Schiff im heulenden Sturm eine Rede, als stünde er auf dem
Areopag’ (Haenchen 679). Certainly the scene is Luke’s construction
and the speech as we read it is Luke’s composition; but there is
nothing improbable in the influence (not on all 276 on board but on
those who mattered) of one man of unshakable faith and imperturb-
able courage. Parallels are as uncommon as such men are, but they
exist. For Haenchen this speech arises out of Paul’s earlier warning
(v. 10), which he takes to have been based upon the reference in v. 12
to a decision by a majority (οἱ πλείονες) to make for Phoenix. Luke
(according to Haenchen) would think of Paul, the centurion, the
shipowner, and the captain. If in this group of four a majority was
against him Paul must have stood alone. This is a rather mechanical
way of estimating the literary form of a historical event, and
overlooks a not unimportant point. Paul makes two predictions of the
outcome of the voyage (vv. 10; 24-26). They contradict each other;
according to the former, lives will be lost; according to the latter
there will be no loss of life. Did Luke simply invent these contra-
dictory predictions? It seems unlikely. If he did not, he must have
received one or possibly both of them from some kind of tradition of
Paul’s attitude and actions on the ship. This need not have been much
more than, ‘In the course of a stormy voyage Paul gave good advice
and greatly encouraged and cheered the crew and his fellow pas-
sengers.’ That the We in this chapter is not due to a convention of
describing sea voyages in this way is shown in FS A. T. Hanson,
52-6; it is part of a source which may have been little more (see
Introduction, p. xxix) than an itinerary. Luke has written up such
traditions in the light of some acquaintance with the sea and of his
estimate of Paul. Neither the acquaintance nor the estimate is to be
too readily dismissed.
1180 COMMENTARY ON ACTS

This is substantially the conclusion reached in Begs. 4.324: ‘Much

the most natural view is that it really represents the actual experience
of Paul and his friends, but it is possible [the view taken here is that it
is virtually certain] that the narrative has been coloured in a few [a
number of] details by traditional accounts of shipwrecks.’ There are
however few who now hold so high a view of the historical value of
Luke’s story. Bauemfeind (270-2) thinks it possible to remove
‘Pauline supplements’ though he believes that these supplements
contain some old features. Roloff (358f.) thinks that 27.1-9a, 12-20,
27-30,32,38-44; 28.1,11-13,14b, 16b come from Aristarchus, but
‘die Wir-Form erweist sich an zwei entscheidenden Stellen (V. 1 und
V. 6) als sinnwidrig und künstlich.’ Luke has put in vv. 9b-11,
21-26, 31, 33-37 to show the centrality of Paul and that nothing can
stop God’s intention that he should reach Rome. On 27.1-28.15
Schmithals (231) writes, ‘Wir haben es anscheinend mit einem von
Lukas bearbeiteten Stück seiner Paulus-Quelle zu tun.’ Schneider’s
view is similar to that of Haenchen; he also (2.382) quotes Pokorný
237: ‘Die unbestreitbaren Berührungspunkte der Romane mit den
Mysterien sind m.E. vor allem dadurch entstanden, dass man im
Roman versucht hat, die Grundprobleme und Krisen der mens-
chlichen Existenz (Tod, Trennung der Liebenden, Beraubung, Angst
vor den Göttern) durch den Hinweis auf die in den Mysterien
gewonnenen Erlebnisse zu lösen.’ Lüdemann (266) notes the pres-
ence of Paul in vv. 9-11, 21-26, 31, 33-36, 43, and that for the rest
he is forgotten. Paul, Lüdemann thinks, was in Malta, but not
shipwrecked. This looks like simple invention.
Johnson makes the sound comments: ‘[Paul] is not portrayed as a
theios aner ... whose will can bend the forces of nature to his own’
(458) ; ‘Paul’s path is one marked not by thaumaturgy but by faith’
(459) . If to faith we add courage and hardihood (amply attested by 2
Cor. 11.23-27), some at least of the details of this chapter become
credible—more credible than some of the objections that have been
quoted above.

1. That Paul should sail to Rome was already implied in the

acceptance of his appeal to the emperor (25.12, ἐπὶ Καίσαρα
πορεύση). ὡς (the temporal use of ὡς is characteristic of Acts; see
16.4) δὲ ἐκρίθη will refer therefore to a decision with regard to the
practical arrangements for the journey. For the (unnecessary) τοῦ
before the infinitive see BDR § 400.7, n. 9, where its frequency in
Lk. and Acts is said to be due to the influence of the LXX, in which it
represents the Hebrew ἡμᾶς means, prima facie, that the writer
accompanied Paul on the journey; for the origin and significance of
the ‘We-passages’ see Introduction, pp. xxvi-xxx. The first person
plural disappears in the reading of P 6 326 2495* pc, which instead
of ἡμᾶς have τοὺς περὶ τὸν Παῦλον, Paul and his party. These MSS
 62. THE SEA VOYAGE. 27.1-44. 1181

do not consistently remove the first person plural from the narrative,
so that there is no attempt to suggest that the author of Acts was not
present, and it is not easy to see how the reading originated; perhaps
ἡμᾶς was thought to lack clarity and directness, ἀποπλεῖν is (for
obvious reasons) used several times in the accounts of Paul’s travels;
see 13.4; 14.26; 20.15.
παρεδίδουν (for the classical –εδίδοσαν) is Hellenistic; for the
tense see on v. 2. The subject is not expressed; in the background lies
the authority of the governor Festus, who acts through his officials
(hence the plural verb). They handed over both Paul and some other
prisoners. The reversal of τόν τε in P74 pc would lead to both handed
over and ...—but there is no second verb. We hear of the other
prisoners again at v. 42, not however in ch. 28. For ἐκατοντάρχης
see on 10.1. Nothing more is known of the Centurion Julius than is
contained in the present narrative, σπεῖρα Σεβαστή translates cohors
Augusta, and there is good epigraphical evidence for the presence of
a cohors Augusta I in Syria in the first century (see T. R. S.
Broughton in Begs. 5.443, quoting H. Dessau, Inscriptiones Latinae
Selectae I (1892), 2683). Broughton rightly rejects the suggestion
that there is a reference to the cohortes Sebastenorum, or cohorts of
Samaritans raised in Sebaste (= Samaria). Augusta (Σεβαστή) was
an honorary title. On this see also NS 1.364. The cohortes Augustae
were Syrian auxiliaries, and Broughton expresses surprise that a
legionary centurion was not found to carry out the task of conveying
an important prisoner to Rome. Perhaps (assuming that the story is
not simply fictitious) Julius was the best man available; or was the
prisoner not thought as important as Broughton assumes? or (since
the appeal was something of a problem) was there a hope that the
prisoner might escape on the way?
For the genitive σπείρης (instead of σπείρας) cf. 10.1 and see
BDR § 43.1, n. 1.
The first nine words of the verse (ὡς ... ’Ιταλίαν) are omitted by h
syP syhmg. These MSS have an addition at the end of 26.32; see on that
verse.
The use of ετέρους implies that there were only two categories of
prisoner, Paul and the rest (Delebecque 128).

2. The imperfect παρεδίδουν is brought to a point in the aorists

ἐπιβάντες ... ἀνήχθημεν. The process of handing over the prisoners
concluded with the actual embarkation, at which point, presumably,
Julius assumed full responsibility, ἐπιβ. ἀνήχ. is used at 21.2, with
reference to a ship sailing to Phoenicia. For the verb with a dative cf.
Thucydides 7.70.5, ἐπειρῶντο ταῖς ἀλλήλων ναυσὶν ἐπιβαίνειν.
Here P74 places τω before πλοίῳ, pointlessly, because we know
nothing of this ship until we read the following words; 614 2495
pc supply ἐν, needlessly—see e.g. the above quotation from
1182 COMMENTARY ON ACTS

Thucydides. The ship’s home port was Adramyttium, and it was used
no doubt for coastwise traffic round the north-eastern comer of the
Mediterranean. It is described by the adjective Άδραμυττηνῷ, for
which there is a respectably attested variant Άδραμυντηνῷ (P74vid A
B* 33 pc). On the variant spelling see Μ. 2.106; BDR § 42.3, n. 3
(‘nasaler ersatz von ττ durch ντ’, with a reference for such substitu-
tions to § 39.7, n. 11). It was probably too much to hope to find a ship
sailing direct to Rome; to reach a port in Asia would take the
travellers well on their way and give them a fair chance of finding
another ship that would cover the rest of the journey. Julius cannot
have intended to take his prisoner ‘to Adramyttium, and then through
the Troad, across the Hellespont and along the via Egnatia to
Dyrrachium, and so by Brundisium to Rome’ (Page 252). Cf. Hemer
(133).
εἰς τοὶς ... τόπους, εἰς is omitted by H L P S and the Byzantine
text (Ropes); the omission ‘must have been deemed good Greek in
the fourth and following centuries, although only in Greek poetry are
parallels found to this usage’ (Begs. 3.240).
Paul’s only named companion is Aristarchus; see 19.29 (named
with Gaius, both described as Macedonians) and 20.4 (named with
others and Secundus, both described as Thessalonians). Here, for
Aristarchus only, both designations are used. Lightfoot (Philippians
35) thought that he left the party at Myra (v. 5), continuing (when the
others changed) in the same ship as far as Adramyttium on his way
home to Thessalonica. Chrysostom (Homily 53.1) thought that he
was carrying news of Paul to the churches of Macedonia. He is not
heard of again in Acts and we do not know what happened. It is
impossible to build on Col. 4.10, but the closely parallel reference in
Philemon 24 shows him with Paul at a time of imprisonment; where
and when however are matters of dispute. Conzelmann (141) takes
him as a pointer to the way in which Luke’s information reached
him. Secundus is introduced here (from 20.4) by 614 (2147) 2495 pc
syh.
Begs. 4.325 notes that in this context Asia must mean the western,
not the southern, coast of Asia Minor. This is correct, but it is clear
that the ship was putting in at south coast ports too.
Metzger (496f.) quotes the reconstruction made by Clark (163), on
the basis of 97 421 syp syhmg, of the Western text of vv. 1, 2:
οὕτως οὖν ἔκρινεν ὁ ἡγεμὼν ἀναπέμπεσθαι αυτόν Καίσαρι.
καὶ τῇ επαύριον προσκαλεσάμενος ἐκατοντάρχην τινὰ
ὀνόματι ’Ιούλιον, σπείρης Σεβαστής, παρεδίδου αὐτῷ τόν
Παύλον σὺν ἑτέροις δεσμώταις. ἀρξάμενοι δέ τοῦ ἀποπλεῖν
εἰς τὴν ’Ιταλίαν ἐπέβημεν πλοίῳ ...
Ropes (Begs. 3.240) thinks that this reading was intended to ‘relieve
the abruptness of the B-text’.
 62. THE SEA VOYAGE. 27.1-44. 1183

3. The ship sailed north along the coast, aided by the current due to
the outflow from the Nile, and on the following day Sidon was
reached. With τῇ ἐτέρα, ἡμέρα must be understood, as at Sophocles,
Oedipus Rex 781f., τὴν μὲν οὖσαν ημέραν μόλις κατέσχον, θἀτέρα
δέ ... Xenophon, Cyropaedia 4.6.10 is somewhat unfortunately
chosen by BA (638) as an example, for here the expression follows
on αὔριον (as at Acts 20.15 it follows on τῇ ἐπιούση) and
accordingly means on the day after tomorrow, on the third day.
Always however it means on the day after the one last mentioned;
here, on the day after leaving Caesarea. According to Marshall the
distance was about 69 nautical miles.
κατήχθημεν corresponds to ἀνήχθημεν (v. 2): we put out to sea,
we put in to land. Sidon, a Phoenician port, was a hellenized city; in
24 BC Augustus made it part of Ituraea. There is some evidence for
the presence of Jews (NS 3.14,15). If τούς φίλους refers to Christian
friends (see below) it will follow that Sidon had been evangelized
and a church established there. We have no other evidence for this at
so early a date. Eusebius (HE 8.13.3) mentions as a martyr Zenobius,
a presbyter of the church at Sidon.
Julius, another of Luke’s ‘good’ centurions (cf. Lk. 7.2; 23.47;
Acts 10.1), showed Paul kindness, φιλανθρωπία was a recognized
virtue in the ancient world. See Plato, Euthyphro 3d; Demosthenes
19.225 (411); Plutarch, Aemilius Paulus 39 (276); lamblichus quoted
in Stobaeus 4.5.76 (Gaisford 46[44], 76), ὅτ’ ἄν χρηστότητι καί
φιλανθρωπία κραθῇ το σεμνόν καὶ αυστηρόν .... For χρῆσθαι cf.
Xenophon, Memorabilia 1.2.48. The centurion permitted Paul to go
πρὸς τοὺς φίλους, to his friends, and these would naturally be
Christians. It is possible (see Begs. 5.379f.) that ‘the Friends’ was
already a term denoting Christians; cf. 3 Jn 15, though here too the
word does not necessarily, and indeed does not probably, have a
technical sense. See G. Stählin in TWNT 9.144-69, especially 159f.
Paul was allowed to go to the friends in order that he might
ἐπιμελείας τυχεῖν. επιμέλεια is care of all kinds: see Prov. 3.8; 1
Macc. 16.14; 2 Macc. 11.23; 3 Macc. 5.1; Josephus, Ant. 2.236, the
young Moses was in his upbringing in Egypt πολλής ἐπιμελείας
τυγχάνων. In the present passage Luke no doubt thinks of what
Ignatius (Polycarp 1.2) describes as ἐπιμέλεια σαρκική τε καὶ
πνευματική. Paul’s Christian friends would give him both.

4. How long the ship stayed with its passengers in Sidon Luke
does not say. In due course they again put out to sea (ἀναχθέντες; cf.
ν. 2). ὑπεπλεύσαμεν τὴν Κύπρον, sailed under (the lee of) Cyprus.
Cf. ὑποδραμεῖν (vv. 7, 16). This is clear, except that it does not tell
us which was the lee side of Cyprus. This depends on the direction of
the prevailing wind. There is a long note with many nautical and
meteorological details in Begs. 4.326. but the only consideration that
1184 COMMENTARY ON ACTS

comes near to being decisive is the statement in v. 5 that they sailed

through the Cilician and Pamphylian seas to Myra. This can hardly
be otherwise interpreted than to mean that they kept to the north of
Cyprus. This, unless the ship had trading connections with Salamis
and Paphos, would in any case be the more probable course in an age
in which shipping hugged the coastline as far as possible.
The winds were contrary (for the expression cf. Mk 6.48 = Mt.
14.24). If they had sailed across in a north-westerly direction they
would have had the Etesian winds in their teeth. Nearer the coast of
Asia Minor they would have been helped by land winds and a
westward-flowing current (Page (252) referring to Smith 67 (better
would be 67-9)). See also Hemer (133).
In 21.3 Paul sailed on the other side of Cyprus.

5. πέλαγος (in the NT elsewhere only at Mt. 18.6) is the high sea,
the open sea, in contrast with landward water. Luke means (and this
is consistent with his remark about the wind—see v. 4) that instead
of keeping to the coast of Cilicia and Pamphylia they sailed straight
across the Pamphylian Bight to arrive at Myra (see the Western text
of 21.1), here correctly described as belonging to Lycia, the use of
the partitive genitive τῆς Λυκίας being also correct (BDR § 164.3, n.
7; Μ. 3.171). For διαπλεύσαντες cf. Thucydides 4.25.1, ἠναγ-
κάσθησαν ὀψὲ τῆς ἡμέρας ναυμαχῆσαι περί πλοίου διαπλέοντος;
Lucian, Hermotimus 28, ... τόν Αἰγαῖον ή τόν Ίόνιον διαπλεῦσαι
θέλοντας.
κατέρχεσθαι (cf. κατάγεσθαι in ν. 3) is the regular term for
‘coming down from the high sea’ to land.
The Western text (represented by 614 2147 pc h vgmss syh**) adds
that the voyage lasted δι’ ημερών δεκαπέντε (for διά and the
genitive expressing extent of time see BDR § 223.2b, n. 4). This was
the time it might have been expected to take; cf. Lucian, Navigium 7,
ἐκεῖθεν [from Sidon] δέ χειμῶνι μεγάλω δεκάτη έπὶ Χελιδονέας
[on the coast of Lycia]... έλθεῖν. Ten days with a strong wind might
easily become fifteen in different conditions. The addition shows
knowledge of local conditions. ‘Unless in the “Western” text the
additional words... are regarded as genuine and accidentally omitted
from the B-text, no explanation is at hand’ (Ropes, Begs. 3.241).
Clark (xlvi) observes that the words would form a στίχος. Begs.
4.326 accepts them as genuine, because they are so good an estimate.
Alternatively, and more probably, they were added by someone who
knew the area at a time when the text of Acts could be handled freely,
and copyists felt at liberty to add what they knew (or believed) to be
true on the basis of their own knowledge.
The port reached at the end of the first stage of the voyage is
given in different forms. The apparatus in NA26 does not distinguish
between Μύρ(ρ)α (neuter plural) and Μύρ(ρ)αν (feminine singular).
 62. THE SEA VOYAGE. 27.1-44. 1185

The neuter plural is more commonly used. Μύρα is read by Ψ <a>

h—a not very distinguished group. B 1175 have Μύρραν. In
addition 69 has Σμύρναν, Ρ74 lat bo have Αύστραν (and A has
Αύστρα). Σμύρνα may be a corruption cf Μύρρα. Begs. 4.327
points out that Myra was also called Άδμυρὰ, Λάμυρα, and Λιμυρά
(or Λιμύρα), and Αύστρα may be derived from one of these. There is
no doubt that Myr(r)a is meant. ‘Alexandrian com ships could rarely
sail directly to Italy from Egypt, but a west wind, or even one slightly
north of west, made it possible to fetch Myra, and thence a north
wind would take them to Sicily, from which another shift of wind
back to the west would make it possible to reach Pozzuoli, or even
Ostia’ (Begs. 4.327). Cf. Hemer 133f.
6. The Alexandrian ship sailing to Italy would probably belong to
the com fleet (see v. 38). Egypt was an indispensable source of supply
for Rome, providing a third of the com used in the year: Josephus,
War 2.386 (παρέχει... τῇ 'Ρώμη σῖτον μηνών τεσσάρων; 383—the
remainder came from Africa). Many of the ships were run by private
enterprise (Suetonius, Claudius 18,19; cf. Seneca, Epistle 77.1f.), but
they were used by official travellers as well as by private persons.
Thus Titus, returning from Jerusalem to Rome, ‘festinans in Italiam,
quum Rhegium, deinde Puteolos oneraria nave appulisset, Romam
inde contendit expeditissimus’ (Suetonius, Titus 5). Some ships were
large; the ship in Lucian’s Navigium 5 was 120 cubits (=180 feet =
54 m.) long, more than 30 cubits (=45 feet = 13.5 m.) wide, and 29
cubits (=43.5 feet = 13.05 m.) deep. Alexandrian sailors had a good
reputation: Philo, Flaccus 26: τάς τε γάρ ἐκεῖθεν ὁλκάδας
ταχυναυτεῖν ἔφασκε καὶ ἐμπειροτάτους εἶναικυβερνήτας ...
ἐμβιβάζειν is regularly used for the transitive sense of ‘to
embark’, ἐμβαίνειν serving for the intransitive. Cf. Lucian, Verae
Historiae 2.26, ἐμβιβάσας ὁ 'Ραδάμανθυς πεντήκοντα τῶν ἡρώων
εις ναῦν ... Cf. Hemer (134).
7,8. The next stage of the journey is dealt with in a long sentence
which runs through the two verses and has a complicated structure.
Moule (IB 101) notes that the present participle βραδυπλοοῦντες is
followed by the aorist indicative ὑπεπλεύσαμεν, and raises the
question whether the participle refers to action before that of the
main verb or covers the whole episode. In fact the sentence is more
complicated than this and it will be necessary to set out its
framework as a whole: Present participle (βραδυπλοοῦντες); aorist
participle (γενόμενοι); present participle (προσεῶντος); aorist indic-
ative (ὑπεπλεύσαμεν); present participle (παραλεγόμενοι); aorist
indicative (ἤλθομεν). When these are considered in the light of a
map, it seems to make sense both of the grammar and of the
geography to suppose that the first three participles describe the
journey as far as Cnidos, ὑπεπλεύσαμεν takes the ship as far as
1186 COMMENTARY ON ACTS

the south coast of Crete, the next participle and finite verb to Fair-
Havens. The ship could sail only slowly (Begs. 4321 suggests that
βραδυπλοεῖν may be a technical term for beating, as ἐυθυδρομεῖν
(16.11; 21.1) may mean to run) in a direct WNW (Etesian) wind
from Myra to Cnidus; it then took advantage of the N or NE wind to
cross over to Crete with a view to taking the sheltered south side of
the island (ὑπεπλεύοαμεν, a sort of inceptive aorist—we took
shelter under the lee of Crete). Even so, however, it was with
difficulty that they coasted half way along the island and eventually
reached (ἤλθομεν) Fair Havens.
ἐν is employed, unusually, to denote the extent of time during
which an action takes place, ἱκανός is a favourite word in Acts to
denote an interval which Luke either is not able, or has no wish, to
give precisely: for a fair number of days.
βραδυπλοεῖν occurs nowhere else in the NT, and is not common
in Greek generally. Its broad meaning is not in doubt (see above),
and the reason for the slow progress it denotes is clear: the wind did
not allow speed, perhaps did not permit the ship to approach Cnidus,
but the compounded πρός (in προσεῶντος) may do no more than
strengthen the meaning of the verb. For the expression cf. Lucian,
Verae Historiae 1.29, οὐ μέντοι ἐπέβημεν αυτής, ού γάρ εἴα τό
πνεύμα, but (since αὐτῆς refers to Νεφελοκοκκυγία) the sense is
somewhat different.
μόλις occurs twice in these two verses; also at 14.18; 27.16; Lk.
9.39 (ν. 1. μόγις); elsewhere in the NT only twice—a Lucan word;
perhaps one should say a word characteristic of the sources of the
present story.
Cnidus is mentioned here only. It lay at the extremity of a peninsula
of the mainland, north west of Rhodes. Since 129 BC the town had
been a civitas libera under Roman rule. See Trebilco (123).
ὑπεπλεύοαμεν was used at v. 4, where it must mean, We sailed
under the lee of. Here, after the reference to Cnidus and followed by
κατά Σαλμώνην, it must mean something like we took refuge, sailed
for refuge (from the troublesome wind) under the lee of Crete. Twice
in these verses (cf. also v. 5) κατά will mean off: they did not call at
Cnidus, but approached it, and reached Crete off Salmone, the
headland at the northeastern extremity of the island. There is some
doubt about the spelling of Salmone (several versions are known,
though there seems to be no variation in the text of Acts), and also
about its precise location; there are two capes, and it is uncertain
which is referred to. Crete itself was brought under Roman rule in 67
BC by Q. Metellus; it was united with Cyrenaica and made a
senatorial province.
Even the sheltered south side of Crete presented difficulties to the
navigators, and it was with difficulty (μόλις) that they coasted
(παραλεγόμενοι) as far as a place called Fair Havens. A port bearing
 62. THE SEA VOYAGE. 27.1-44. 1187

this name still exists; it is sheltered on the west, open to the east, and
would therefore give protection from northerly and westerly winds. It
was about five or six miles from a place that has been identified with
(and is said to be still called by local peasants) Lasaea. Apparently
the place had little but safety to commend it. For the language cf.
Cicero, Ad Familiäres 14.5 (236a), Athenas venimus, cum sane
adversis ventis usi essemus, tardeque et incommode navigassemus.
εγγύς is here constructed with the dative, as at 9.38; at 1.12 it is
used with the indeclinable ’Ιερουσαλήμ. Elsewhere in the NT where
the case is determined it is always genitive (notwithstanding the
statement in Μ. 3.216).
According to Titus 1.5 Paul left Titus in Crete; this epistle
presupposes the existence of a fairly developed and numerous church
in the island. There is no confirmation of these things in the present
narrative. However considerately Julius treated Paul he will hardly
have permitted him to set about evangelizing the island.
The name of Lasaea is variously given: Αασαία ((*) Ψ <a>);
Λασέα (B 33 1175 1739 1891 2464 al); Λασία (36 81 453 945 pc);
Λαίσσα (Κ2); Άλασσα (A syhmß sa); Thalassa (lat).
h has the curious reading, devenimus in portum bonum, ubi
Anchis ci[vitas er]at. Ropes (Begs. 3.243) suggests that Anchis may
be a misunderstanding of ἄγχι, which the Western text may have
used instead of εγγύς. There is no better suggestion.
Hemer (134-6, cf. Smith 75f.) in a long note argues, in depend-
ence on L. Casson (see Bibliography, p. 1177) that it was usual, at
least for sailing ships, to take a course to the south of Crete. This may
now be taken as established.
It is often held that v. 8 (or v. 9a) connects with v. 12, vv. 9-11
being a Pauline insertion in a non-Pauline account of a voyage. See
p. 1191; also Schille (461), who takes the view that vv. 9-11 do not
interrupt the sequence of the narrative.

9. In Luke’s narrative as it stands it is clear that the continuation of

the voyage was under discussion, clear and also understandable. A
considerable time (ικανού χρόνου; cf. v. 7) had now elapsed (since
they left Caesarea? or since they left Myra?), and Fair Havens was
not a desirable place in which to spend the winter. Claudius however
had taken steps to make even risky winter voyages for the com fleet
profitable (Suetonius, Claudius 18). On the other hand the danger of
sailing was sufficient to constitute a good case for remaining where
they were. In this situation, Paul (according to Luke) intervened. Is it
likely that he would be in a position to do so? Prima facie one might
think not; but it must be remembered that though under guard, and
not free, he was in a sense a privileged person, who must be
delivered to the Emperor. He had not been found guilty of any crime,
and it was the opinion of the governor that he was innocent.
1188 COMMENTARY ON ACTS

The first two statements in the verse are made in absolute

genitives, the latter expanded by διὰ τό (the use of the article is
characteristic of Acts) and an accusative and infinitive: because the
Fast had now already passed. The Fast is the Day of Atonement, the
only fast in the Jewish calendar. 'On the Day of Atonement, eating,
drinking, washing, anointing, putting on sandals, and marital inter-
course are forbidden’ (Yoma 8.1). Many other passages connect the
Day of Atonement with fasting, e.g. Josephus, Ant. 17.165: ... μίαν
ἡμέραν, ἥν ’Ιουδαῖοι νηστείαν ἄγουσιν; cf. 14.66; Philo, Moses
2.23; De Specialìbus Legibus 1.186; 2.193: μετά δέ τήν τών
σαλπίγγων άγεται νηστεία ἑορτή; Gaius 306. The Day of Atone-
ment fell on 10 Tishri; Tishri corresponds to the latter part of
September and the former of October. The 10th would fall at the end
of September or the beginning of October. By this time sailing was
unsafe, ἐπισφαλής (here only in the NT, but not an uncommon
word). Cf. Josephus, Ant. 16.15: τόν γάρ πλοῦν, ἐμβαίνοντος τοῦ
χειμῶνος, οὐκ ἐνόμιζον ασφαλή. According to Vegetius, De re
militari 4.39 sailing became dangerous after 15 September and
ceased after 11 November. The reference to the Fast has been used as
a means of dating the journey. The Fast, τήν νηστείαν, is introduced
by καί which should mean The Fast too, or even the Fast. This (see
W. P. Workman, ExpT 11 (1900), 316-19 and many commentaries,
e.g. Begs. 4.328f.; see also J. Behm in TWNT 4.928-32) has been
held to mean that this year the Fast (= 10 Tishri) fell late, later at
least than the autumnal equinox. In AD 59 the date was 5 October; it
was earlier in 57, 58, 60, 61, 62. From this some have inferred that
the date of Paul’s journey was AD 59. It is true that problems arise at
28.11 (see the note there); the voyage to Malta must have ended, and
the stay in Malta begun, by about the end of October; the last part of
the voyage, after three months, will therefore have begun at the end
of January, that is, in the period when sailing would be considered
impossibly dangerous. Perhaps in view of Claudius’s encouragement
of the com trade this matters little. But it seems that a great deal is
being made to rest on the word καί, to which there is no need to give
this quasi-superlative force. It was late; even the day usually
considered as marking the end of navigation had gone by. There must
have been an uncertain period; p. Shabbath 2.5b.25 reckons sailing to
be unsafe after the Feast of Tabernacles, 15 Tishri. On this reckoning
the Day of Atonement was late but not impossibly late. See Marshall
(406).
Hanson (245) speculates on where Paul had observed the Day of
Atonement, whether on shipboard or at Lasaea (if there was a
synagogue there). It might be more useful to ask whether he would
have observed it at all; see Gal. 4.10.
In view of the situation, Paul offered advice, παρῇνει. One might
have expected the verb παρακαλεῖν; παραῖνειν is used in the NT
 62. THE SEA VOYAGE. 27.1-44. 1189

only here and at v. 22. Like μόλις (above) it could be a mark of a

special source for this chapter; it is not a word particularly appro-
priate to a sea voyage—and it does nothing to confirm the view that
we have here a Lucan insertion. Since the prediction that follows in
ν. 10 was proved false we may suppose that Luke (if he thought
about the matter) took it to be Paul’s human opinion, whereas his
later prediction (vv. 22-26) was based on a supernatural communica-
tion from an angel.

10. Paul addresses the company. The vocative ἄνδρες is in nearly

every other place in Acts accompanied by some other noun (such
as αδελφοί); here, and at vv. 21, 25, it is alone, probably because
there was no other word suitably descriptive of all Paul’s fellow
travellers.
The sentence that follows the vocative has a mixed construction. It
begins θεωρῶ ὅτι; this should be taken up by a finite verb, but in fact
the sentence continues with an accusative and infinitive (μέλλειν
ἔσεσθαι τόν πλοῦν), which should have followed upon θεωρώ
without ὅτι. The construction has however been defended. Μ. 1.213
says that there are classical parallels (from various sources the
following may be noted: Xenophon, Hellenica 2.2.2,... εἰδὼς ὅτι...
θᾶττον τῶν ἐπιτηδείων ἔνδειαν ἔσεσθαι; Cyropaedia 8.1.25; Thu-
cydides 5.46.3; Plato, Phaedo 63c; we may add POxy 2.237.col.5.8,
δηλῶν ὅτι εἰ τὰ ἀληθή φανείη μηδὲ κρίσεως δεῖσθαι τὸ πράγμα).
Blass (274): ‘sine ὅτι erat periculum ne coniungerentur θεωρώ μετά
κτἑ, et idem ὅτι multis interiectis facile e mente elapsum est.’
Delebecque (130) calls ὅτι with the infinitive a ‘tour classique’,
which perhaps stretches the evidence somewhat. To Knowling (520)
it is ‘a vivid dramatic touch’. See also Begs. 4.329; H. J. Cadbury in
JBL 48 (1929), 419f. Cf. Acts 16.19 D.
The future infinitive (found in the NT only in Acts and Hebrews)
is rightly used with μέλλειν.
Paul foresees that the voyage if persisted in will be μετά ύβρεως
καί πολλής ζημίας. ὕβρις (also at ν. 21) is an unexpected word. LS
1841, in addition to giving the usual meanings of the word (wanton
violence, insolence, lust, lewdness, an outrage, especially on the
person) have a third division, ‘III used of a loss by sea' for which
they cite only our passage and Pindar, with a cross-reference to
ναυσίστανος. Here (LS 1162) ναυσίστανος ὕβρις is rendered ‘the
lamentable loss of the ships', with reference to Pindar, Pythian Odes
1.72 (ναυσίστανον ὕβριν ἰδών τἀν πρὸ Κύμας). Their interpretation
may not be correct. B. L. Gildersleeve (Pindar: the Olympian and
Pythian Odes; 1885) explains the sentence as equivalent to ὅτι ή
ὕβρις ή πρὸ Κύμης ναυσίοτονος (sic Gildersleeve) ἐγένετο, and
writes (p. 249), ‘There is no Pindaric warrant for the use of ὕβρις as
“loss”, “damage”. The reflection that their overweening insolence
1190 COMMENTARY ON ACTS

[ὕβρις] off Cumae had brought groans and lamentation to the ships
(cf. P[yth]. 2.28) would silence their savage yell and keep them quiet
at home.’ Luke’s repeated use of the word however suggests that he
had reason to think it appropriate, and there is ground for this. See
BA 1660, with e.g. Angiologia Palatina 7.291.4 (7.29.3f.), δείσασα
θαλάττης ὕβριν (a personal outrage inflicted by the sea); Josephus,
Ant. 3.133, τὴν ἀπό τών όμβρων ὕβριν. ζημία gives rise to no
problem: ‘damage and much loss not only of the cargo and the ship
but also of our lives’. For the last five words contrast v. 24.
Bruce (2.482) counters the argument that it is unlikely that Paul
the prisoner would be allowed to attend and address a ship’s council
with the observation that Paul was on good terms with the centurion
and had in the first instance given him his opinion. But Paul’s advice
is given not to the centurion but to ‘ἄνδρες’. For the advice cf. that of
Apollonius in Philostratus, 5.18: ἀποβώμεν, ἔφη, τής νεὼς ταύτης,
οὐ γὰρ λῷον αυτή ἐς Άχαίαν πλεῦσαι. Only Apollonius’s friends
heeded his warning; the ship sank.

11. Julius’s favourable treatment of Paul did not extend to

preference for his advice over that of the professionals, who
evidently counselled continuation of the voyage as far as Phoenix.
Whether it was this short voyage, or the whole journey to Italy, that
Paul had in mind in v. 10 is not clear.
The κυβερνήτης was the steersman, and the primary and ulti-
mately responsible steersman was the captain of the ship. The word
is used in this sense; so e.g. Plato, Republic 341cd, τί δέ κυβερνήτης;
ὁ ὀρθῶς κυβερνήτης ναυτών ἄρχων ἐστιν ἢ ναύτης; ναυτών
ἄρχων. A special use of the word is for the captain of a Nile-boat (LS
1004). The role of the ναύκληρος is not clear. In some contexts the
word denotes the ship-owner (who would certainly even if not
travelling be very interested in the possibility of shipwreck). This
may well be the meaning here. Many of the ships in the com fleet
were in private ownership (see v. 6); some however were not, and in
such a case the ναύκληρος would presumably be the man who
represented the ‘owner’—that is, the official responsible for the fleet.
It may be said that the κυβερνήτης and the ναύκληρος would
represent respectively the nautical and the financial interest. If these
agreed it would be most unlikely that the centurion would accept the
contrary view of Paul. There is an important note on ναύκληρος in
MM 422f.; see also BA 1081. A passage in Plutarch, Praecepta
Politica 13 (807b), is sometimes quoted as if it settled the relation
between ναύκληρος and κυβερνήτης: ναύτας μὲν ἐκλέγεται
κυβερνήτης καί κυβερνήτην ναύκληρος. This is however a very
difficult passage. It is introduced by the words δεινόν γάρ ὡς
ἀληθῶς καί σχέτλιον, εἰ ναύτας μὲν ἐκλέγεται, κτλ. See Hemer
(138f.).
 62. THE SEA VOYAGE. 27.1-44. 1191

Conzelmann (141) notes that the κυβ. and the ναυκ. are not
mentioned in the account of the wreck and infers that the present
verse is an insertion. Stählin (316) thinks that two sources may have
been combined, since v. 11 suggests a Führungsgruppe, v. 12 a
democratic decision among all (or at least all free) travellers.
The centurion was persuaded by the captain and owner (or man-
ager) rather than by what Paul said; in fact, he did what they said, and
did not do what Paul advised. Zerwick (§ 445) notes the use of μάλλον
... ἤ in an exclusive sense, and adduces in addition to this passage Jn
3.19; 12.43; Acts 4.19; 5.29; 27.11; 1 Tim. 1.4 (correcting Zerwick’s
reference); 2 Tim. 3.4; Heb. 11.25. BDR § 246.2a, n. 4 add Acts 20.35,
and note the parallel in Plato, Apology 29d, πείσομαι δέ μάλλον τω
θεῷ ἢ ὑμῖν, which means, I will obey God and I will not obey you. It
is correct that Luke’s interest is in the fact that in face of all opposition
Paul is proved right by events (Conzelmann 142).
Opinions of the historicity of this verse and of the whole paragraph
of which it is part differ. Some think it impossible that the opinions
of a prisoner should even be considered. Haenchen (670) makes
much of Paul as a schwerverdächtiger Gefangener; he is not so
represented in Acts. Bruce (2.482) quotes Nock (Essays 2.823): it is
‘an authentic transcript of the recollections of an eyewitness, with the
confusion and colouring which so easily attach themselves to
recollections’.

12. This verse could be connected with v. 8 or v. 9a; hence the

suggestion that vv. 9(b)—11 are an insertion; see above.
The sentence opens with a genitive absolute which states the main
fact from which the discussion and disagreement arise. The harbour
at Fair Havens was inconvenient (ἀνεύθετος: here only in the NT,
but εύθετος occurs at Lk. 9.62; 14.35; also Heb. 6.7) for wintering
(παραχειμασία: here only in the NT; see MM 491; ND 4.166). What
made it inconvenient is not stated: perhaps insecure anchorage for
the ship; perhaps a lack of social amenities for the crew.
This being so the majority (οἱ πλείονες) formed a plan, εθεντο
βουλήν. For this expression cf. Judges 19.30; Ps. 12( 13).3; Plutarch,
Galba 4 (1054), προὔθηκε βουλήν τοῖς φίλοις. προτίθεσθαι β.
seems to be more common than τίθεσθαι β. Who were the majority?
The majority of the whole ship’s company? The captain and owner
against Paul? See on v. 11. Parallels with the emperor’s council,
sometimes drawn, are pointless. In a difficult and doubtful situation
there is no need for imperial precedent; an intelligent man like Julius
knows that it will be wise to call together the interested and
knowledgeable parties and find out their views. The majority view is
that they should not stay in Fair Havens, equally that they should not
attempt the long voyage to Rome, but should make for Phoenix, a
little way along the coast.
1192 COMMENTARY ON ACTS

There was doubt about even this objective; they were to get to
Phoenix εἴ πως δύναιντο. Μ. 3.127 says that such uses of εἰ ‘are not
so much real conditions as final clauses’, and adds Acts 17.27. Others
recognize the uncertainty imported into the clause by πως and the
optative: Moule (IB 154), the optative gives to the indirect question
‘a more tentative and cautious tone’; Zerwick (§ 403); Radermacher
(131), ‘Die Partikel εἰ leitet, parallel unserm “ob”, einen Satz ein,
der den erwarteten Erfolg einer Ueberlegung angibt.’ BDR 375, n. 4
compare Xenophon, Anabasis 4.1.21, εἴ πως δυναίμην. Delebecque
(130) speaks of the indirect question depending on a verb of action
(ἀναχθήναι) as a ‘... tour, constant depuis Homère’. εἰ means
‘ “pour voir si”, “au cas où”, “dans l’espoir de’”.
The hoped for objective was Phoenix, the precise location of
which is disputed. The name, with minor variations in spelling, is
found in ancient authors, and the approximate location of the town is
known. It was about 50 miles west of Fair Havens (Stählin 316), and
it can be deduced from Ptolemy, Geographia 3.15 [3.17.3] (see Begs.
4.330) that it was about 34 miles east of the western extremity of
Crete. It has been customary to identify its harbour with that now
known as Loutro (see W. J. Woodhouse in Ency. Bib. s.v. Phenice,
3690-92, for a good account of earlier discussion; also Smith
(87-93)), which has been described as the best harbour on the
relevant part of the southern coast of Crete. Difficulty however arises
in the description of the harbour as βλέποντα κατά λίβα καί κατά
χώρον. The meaning of λίψ is clear (though it has other uses, e.g. in
the LXX); it means the south west wind. χῶρος is not a native Greek
word: it transliterates the Latin Caurus or Corus (e.g. Vergil,
Georgies 3.356, Semper hiems, semper spirantes frigora Cauri;
Caesar, De Bello Gallico 5.7, Corus ventus navigationem impedie-
bat). This was a north west wind. βλέπειν κατά is most naturally
taken to mean to look into a given direction; here to look into the
direction of Lips and of Corus, that is, with the sort of approximation
inevitable when directions are given with no greater precision than
the names of winds afford, to look west. Hence an impasse: the
harbour Loutro looks east. No attempt to make βλέπειν κατά λίβα
καί κατά χώρον mean something different can be judged successful.
Cf. Caesar’s spectare in De Bello Gallico 7.69 (Sub muro, quae pars
collis ad orientem solem spectabat ...). It is hard to believe that
βλέπειν κατά means to look down wind, hard to believe that βλέπειν
implies a sailor’s view as he enters harbour—entering an east facing
harbour he is looking west. Sailors leave harbour as often as they
enter. RV’s ‘looking NE and SE’ means taking either κατά λίβα καί
κατά χώρον rightly but βλέποντα wrongly, or taking κατά λ. καί κ.
χ. wrongly but βλ. rightly (Page 254f.) The right explanation is
almost certainly that given by Begs. 4.330 (see the sketch map there).
The harbour Loutro is formed by a promontory, Muros, projecting
 62. THE SEA VOYAGE. 27.1-44. 1193

south from the coast of Crete. On the other side of the harbour is
another harbour whose modern name is Phineha—almost certainly a
modification of Φοῖνιξ. Phineha faces west as clearly as Loutro
faces east. It has been objected that Phineha offers no suitable
anchorage. This objection (with some others) is dealt with in an
article by R. Μ. Ogilvie (JTS 9 (1958), 308-14). Astonishingly, he
refers to commentaries by Bruce and Haenchen but makes no
reference to Begs. 4, though he supplies the archaeological and other
evidence the authors there desiderated. Since Paul’s time there have
been notable geological changes in Crete, and there is no difficulty in
supposing that in the first century the bay of Phineha could be used
by wintering vessels. It may now be regarded as virtually certain that
it was the protection of the western harbour that the centurion and his
professional advisers sought. It was in the end a north-easterly wind
that destroyed the ship (v. 14). On the winds see much detail in Begs.
5.338-44; also Hemer (139-41).
Did Luke at this point recall Hesiod, Works and Days 618-630
(Delebecque 130)? It seems unlikely.
13. Again a sentence begins with a genitive absolute, ὑποπνεῖν is
to blow gently, ‘ὑπό in composition = “slightly”, cf. Homer, Iliad
4.423, Ζεφύρου ὑποκινήσαντος’ (Page 256; OCT prints separatimi
ὕπο κινήσαντος). A light south wind was exactly what the seamen
wanted; it would enable them to reach Phoenix without difficulty,
and they supposed that they had already achieved (κεκρατηκέναι)
their intention, κρατεῖv here with an abstract noun does not readily
fall in with the ‘rule’ for the use with it of accusative and genitive
(see BDR § 170.2, n. 3; Μ. 3.232). The voyage seemed as good as
complete before it was begun.
ἄραντες: with this verb ἀγκύρας may once have been understood
but it had come to mean no more than to move (away), depart. Thus
Josephus, Ant. 13.86, ἄρας από τής Κρήτης κατέπλευσεν εἰς
Κιλικίαν; at 7.97 and 9.229 it is used of journeys by land. Page and
others add further examples from Thucydides, Plutarch, Lucian, and
Philo.
άσσον is the comparative of ἄγχι (which does not occur in the
NT) and is used as an elative, as near as possible (nearer than was
usual); see BDR § 244.2, n. 3. It has been conjectured that άσσον
may be an error for θᾶσσον (comparative of ταχύς), used similarly:
as quickly as possible. The Latin versions, not recognizing the word,
have de Asso(u); for Assos see 20.13.
For παραλέγεσθαι see v. 8.
In this verse h has the first person plural, tulimus ... sub-
legebamus. The Peshitto has we sailed.
14. The calm sea and the prosperous voyage did not last long (ού
πολύ; cf. 1.5). βάλλειν is nearly always a transitive verb, but here it
1194 COMMENTARY ON ACTS

is intransitive, the wind rushed down upon them (in the background
there may lie the thought of a reflexive—the wind hurled itself upon
them). Μ. 3.52 notes parallels in Aeschylus, Euripides, Aeschines,
and 1 Enoch 18.6.
κατ' αὐτῆς may mean against (the ship; but what would be the
antecedent?); more probably it is down from it (that is, from Crete).
Sudden and violent off-shore winds are known in this area (Page
257). ‘There is a noted tendency of a south wind in these climes to
back suddenly to a violent north-easter, the well-known gregale'
(Hemer 141). Begs. 4.331 suggests that ‘the squall from Mount Ida
drove them from the lee of Crete into the steady gale which had
passed right over them while they were coasting’. The wind is
described as τυφωνικός, translated by BA 1656 as Wirbelsturm, but
LS 1838 tempestuous is better, if it is strong enough. The word (and
some would have added, the wind itself) came from Τυφῶν, the
father of the winds; see Begs. 5.338-44.
As a recognized (and probably not unfamiliar) phenomenon the
wind had a name, Εὐρακύλων. Like Χώρος this is probably a
Latinism, at least in part, a compound of Εὖρος, south east wind
(Bruce 1.458, east), and Aquila, north wind. For the compound cf.
Εὐρόνοτος, a compound of εύρος and νότος (Aristotle, Meteor-
ologica 363b 22 (2.6)). The name Euraquilo suggests a wind blowing
somewhat east of north. Εύρακύλων is the reading of Ρ74 A B* latt
(co); Ψ <a> sy have Εὐροκλύδων, a south east wind that stirs up
the waves (κλύδωνες). It may have been suggested to copyists
by Etymologicon Magnum (ed. Gaisford) 772.30, τυφῶν γάρ έστιν
ή τού ανέμου σφόδρα πνοή, ὃς καί εὐροκλύδων καλείται.
See further C. J. Hemer, ‘Euraquilo and Malta’, JTS 26 (1975),
100-11.15

15. Again the narrative is carried forward by a genitive absolute—

a double genitive absolute. The ship was seized (by the wind) and
was unable to head into (ἀντοφθαλμεῖν, face up to, used metaphor-
ically at Wisdom 12.14 and in the reading of D al at 6.10) it. There
was nothing to do but let go and be carried along by it.
Composition with συν-, and the position of the participle,
strengthen the sense: ‘... été saisi d’un coup’ (Delebecque 131). Cf.
Sophocles, Electra 1150f., πὰντα γὰρ συναρπάσας θύελλ’ ὅπως
βέβηκας. This is metaphor. Cf. Thucydides 6.104.3, ἄρας παρέπλει
τήν ’Ιταλίαν, καί ἁρπασθείς ὑπ’ ανέμου ... ἀποφέρεται ές τό
πέλαγος; Lucian, Verae Historiae 1.34, ἁρπασθέντες ἀνέμω
σφοδρώ.
ἐπιδόντες stands oddly on its own (see below). According to ND
1.49 it needs a reflexive pronoun. BDR § 241.5, n. 7 suggests an
ellipse of τῷ ἀνέμῳ but τῷ ἀνέμῳ stands in the text and might be
taken with ἐπιδόντες rather than with ἀντοφθαλμεῖν; possibly with
 62. THE SEA VOYAGE. 27.1-44. 1195

both? It would be awkward to repeat it. In such a context as this one

thinks of the English ‘We let go’, with no object, reflexive or other.
Western editors evidently felt the text to be unsatisfactory and
added after ἐπιδόντες, τῷ πνέοντι [614 1518 have by a minor slip τφ
πλέοντι] καί συστείλαντες τὰ ἵστια (Metzger 497 gives as the
authorities for this reading 82 (614) (1518) 2125 syh *). It would not
be easy to take τῷ πνέοντι as a complement to ἀνέμω (before
ἐπιδόντες); it might be a substitute for a noun, άνεμος having
already been used. The meaning is clear and is no doubt what Luke
intended: We gave way to the wind, furled the sails, and ran before it.
Marshall (408) notes that the foresail must have been left in place or
it would have been impossible to steer. It does not however appear
that much steering was being done. For the use of φέρεσθαι and the
general picture, cf. Homer, Odyssey 5.343f.: σχεδίην ἀνέμοισι
φέρεσθαι κάλλιπε.
16. Except that τρέχειν adds the notion of speed, ὑποτρέχειν does
not differ in meaning from ὑποπλεῖν (vv. 4, 7): to sail under the lee
of—in this case of a certain island. Even in this relatively favourable
position we (the writer presumably associates himself with the crew
as passengers are apt to do—see FS Hanson 53) were scarcely able
to gain control of the dinghy, which presumably in ordinary circum-
stances would be towed behind the ship but would now (v. 17) be
lifted in board lest is should be damaged, or should damage the ship,
through a collision caused by the storm.
νησίον is a diminutive, a little island; so Μ. 2.345 (but the whole
section, 2.340-47, on Neuter Nouns in –ιον, should be studied; the
matter is not a simple one). The island is named, but with consider-
able textual variation. NA26 has Καῦδα, with P74 2 Β (Ψ has
Γαύδην, an orthographical and declensional variant) 1175 lat syp.
Avid 33 81 614 945 1739 2495 pc vgmss syh have Κλαῦδα; <a> has
Κλαύδην. Whatever the original text, the reference must be to a
small island south of Crete, whose modern name is Gaudes (or
Γαυδουῆσι, or Guzzo). It is not only in the NT MSS that the spelling
of the name varies, and it is impossible to be confident of the original
text. The Latin Gaudus (Gaudos) appears in Pliny, Natural History
4.61 (12); Pomponius Mela 2.7 (114), Ceudos. See further Hemer
(142). Metzger (498) quotes Begs. 4.332 (also Haenchen 672) for the
suggestion that Κλαῦδα was the Alexandrian, Καῦδα the Latin form
of the name.
σκάφη was borrowed into Latin (e.g. Cicero, De Inventione 2.51
(154), funiculo qui a puppi religatur scapham annexam trahebat), and
tended to drop out of Greek (Hemer 143). It is used in POxy
46.3269.9 (ND 3.17).
17. If in ν. 13 ἄραντες referred (not simply to setting out but) to
pulling up (the anchor), they (third person now, referring to the
1196 COMMENTARY ON ACTS

ship’s crew) at this point did the same thing (ἄραντες) with the
probably water-logged dinghy (σκάφη, v. 16).
Further steps were necessary. χρῆσθαι with the dative (βοήθειαν,
accusative singular in gig, and βοήθειας, genitive singular or
accusative plural, in 6 36 81 453 614 1241 2464 pc, are not to be
accepted) means to make use of; but in this context the meaning of
βοήθεια is not clear. It may mean generally help; that is, the sailors
made use of unspecified helping devices, designed to reduce danger
by increasing their control of the ship and its stability. Or it may have
a specific meaning as a nautical technical term. The latter is strongly
suggested by a few passages such as Philo, Joseph 33, κυβερνήτης
ταῖς τῶν πνευμάτων μεταβολαῖς συμμεταβάλλει τὰς πρὸς
εὔπλοιαν βοηθειας, εὐθύνων τὸ σκάφος οὐχ ἑνι τρόπῳ. Cf.
Aristotle, Rhetoric 2.5.18. If however the word does refer to specific
procedures or instruments we do not know what they were. This does
not justify the conjecture (S. A. Naber, Mnemosyne 23 (1895),
267-9) of βοείαις, ‘ropes of ox-hide’, though these might fit with
one possible interpretation of ὑποζωννύντες. This word (for the
participle in –υς see Μ. 2.205) occurs here only in the NT; see
Polybius 27.3.3. It appears to mean to provide a ship with
ὑποζώματα. Plato, Republic 616c, says that the rainbow holds
together την περιφεράν, the circle of the universe, οἷον τά
ὑποζώματα τῶν τριήρων, ‘the ropes or braces used to strengthen
the hull of a trireme' (LS 1811). It seems that the sailors perhaps
used the recovered dinghy to frap the ship, running ropes round it,
presumably at right angles to its axis, in order to prevent it from
breaking up under the violence of the waves. The main objection to
this interpretation is that the process described does not seem likely
to have done anything to prevent what, according to the next words,
the sailors feared. That fear, however, was dealt with by what
follows.
They feared that they might be cast upon the Syrtis. ἐκπίπτειν
serves as the passive of ἐκβάλλειν: BDR § 315, n. 1; Μ. 3.53. Thus
Herodotus 8.13.1, φερόμενοι τῷ πνεύματι καὶ οὐκ εἰδότες τῇ
ἐφέροντο ἐξέπιπτον πρὸς τάς πέτρας; Euripides, Helena 409(416):
1211(1227); Xenophon, Anabasis 7.5.12. The Greater Syrtis (now
the Gulf of Sidra) was the eastern, the Lesser Syrtis (now the Gulf of
Gabès) the western, part of the wide stretch of water between
Tunisia, Tripolitania, and Cyrenaica. It was reputed to be dangerous
to shipping because the water was shallow and the tides caused the
sandbanks to shift unpredictably. In 253 BC the Consuls Gnaeus
Servilius and Gaius Sempronius had been obliged to jettison their
stores in order to escape (Polybius 1.39.2-4). The area was feared
and avoided if possible (aἱ φοβεραὶ καὶ τοῖς ἀκούουσιν Σύρτεις,
Josephus, War 2.381 ; Sive per Syrtes iter aestuosas..., Horace, Odes
1.22). Would the sailors have feared this danger so soon? The Syrtes
 62. THE SEA VOYAGE. 27.1-44. 1197

were about 375 miles from Cauda (Stählin 317). According to Acts
they did so fear, and in consequence χαλάσαντες τό σκεύος, ούτως
ἐφέροντο (for φέρεσθαι cf. the passage from Herodotus quoted
above). ‘Die klass. Freiheit, ούτως zur Zusammenfassung des Inhalts
einer vorangegangenen Partizipialkonstruktion zu verwenden, findet
sich im NT nur Apg 20.11 ... und 27.17’ (BDR § 425.6; cf.
Delebecque 131).
What it was that the crew did we do not know. χαλᾶν means to
loose or to let down; σκεῦος means gear, equipment in general, or
might refer to a particular piece of equipment as a technical term
whose meaning is now lost. BA (1507) think that probably the sea-
anchor was meant; to drop this would slow the boat’s movement
towards the danger they feared. Similarly Preuschen (152), quoting
Plutarch, De Garrulitate 10 (507A), νεῶς μὲν γάρ ἁρπαγείσης ὑπό
πνεύματος επιλαμβάνονται σπείραις καἰ ἀγκύραις τό τάχος
ἀμβλύνοντες. See also Schneider (2.391). This seems to be the
meaning of the reading of g (not given in the apparatus of NA26; see
Begs. 3.243): vas quoddam dimiserunt quod traheret. On the other
hand 2495 pc s (syp) have (for τὸ σκεῦος) τά ἰστία (τό ἰστίον); this
probably means that the sailors loosed the previously reefed sails,
with a view to sailing as close to the wind as they could (though
Roloff (362) thinks that reefing of the sails is ‘nicht ganz ausges-
chlossen’). It will be remembered that the word σκεύος was used for
the sheet (?) of 10.11. These two variant readings (of g and of the
Peshitto) represent guesses at the meaning of χαλάσαντες τό σκεύος.
They are interesting; one of them might be right; neither is the
original text, which remains obscure. Note the occurrence of τήν
σκευήν in v. 19.
For ἐφέροντο, 36 453 pc syp bo have ἐφερόμεθα, by assimilation
to ν. 15. ἐφέροντο is required by ἐχρῶντο. The sailors who handle
the tackle are the subject in this verse.
On the verse see Hemer (143f.), and for a very clear statement of
various possibilities Marshall (409).

18. The sentence begins with yet another genitive absolute, and the
narrative returns briefly to the first person (ἡμῶν), since all those on
board were storm-tossed (χειμαζομένων), but it immediately reverts
to the third person, for it was the crew who, on the next day (τῇ ἑξῆς;
cf. 21.1; 25.17), jettisoned the cargo (cf. Jonah 1.5, oἱ ναυτικοὶ...
ἐκβολὴν ἐποιήσαντο τῶν σκευῶν τῶν ἐν τῷ πλοίῳ; there are
classical and Hellenistic parallels too). As in Jonah, the middle of
ποιεῖν is correctly used, since ἐκβολήν ποιεῖσθαι is equivalent to a
single verb. On the story of Jonah see the long passage of Pirqe R.
Eliezer 10, quoted in StrB 1.644-647; also the Targum of Eccles.
3.6; (There is a time for
throwing a thing into the sea, namely a time of great tempest). Cf.
1198 COMMENTARY ON ACTS

Josephus, War 1.280, τοῦ φόρτου τό πλεῖον ἐκβαλὼν μόλις εἰς

'Ρόδον διασώζεται—a winter voyage.
ἐκβολή is a throwing out, presumably of part of the cargo. Some
was left at v. 38, which constitutes a difficulty (Begs. 4.333) only if
we suppose that a full clearance was made at this point. ἐποιοῦντο
marks the beginning of the process, completed (as far as this stage
was concerned) next day with the aorist ἔρριψαν. Schille thinks that
they threw out what was on deck.
For χειμάζεσθαι see Thucydides 2.25.3; 8.99.3; Plato, Ion 540b.
For ἐκβολή, see Aeschylus, Seven against Thebes 769-771; Aris-
totle, Nicomachean Ethics 3.1.5 (1110a); Lucian, De Mercede
Conductis 1. The meaning of each word is clear and undisputed.
See Hemer (144).
19. On the third day (the next after the day of v. 18) the crew
(reading ἔρριψαν, third person, with Ρ74 A B C 33 36 81 323 453
945 1175 1739 1891 al latt co; against ἐρρίψαμεν, first person, with
Ψ <a>sy) with their own hands (αὐτόχειρες, here only in the NT, but
not uncommon elsewhere) threw out the ship’s tackle (σκευή; see
further below); not all the tackle (if this is what σκευή means)—see
vv. 30, 32, (38), 40.
αὐτόχειρες: on the use of the adjective with the value of an adverb
Delebecque (131) comments that it is ‘très classique’, and is almost
confined (in the NT) to Luke; see Acts 12.10; 14.10; 20.6 D. He
takes σκευή to mean “les agrès’’: ‘les sacrifices sont chaque jour
plus graves’.
LS 1607 take σκευή to mean tackle; that is, they do not distinguish
it from σκεύος (v. 17). This may well be correct; Luke character-
istically varies his vocabulary. There are however other possibilities
which, even if they are not convincing, are worthy of consideration.
Smith (116), accepting the reading ἐρρίψαμεν (but the editor of the
4th edition corrects this), thinks of the mainyard, an extremely heavy
weight, which the crew alone (v. 18) could not deal with, and which
was thrown overboard only with the assistance of the passengers. A
variant on this is noted by Hemer (145), who refers to the suggestion
of D. J. Clark (BT 26 (1975), 144-6) that in v. 18 ἐποιούντο is
conative; they tried by mechanical means to get rid of the mainyard
and failed; next day their achieved their object with their own hands.
But what other means would have been available? Cf. Jonah 1.5,
where σκευών, usually taken to be the genitive plural of σκεύος
could be the genitive plural of σκευή.
At the end of the verse 614 2147 (pc) it vgmss syh** sa add εἰς τήν
θάλασσαν. This must probably be regarded as another example of
the Western inability to know when to stop.
20. Again the narrative proceeds by means of two absolute
genitives. For many days (as at 13.31; 21.10, except that these have
 62. THE SEA VOYAGE. 27.1-44. 1199

πλείους) neither sun nor stars were to be seen. Neither noun has the
article, and BDR § 253.1, n. 2 suggest that this adds emphasis: ‘...
scheint durch die Auslassung [of the articles] der Sinn “weder etwas
von Sonne“ verstärkt zu sein’. ἐπιφαίνειν is here intransitive; the
English ‘neither sun nor stars showed’ is similar.
χειμῶνος οὐκ ολίγου is characteristic Lucan litotes; see 12.18. τε
simply adds a new statement Plutarch, Timoleon 28 (250), τού
χειμῶνος ἐπικειμένου, is very similar.
λοιπόν, used adverbially in a variety of senses, is here finally; see
BDR § 160.2, n. 3, with references; also BA 974, with examples.
This meaning however is not easily reconciled with BA 1301, where
the imperfect περιηρεῖτο is rendered ‘jede Hoffnung entschwand
allmählich’. One could perhaps say in English, ‘All hope was finally
disappearing’. The use of τού with the infinitive is characteristic of
Luke. The genitive of the cognate noun could also be used. Cf.
Aristophanes, Thesmophoriazousai 946 (953), κοὐκ ἔστιν ἔτ’ ελπίς
οὐδεμία σωτηρίας; Thucydides 1.65.1, ἐλπίδα οὐδεμίαν ἔχων
σωτηρίας.

21. A new feature of the situation is introduced by another genitive

absolute. People on board were short of food, not because there was
nothing to eat (it is not till v. 38 that food is thrown overboard) but
because it was not being eaten—through preoccupation, fear, and no
doubt seasickness. The Vulgate has jejunatio. For πολλής ασιτίας
ὑπαρχούσης cf. 21.40, πολλής δέ σιγής γενομένης. With this
situation Paul attempts to deal by allaying fear; he seems to have no
success till ν. 36. ‘Wogegen sich Paulus wendet, das ist die
Hoffnungslosigkeit’ (Bauemfeind 274).
τότε σταθείς: the reference of τότε is not clear. For Paul’s
standing in the midst of them (the ship’s company as a whole?) cf.
17.22, σταθείς ... ἐν μέσῳ τού Άρείου Πάγου. Page (260) recalls
also the 'Iustum et tenacem propositi virum ... inquieti... Hadriae’
(Horace, Odes 3.3.1,5); an interesting coincidence. One would be
inclined to think Paul’s 'I-told-you-so’ approach unlikely to win him
friends (and thus a mark of Lucan fiction?), but Page (260), noting
the μέν solitarium (cf. 1.1; 3.21; 28.22; and see BDR § 447.2c, n. 15)
comments, ‘The words “but you did not listen” are omitted in
courtesy.’
For the vocative ἄνδρες cf. v. 10; no further definition would suit
them all. The vocative is correctly preceded by ὦ (BDR § 146.3, n. 4;
Μ. 3.33). Paul refers back to the advice he gave at Fair Havens (v.
10). πειθαρχεῖν takes as usual a dative; see MM 500. For ἀνάγεσθαι
see ν. 2, for ὕβρις and ζημία, v. 10. In the latter part of the verse
there is an ambiguity which is best explained in a detailed note in
Begs. 4.334. There are, for κερδῆσαι... ζημίαν, two possibilities, (i)
The clause means (and there are many parallels in Greek) to gain by
1200 COMMENTARY ON ACTS

avoiding what is detrimental, that is “to avoid this danger and loss” ;
(ii) to incur something disadvantageous. If (ii) is chosen, the μή
negates both ἀνάγεσθαι and κερδῆσαι: If you had listened to me you
would not have set out and would not have incurred damage and loss;
if (i) is chosen, the μή negatives ἀνάγεσθαι only: ... you would not
have set out and would have avoided danger and loss. Both ways of
taking the clause make good—indeed the same—sense.
The verse has three Lucan features: ἔδει, μέν solitarium,
πειθαρχεῖν. Is it then an insertion into a secular sea-story designed
to bring it into the story of Paul? Certainly this verse (and the whole
of vv. 21-26) makes Paul stand out as a commanding character. But
perhaps he was one. See pp. 1178f.
22. Paul is now (τά νυν is perhaps more emphatic than the simple
νῦν would be; cf. e.g. 4.29; 17.30) able to urge his hearers to be of
good cheer. παραινεῖv is followed by an accusative and infinitive; of
this Μ. 3.138 says, ‘the accus. of the object with the infin. after
παραινέω is a mark of literary style’. This is at first sight surprising,
since the usual Attic construction is that the dative should follow
παραινεῖν; so BA 1246, BDR § 152.3, n. 3, and examples in LS
1310; also Blass (278); Radermacher (98). The point presumably is
that Turner wishes to indicate that υμάς should be taken with the
infinitive rather than with παραινώ: I urge that you be of good cheer
(not, I urge you to be of good cheer). So perhaps BDR § 392.1,
n. 5.
There will be no throwing away, loss (αποβολήν) of any human
life; (loss) only of the ship. For πλήν = only see BDR § 449.2. This
prediction differs from that of v. 10; it is given on supernatural
authority (vv. 23, 24). For the loss of the ship see v. 41; for the safe
arrival of all the travellers see v. 44.
23. Paul introduces the grounds (γάρ) for the encouragement that
he offers; it derives from an angelic message, delivered ταύτῃ τῇ
νυκτί; the dative expresses ‘time when’, and, unless Paul is speaking
in the night (which seems prima facie unlikely, but see on v. 33),
must mean ‘the night just past’—last night. Rapske (The Book of
Acts 3.359) quotes Krodel to the effect that Christian readers would
take the angel to be Christ—an ‘angel Christology’. In view of what
is said elsewhere in Acts about Christ and about angels this seems
very improbable.
The order of words is striking. According to Haenchen (674) it is
Lucan, but even so it is unusual. παρέστη stands first, though there
seems to be no reason why it should bear the emphasis that this
position would give it. The word itself is common in narratives of
epiphanies, angelic visitations, and the like. Cf. 12.7 (ἐπέστη);
Josephus, Ant. 1.341, ὁ θεός παραστὰς [Jacob] ἐκέλευσε θαρρεῖν;
Plutarch, Lysander 20 (444), ἔνιοι... φασιν αὐτῷ παραστῆναι τον
 62. THE SEA VOYAGE. 27.1-44. 1201

' μμωνα. It is interesting that in the next verse another semi-

Ά
technical use of the same verb appears. τοῦ θεοῦ not only precedes
άγγελος but is separated from it by the relative clause, which thus
receives some prominence. This is not unimportant. It has often been
pointed out (e.g. Conzelmann 144) that supernatural visitations and
supernatural protection are characteristic of the θεῖος ἀνήρ; it is
characteristic of Luke, recognizing this, to point out that though in
some respects Paul may resemble figures represented as θεῖοι
ἄνδρες, he is in fact no such thing. He is the property (οὗ εἰμι) and
the servant (ᾦ λατρεύω) of another, who alone can bear the title
θεός. See 10.26; 14.15; et al. Elsewhere in Acts λατρεύειν is used for
service of the God of Israel (7.7, 42; 26.7); 24.14 (λατρεύω τῷ
πατρῴῳ θεῷ) is particularly important. Paul is not speaking here of
Christ as θεός. The service implied here is the same as that described
in Rom. 1.9 (ᾦ λατρεύω ... ἐν τω εὐαγγελίῳ τού υἱοῦ αὐτοῦ). It is
the service of preacher and evangelist.
‘DEI esse, summa religionis; qua fides, amor, spes comprehendi-
tur. Correlatum, DEO servire’ (Bengel 485).

24. The words of the angel are reported. μὴ φοβοῦ is characteristic

of such visitations (cf. 18.9; also Lk 1.13,30; 2.10; 5.10; also Lucian,
Dialogi Deorum 20.7). Paul is assured that what is (for him) the goal
of the journey will be reached: he will stand before Caesar, to whom
he has appealed (25.11,12). παραστῆναι (used in v. 23 of the angel)
is used of appearing before a judge, but especially in the (transitive)
middle for producing a witness or defendant in court (e.g. Plato,
Republic 555b, παραστησώμεθ’ αὐτὸν εἰς κρίσιν). Cf. also Rom.
14.10.
καὶ ιδού is not translation Greek, but an imitation of the style of
the Greek OT (suitable for angels) where it translates For Paul
to appear before Caesar it is necessary only that he should escape the
storm, but as a favour (χάρις) God has granted (κεχάρισται) to him
(the lives of) all those who are sailing with him. It is implied, not
quite necessarily, that Paul has prayed for his fellow travellers.

25. εὐθυμεῖτε takes up εὐθυμεῖν in v. 22. Paul believes God; that

is, not, he trusts in God, but he accepts as true the message that God
has sent. God will do what he has said he will do; hence (διό),
ευθυμείτε.
καθ’ ὄν τρόπον: cf. 15.11. ούτως is not strictly necessary. Cf.
Thucydides 4.92.7, πιστεύσαντες τῷ θεῷ πρὸς ἡμῶν εσεσθαι.

26. There is a qualification of the promise of safety in the storm.

The ship will be cast (ἐκπίπτειν serves as a passive to ἐκβάλλειν, as
at ν. 17; see the note there) on a certain island. The angel did not
specify the island; it is not identified till 28.1. Schneider (2.394),
1202 COMMENTARY ON ACTS

quoting Wikenhauser, distinguishes the prediction of shipwreck from

the angel’s message, from which it is an inference.
δεῖ: ‘Das Schiff “muss” an einer Insel stranden. Damit ist die
Peripetie in der Darstellung der bedrohlichen Irrfahrt vorbereitet und
die kommende Rettung vorweg theologisch gedeutet’ (Pesch 2.291).
This makes too much of δει, but undoubtedly Luke means to
represent the whole course of action as leading Paul, under God’s
providence, to Rome.

27. The time of action of the main clause (ὑπενόουν οἱ ναῦται) is

given threefold determination: (a) in a temporal clause introduced by
ὡς (temporal ὡς is characteristic of Acts; see v. 1), When the 14th
night came (ἐπεγένετο, A 81 pc vg, differs little from the simple
ἐγένετο); (b) by a genitive absolute, διαφερομένων (for the meaning
of this verb see below) ἐν τῷ Άδρίᾳ (see below); (c) by the adverbial
phrase κατά μέσον τής νυκτός, about the middle of the night.
Smith (124-8) calculates that 13 or 14 days would be the time
required, under the conditions described, for the passage from Cauda
to Malta. The converse also applies (128): ‘... there is no other place
[than Malta] agreeing, either in name or description, within the limits
to which we are tied down by calculations founded upon the
narrative’. On this however see further on 28.1-10.
LS 417 take διαφέρεσθαι to mean tossed about; Delebecque (132)
agrees with this, rendering ballottés, and citing Plutarch, Galba 26
(1065); De his qui sero a Numine puniuntur 6 (552C). To these
passages may be added Philo, De Migratione Abrahami 148, ώσπερ
σκάφος ὑπ’ ἐναντίων πνευμάτων διαφερόμενον, Bruce (1.462)
takes the word to mean drifting (not up and down but) across, and
compares διαπλεύσαντες in v. 5. It should be noted however that
διαπλεύσαντες takes an accusative object (τό πέλαγος) whereas
διαφερομένων is followed by ἐν τῷ ’Αδρίᾳ; it would be hard to
translate this ‘drifting across Adria’. Adria is more than what is now
known as the Adriatic Sea. ‘At its maximum the Ionian Sea extended
all the way West to East from Gibraltar to the Levant, and ... the
Adriatic Sea at its maximum stretched North to South from Venice to
North Africa. Hence the area between Italy, Sicily, and Epirus
sometimes was all Ionian, sometimes all Adria, sometimes Adria in
the northern part, Ionian in the southern part’ (Begs. 4.335). See
Ptolemy, Geography 3.4.1; 3.15.1; 3.17.1; cf. Josephus, Life 15.
μέσον τῆς νυκτός; cf. 16.25, μεσονύκτιον; Xenophon, Anabasis
1.8.8, ἦν μέσον ημέρας.
At this time the sailors supposed, thought, προσάγειν τινὰ αὐτοῖς
χώραν. This is the text of NA26; it is read by A C Ψ <a>. This is a
strange expression; they thought that a certain land was approaching
them. This does in fact make sense as a matter of relative motion, and
it is not without precedent. Blass (279) quotes the sixth century
 62. THE SEA VOYAGE. 27.1-44. 1203

commentator on Epictetus, Simplicius 38: δι’ απειρίαν [sailors]

δοκοῦσιν οὐκ αυτοί προσιέναι τῆ πέτρα, αλλά τήν πέτραν κατ’
ολίγον ἐπ’ αυτούς ἰέναι. Cf. Vergil, Aeneid 3.72, provehimur portu
terraeque urbesque recedunt. There are several variants.
προσεγγίζειν (614 2147 2495 (sy)) makes no difference in meaning:
a certain land was drawing near to them. προαγαγεῖν ( (pc))
might be taken to mean that land was ahead of them, but it is hard to
understand why anyone should express himself in this way. B2 has
προσανέχειν. It might be just possible to make this mean that land
was awaiting them, that is, that it was just ahead. But B* has
προσαχεῖν. This could be regarded as a Doric form (a for η) of
προσηχεῖν to resound or re-echo (so LS 1513). What gives this
significance is that g has resonare sibi aliquam regionem, and s,
resonare sibi quondam regionem. Μ. 2.71 describes this as a very
attractive reading ‘which accounts for the variants’. It ‘has the
difficulty of being a Doric (etc.) form which disagrees with the
common derivatives of the same root: κατηχεῖν and ήχος are
conspicuous in the NT. Could it have been used by sailors from
Crete, Cyprus, Lesbos, Corinth, or some other maritime country
outside the Ionic-Attic area, appropriated as a t.t.?’ See on the other
hand Μ. 3.51, where προσάγειν is assumed without reference to
variants. Metzger (498) prefers προσάγειν on the ground that its
‘harshness’ may have given rise to the other readings, and so does
Ropes (Begs. 3.245) (‘προσάγειν, although itself difficult, is to be
preferred’). Clark (385) however prefers προσαχεῖν (= προσηχεῖν
= resonare). Hemer (146) takes up the suggestion of Smith (119-22)
that the sailors became aware of the breakers on the rocky point of
Koura, in the neighbourhood of St Paul’s Bay (see on v. 28 and
28.1). The reading must remain doubtful; fortunately the meaning is
not. The sailors became aware that they were not far from land.

28. Suspicion that the ship was approaching land naturally led the
sailors to take soundings (βολίζειν, a word of very scanty attesta-
tion). The result was 20 ὀργυιαί (for the accent see below). The
ὄργυια was the combined length of the outstretched arms, about 6
feet, or 1 fathom (1.85 m.); a nautical technical term. The water
(at 120 feet) was fairly deep, but it was growing shallower.
διαστήσαντες implies τήν ναῦν as object (BA 393); having moved
the boat (the narrative suggests that the wind was in any case moving
the boat) a little they sounded again and found a depth of 15 fathoms,
30 feet less water. Blass (279) understands the construction differ-
ently: ‘βραχύ διαστησ. = βρ. διάστημα (5.7) ποιήσαντες, quo et de
tempore et de loco accipi potest.’ The figures seem to be compatible
with data derived from St Paul’s Bay, Malta.
On ὄργυια LS 1246 have ‘Proparox. in nom. and acc. sg.’ as used
in Homer, ‘oxyt. or perispom. in other cases.’ More fully Winer-
1204 COMMENTARY ON ACTS

Schmiedel (72 = § 6.7): ‘Zwischen ὄργυιἄ und ὀργυκ<c> (AG 27.28)

schwankte die Sprache so, dass die alten Grammatiker ernstlich
vorzuschreiben scheinen, die Endung sei theils (sei es im Plural oder
auf allen zweifellos langen Vocalen) betont, theils (im Sing. bzw. nur
im Nom. Sing. u. pl.) unbetont.’ Cf. Μ. 2.58; BA (1174); Hemer
(147).

29. The rapidly shelving seabed meant that it was imperative to

halt the ship’s motion or at least reduce its speed. φοβούμενοι and
<b>ίψαντες will go with the third person εὔχοντο and refer to the
sailors; the first person ἐκπέσωμεν (81 326 945 1739 1891 al have
ἐκπέσωσιν) refers in a natural way (see FS A. T. Hanson 53) to all
who were on board.
τε (A B C <a>) gives a better connection than δέ (P74 Ψ 33 81 614
1175 2495 al syh); φοβούμενοι is positively connected with v. 28
and an adversative particle is not called for. The force of μή (lest,
becoming lest perhaps) is strengthened by the addition of που (for
which πώς and ποτε are more frequently used). As at vv. 17, 26
ἐκπίπτειν is used as the passive of ἐκβάλλειν. The danger was that
the ship might be thrown by wind and wave upon τραχείς τόπους,
rocky (literally rough, jagged) places. τραχύς is a stock epithet of
Ithaca (Homer, Odyssey 9.27; 10.417). For τραχείς, P74 104 2495 pc
have βραχείς. This may be a slip; otherwise it is hard to make a
better guess at its meaning than ‘auf enge [?] Stellen’ (Schneider
2.386). To avoid the danger of running into the rocks the sailors let
down (literally, threw, <b>ίψαντες) four anchors from the stem. It was
usual to anchor by the bows (e.g. Vergil, Aeneid 3.277, ancora de
prora iacitur), but since the ship was driving before the wind (vv. 15,
17) this would be the natural way to check its progress; see however
v. 30. What had to be avoded at all costs was allowing the ship to
take heavy seas broadside on. See Smith (133).
ηὔχοντο ἡμέραν γενέσθαι, primarily no doubt in order to be able
to see and choose the best place at which to beach the ship. Homer,
Odyssey 9.151, 436 (ἐμείναμεν Ήῶ δίαν) is sometimes quoted as a
parallel, but in these two lines Odysseus and his men are on shore.
See also Longus, Daphnis and Chloe 2.8, διὰ τούτο θᾶττον
εὐχόμεθα γενέσθαι τήν ημέραν.
At the end of the verse g vgmss add ut sciremus an salvi esse
possimus (vgmss possemus). This seems to represent τού εἰδέναι εἰ
σωθῆναι δυνάμεθα. But the clause may never have existed other-
wise than as a Latin Western expansion.

30. Again the narrative is taken forward by means of a genitive

absolute, the main indicative verb not appearing till v. 31. Having let
down anchors from the stem (v. 29) the sailors prepared to let down
more from the bow (πρφρης, instead of the expected πρώρας—see
 62. THE SEA VOYAGE. 27.1-44. 1205

BDR § 43.1, n. 1). It may be supposed that the stem anchors had
checked the forward motion of the ship and that the sailors’ intention
(if it was not that which Luke attributes to them) was to maintain the
ship’s position in the line of the wind, and so to prevent it from being
struck broadside by the heavy seas. This, it seems, would make
sense, and whereas the sailors would simply drop anchors over the
stem they would need the dinghy to stretch out (ἐκτείνειν) the anchor
lines from the bow. For χαλάν see v. 17; the dinghy would have to be
let down by rope (see v. 32) over the high prow of the ship. For
σκάφη see v. 16.
According to Luke, the sailors were deceiving the passengers; it
was not their intention to lower anchors but to escape from what they
presumably regarded as a doomed ship. Deceit is indicated both by
προφάσει, frequently used for falsely alleged motive (so LS 1539,
s.v. I 2; as e.g. Thucydides 3.86.2, οἱ ’Αθηναῖοι ... μὲν ...
προφάσει, βουλόμενοι δέ ...; 6.76.1), and by ὡς, which itself means
under pretence of (Μ. 3.158), giving not so much the subjective basis
(BDR § 425.3) of an action as the basis suggested, affirmed, by the
actor. For ὡς, see 23.15, 20.
Begs. 4.335f. argues strongly that Luke here misrepresents the
sailors; they were in fact attempting to do what they said they were
doing, would not have attempted to escape in a dark and stormy night
in a tiny boat that would hold very few of them, and must have been
very sorry to lose the dinghy through the soldiers’ precipitate action.
The dinghy would have helped them to beach the boat properly and
would have assisted passage from the boat to land. It is worth
observing that if Luke was making a mistake he was probably
making a mistake about something that really happened; one can
hardly suppose that he invented an action and invented also, without
letting his reader into the secret, a mistaken interpretation of it.
Hanson (249) agrees with Begs. Schille (466) argues that this
observation fails to recognize the literary character of the story; Luke
is describing a panic. Conzelmann (145) points out that the flight of
the crew is a regular feature in Greek stories; see Achilles Tatius 3.3;
Petronius 102, cf. 114.
At the end of the verse (cf v. 29), gig vgmss add, ‘ut tutius navis
staret’. This Clark includes as part of the Western text in the form
τοῦ ἀσφαλέστερον ... τό πλοῖον ἑστάναι.

31. Paul evidently (according to Luke) viewed with dismay what

he thought to be the impending departure of the only persons with
sailing skills. With them all prospect of safety would disappear. (But
could he not as before rely on the divine promise?) It is surprising
that he should address the soldiers directly, as well as their com-
manding officer. And one would have thought that even if some
sailors escaped (presumably to the land which the boat seemed to be
 1206 COMMENTARY ON ACTS

approaching (v. 28)) enough would have been left, unable to get into
the dinghy, to run the ship.
It is worth observing, without prejudice, that what seems improb-
able as an account of what really happened must also seem im-
probable in a work of fiction written by an author with any concern
for verisimilitude. It may be that Luke’s motive was to show (in v.
32) the respect shown to Paul, ‘der auch in der Mitternacht wachsam
auf Posten ist’ (Stählin 319), by the soldiers, who immediately acted
on his advice.
32. If the sailors were not to be allowed to escape in the dinghy the
simplest, but not necessarily the wisest, course was to get rid of the
dinghy. This the soldiers did. It was suspended over the side of the
ship (v. 30) and the soldiers simply cut the ropes and permitted
(εἴασαν; ἐᾶν occurs 7 times in Acts, twice in Lk., thrice in the rest of
the NT) it to fall away. This act was, according to Begs. 4.336,
responsible for the loss of the ship. ‘It was only necessary to wait for
the end of the gale, and row ashore in comfort.’ It would have been
unnecessary to beach the ship and expose it to the violence of the sea
(vv. 40, 41).
Delebecque (133) thinks that Luke may have remembered Homer,
Odyssey 10.27; Xenophon, Hellenica 1.6.21. It seems very unlikely.
33. This verse (and the following paragraph) might well have been
introduced by v. 21a; see on that verse. It is surprising that the
reference to food should come a second time; it may be that Luke
took what we read in v. 21 out of the present material and used it to
introduce the prediction he ascribes to Paul in vv. 22-26. Con-
zelmann (145), and others, take vv. 33-36 to be an insertion; see on
v. 37.
According to Begs. 4.336 (comparing μέχρι at 10.30, and the
African rendering—in die quo—of ἄχρι in 1.2), ‘ἄχρι gives no good
sense if it be rendered ‘ ‘until’ ’ ’. The clause is translated, ‘And when it
was nearly day’. It is true that this is what the reader would expect, but
it is hardly justifiable as a rendering of ἄχρι δὲ οὗ, especially as this is
followed by the imperfect παρεκάλει. The imperfect is indeed often
used in verbs of requesting, entreating, and the like (these in them-
selves represent incomplete action; see Μ. 3.65; BDR § 455.3b, n. 6),
but παρεκάλει may well be iterative: Until day was about to break,
Paul kept on exhorting... It may be that this should be connected with
ν. 23, ταύτῃ τῇ νυκτί, if this means, In this night, which is not yet over
(see the note). This could bear on the construction of the chapter,
especially if it is thought to include ‘Pauline’ insertions into an
independent sea-voyage story; on this see pp. 1178f.
Paul exhorted them all to partake of food. If ἅπαντας is to be
distinguished as a stronger word than πάντας, it will be intended to
make the point that Paul had in mind his fellow passengers, the
 62. THE SEA VOYAGE. 27.1-44. 1207

soldiers, and the sailors. This however would be the natural meaning if
πάντας were used. It may be no more than a mannerism: άπας occurs
in Lk. 11 times, in Acts 10 times, in the rest of the NT 11 times. For
taking food Paul uses here μεταλαμβάνειν with the genitive; later in
the verse, προσλαμβάνεσθαι with the accusative; in v. 34
μεταλαμβάνειν (but H L P Ψ 049 326 1241 2495 have
προσλαμβάνεσθαι) and the genitive; in v. 35, ἐσθίειν with no object;
in v. 36, προσλαμβάνεσθαι with the genitive. LS 1113 quote only
Acts for μεταλαμβάνειν with the meaning to partake offood, but it is
clear that this meaning was coming into use in the post-classical
period; so Josephus, War 2.143; PRyl 2.77.19 (BA 1035). For the use
of προσλαμβάνεσθαι for taking food MM 549f. offer nothing and BA
1436 only Clementine Homilies 3.21 (active). See BDR § 169.2, n. 5.
Paul’s exhortation begins with another curious expression of time.
(cf. v. 27). BDR § 161.3, n. 6 describe τεσσαρεσκαιδεκάτην
σήμερον ἡμέραν προσδοκῶντες as a ‘besondere Redensart’ (as it
is), and translate, ‘jetzt schon den 14. Tag wartend’. This can only
mean (as prima facie the Greek can only mean), ‘We are now
awaiting the 14th day’. This presumably is to be associated with the
14th night (in v. 27), and the statement at the beginning of the present
verse that day was about to break. If nights are counted before days,
we have now had almost the whole of the 14th night and are now
awaiting the imminent beginning of the 14th day (the omission of the
article before τεσσαρεσκαιδεκάτην follows old custom) of flying
before the NE wind. We know already from v. 21 that this period had
been marked by πολλή ασιτία. This is now expressed in the words
ἄσιτοι διατελεῖτε. Radermacher (169) argues from Phrynichus’s
(244—he says 277) treatment of τυγχάνειν that ‘correct’ use would
have included the participle (in this verse the plural ὄντες); similarly
BDR § 414.1, n. 4. But (if it is in any case legitimate to argue from
τυγχάνειν to διατελεῖν) Rutherford’s (342) comment should be
observed. ‘Even in the best age the participle of the substantive verb
was sometimes carelessly omitted after τυγχάνω. If the prose
instances are set aside as of no importance in such an inquiry,
there is a line of Aristophanes to confute such scholars as would
correct the texts of prose writers by the dictum of Phrynichus—
καὶ τῶν θεατῶν εἴ τις εὔνους τυγχάνει.
Ecclesiazusae 1141.’
Thus ἄσιτοι διατελεῖτε is You continue without food, or better, You
have been (and still are) continuing without food. For the adjective
cf. Sophocles, Ajax 324, ἄσιτος ἀνήρ, ἄποτος; Galen, On Phle-
botomy 11.242.
μηθείς occurs here only in the NT; for οὐθείς see 15.9; 19.27;
20.33; 26.26.
’Nachtstunden der äussersten Spannung. Ein Deuteropaulus
würde jetzt gewiss gesagt haben: “Diese Stunden wollen wir in
1208 COMMENTARY ON ACTS

ununterbrochenem Gebet durchleben.” Der stattdessen sagte: “Wir

wollen endlich etwas essen” ... das ist der historische Paulus
gewesen’ (Bauemfeind 275). A good remark; but one cannot exclude
the possibility that a Deuteropaulus may have learned something
from the historische Paulus.
34. The request of v. 33 is now repeated in the first person of direct
speech, τούτο refers back to the infinitive, μεταλαβεῖv τροφής. For
μεταλαβεῖν, H L P Ψ 049 326 1241 2495 <a> have προσλαβεῖν.
ὑπάρχει seems unnecessary; ἐστιν would suffice. πρὸς τῇς ὑμετέρας
(ALP 326 614 1241 pm w syh have ἡμετέρας) σωτηρίας is the only
occurrence in the NT of πρός with the genitive. Cf. Thucydides
3.59.1, οὐ πρὸς τῆς ὑμετέρας δόξης... τάδε; Josephus, Ant. 16.313,
πρὸς τῆς τού βασιλεύοντος ἐδόκει σωτηρίας, σωτηρία must here
refer to physical well-being—a not unimportant consideration.
In the next clause οὐδενός stands first for emphasis: Not one of
you will lose a hair ... The promise recalls that of v. 22; the
expression is proverbial; cf. 1 Sam. 14.45; 2 Sam. 14.11; 1 Kings
1.52; Lk. 12.7 = Mt. 10.30; Lk. 21.18 is important because it is a
Lucan insertion. Luke, we may suppose, liked the image. For
ἀπολειται, Ψ <a> gig syh sa have πεσεῖται. After ὑπάρχει gig adds
spero enim in deo meo quia; this MS shows great freedom in
handling the wording of the text.
Bruce 1.465 observes that if γάρ is omitted from the sentence
οὐδενός ... ἀπολεῖται what is left is a perfect hexameter. Inter-
esting; but probably a pure coincidence.
35,36. To give weight to his words Paul initiates a meal in words
and acts that have important parallels; see FS A. T. Hanson (60).
These may be set out as follows:
λαβών άρτον: cf. λαβὼν ἄρτον, Lk. 22.19; λαβών δέ τούς
πέντε άρτους, Lk. 9.16
εὐχαρίστησεν τῷ θεῷ; cf. εὐχαριστήσας ... εὐχαριστήσας, Lk.
22.17, 19; εὐλόγησεν, Lk. 9.16
κλάσας: cf. ἔκλασεν, Lk. 22.19; κατέκλασεν, Lk. 9.16
ἤρξατο ἐσθίειν: cf. ἐπιθυμία ἐπεθύμησα τούτο τό πάσχα
φαγεῖν, Lk. 22.15
ἐπιδιδούς καὶ ἡμῖν (614 2147 pc syh** sa): cf. διαμερίσατε εις
εαυτούς, Lk. 22.17; ἔδωκεν αὐτοῖς, Lk. 22.19; ἐδίδου τοῖς
μαθηταῖς, Lk. 9.16
προσελάβοντο τροφής: cf. ἔφαγον καί ἐχορτάσθησαν πάντες,
Lk. 9.17
We may compare also the references to the breaking of bread at Lk.
24.30, 35; Acts 2.42, 46; 20.7, 11. As far as language goes, this is
more ‘eucharistic’ than any other passage in Acts. Here only does the
verb εὐχαριστεῖν occur, and here only is the leading figure said to
 62. THE SEA VOYAGE. 27.1-44. 1209

take (λαμβάνειν) the loaf before breaking it. Black, AA 125 (cf.
Wilcox 125) may be right in describing λαμβάνειν thus used as an
auxiliary and seeing its origin in Aramaic usage, but the coincidence
in language with that of the Last Supper cannot be missed and can
hardly be accidental. At the same time the context demands not a
purely symbolic meal but a meal, eaten to overcome the physical
hunger of men who for days been too busy and too preoccupied
to eat. Paul began to eat and the others—all of them, πάντες—
followed his example. The Christians did not withdraw to hold a
special, still less a secret, rite of their own. ενώπιον πάντων is
clear.
Luke’s readers can hardly have failed to note the eucharistic
allusions; yet these terms and allusions are all such as are rooted in
ordinary Jewish practice—a meal began with the blessing of God
and the breaking of bread. No explanation of the data is satisfactory
that does not include the inference that they belong to a time,
probably lasting at least till the date of the composition of Acts, in
which the eucharist (understood in the narrow sense of a rite
involving the symbolic eating and drinking of bread and wine to
which a theological interpretation is attached) was not yet separated
from a fellowship meal in which normal quantities of food and drink
were consumed. This inference is consistent with the other references
in Acts to the breaking of bread (and, it may be added, with Pauline
and other NT evidence; see 1 Corinthians 231-5, 261-77; CMS
60-76).
The question, Is the meal described in this chapter a eucharistic
meal or not? has been given diverse answers. Calvin (2.295) makes
no reference to the possibility that the eucharist may be in mind. It
was not a eucharist but ‘einen Akt der Ermutigung und vielleicht
auch ... ein Wunder, durch das Paulus die Seekrankheit geheilt hat’
(Roloff 364). The meal was a eucharist understood as a φάρμακον
ἀθανασίας (Schille 467). ‘Lukas deutet dergleichen [a prefiguring of
the eucharist] nicht an’ (Haenchen 676). It was certainly not ‘das
christliche Abendmahl’ (Schmithals 232). Weiser (664f.) attempts a
difficult distinction: ‘Auf diese Weise und durch die auffällige
Aussage, dass Paulus “zu essen begann” und erst “dann alle
Speisen zu sich nahmen”, macht Lukas deutlich, dass zwar nicht alle
das eucharistische Brot empfingen, aber doch alle Anteil erhielten an
der im eucharistischen Mahl gegenwärtigen Rettungsmahl des aufer-
standenen erhöhten Herrn.’ Pesch (2.292) is somewhat similar: The
‘Beschreibung an die Speisungswundererzählungen erinnert, aber
nicht auf eine Eucharistiefeier hinweist.’ Cf. Rackham (477).
Schneider (2.397) goes further: ‘Die Mahlzeit hat eine ähnliche
Transparenz auf die Eucharistie hin, wie das Stichwort σωτηρία in
V.34b auf die Bedeutung “Heil” hin offen ist. Der Christ weiss, dass
das Herrenmahl als Ausdruck der “Hoffnung” (vgl. V.20) dem
1210 COMMENTARY ON ACTS

“Heil” dient (vgl. V.34b).’ Stählin (320) goes farther still. There ‘...
kann kein Zweifel sein, dass er [Luke] ein irgendwie sakramentales
Mahl ... skizzieren wollte, wenn auch vielleicht nicht ein eigen-
tliches Herrenmahl'.
All these interpretations, instructive as they are, fail because they
assume something like the clear developed distinction between an
‘ordinary’ meal and a ‘sacramental’ or ‘eucharistic’ meal, between
bread and eucharistic bread.
For the breaking of bread (κλάσας) see on 2.42; also for the lack
of any reference to wine. Of course, the use of wine during a storm at
sea would have presented very considerable practical difficulties.
For εύθυμοι see on vv. 22, 25 (εὐθυμεῖv).
προσελάβοντο, aorist middle, the ‘correct’ form (see v. 33), is
read by B C <a>. There are several variants, μετελάμβανον (614 2147
pc) assimilates to μεταλάβεῖv in v. 34, and so (in the use of active
rather than middle) do προσέλαβεν (Α Ψ 1175 pc) and
προσελάμβανον (2495). μεταλαβαν (sic) though read by as well
as 1241, bears witness only to careless inattention.

37. The numbering of those on board the ship follows upon the
πάντες of v. 36: Luke will tell his readers what πάντες means.
Schneider (2.397) however thinks that the number would link
originally with v. 32. The article before πᾶσαι indicates the totality
of persons present (Μ. 3.201—‘We were in all...’; Zerwick § 188;
BDR § 275.3, n. 6). Unfortunately the number is textually uncertain.
The majority of witnesses have διακόσιαι ἑβδομήκοντα ἕξ, 276; but
B (pc) sa have ὡς ἑβδομήκοντα ἕξ, about 76. The textual problem is
complicated by the fact that 276, if not written in words, would be
written COς, and 76 as Oς. Metzger (499f.) represents a common
opinion in the words, The reading of B sa ‘probably arose by taking
ΠΑΟΙΩΟΟς as ΠΛΟΙΩΩΟΟς. In any case, ὡς with an exact
statement of number is inappropriate (despite Luke’s penchant for
qualifying numbers by using ὡς or ὡσεί, cf. Lk. 3.23; Acts 2.41; 4.4;
5.7, 36; 10.3; 13.18, 20; 19.7, 34).’ Metzger notes other variants: A
has 275; 69 and Ephraim have 270; bomss have 176 or 876; 522 and
1680 have 76; Epiphanius has about (ὡς) 70. Metzger (similarly
Ropes, Begs. 3.247) is probably right but like most commentators
does not note the problem of the iota subscript, which in uncials is
often though not always written adscript. Thus the two readings
discussed might well be not as given above but ΠΛΟΙΩΙΟΟς and
ΠΛΟΙΩΙΩΟΟζ. This makes simple confusion less likely. The
number 276 is not impossibly large; Josephus (Life 15) records his
own experience of shipwreck (in Adria), as a result of which about
600 were obliged to swim all night. On the size of ships see Smith
(187-90) and Hemer (149f.) A surprising number of commentators
repeat the statement that 276 is the sum of the digits from 1 to 24. It
 62. THE SEA VOYAGE. 27.1-44. 1211

is not; it is the sum of the digits from 1 to 23. The fact seems in any
case irrelevant; there is no hint of numerical or other symbolism
here.

38. When they were satisfied (κορεννύναι occurs only here and at
1 Cor. 4.8 in the NT; a literary word—see LS 980) with food
(τροφής, genitive; again, like the verb, a literary touch—BDR §
169.3, n. 7), they lightened, perhaps they set about lightening
(ἐκούφιζον, imperfect, perhaps inceptive) the ship. They did this,
rashly, one may think, unless they knew that they would soon be on
land, by throwing overboard (modal use of the participle
ἐκβαλλόμενοι) the supplies of food. It is not clear what may have
been thrown overboard at an earlier stage; no doubt it now seemed
safe enough to get rid of the food. τὸν σίτον is not difficult enough to
justify the conjecture τόν ἵστoν, the main mast. See Hemer (150). Cf.
Polybius 1.39.4, referred to above on p. 1196.

39. V. 33 describes the ship’s company waiting for day. Now it

came; they could see land, but not recognize it. It is identified in 28.1 ;
see the note there (also A. Acworth, JTS 24 (1973) 190-193; C. J.
Hemer, JTS 26 (1975) 100-111; also Hemer 150).
Μ. 1.117 notes that ‘durative action is most certainly represented in
the present κατανοεῖv, except Acts 27.39’; for this imperfect
Moulton tentatively suggests ‘noticed one after another*. If one is to
make such guesses, the suggestion ‘as daylight increased they
perceived more and more clearly’ might do as well; but it is probably
best to suppose that on this occasion Luke did not choose the most
suitable tense. What they saw was a bay, with a beach. This offered a
way of escape from the storm. Had the dinghy still been available it
might have been used for successive trips to the shore; it was not
available (v. 32), and the alternative was to run the ship on shore
(ἐξώσαι). ἐξωθεῖν has other meanings, but this is attested in
Thucydides 2.90.4; 7.52.2; 8.104.3. B* C pc have ἐκσῶσαι, identical
in sound (the variant being due to dictation?) but weaker in meaning
(Μ. 2.108; but Begs. 4.337 think B C may be right). They did beach
the ship, and they did not save it. Their planning (ἐβουλεύοντο) was
qualified by the parenthetical (so Μ. 3.196, ‘if possible’; cf. v. 12 and
20.16, but see below) εἰ δύναιντο. On the optative see Μ. 3.196; also
Radermacher (131). Moule (IB 151) finds the meaning unclear. We
have to choose between Planning whether they could ... and
Planning (if they could) to ... In the former alternative, considering
might be a preferable alternative. But it seems better to regard εἰ
δύναιντο as parenthetical.
For the manoeuvre cf. Caesar, De Bello Civile 3.25, ... atque eo
naves ejicere possent.
Haenchen (677) considers that this verse connects perfectly with
1212 COMMENTARY ON ACTS

v. 32. It does not seem to connect less well with v. 38. See Hemer
(150).
For δύναιντο ( A Β Ψ 33 81 323 (614) 945 1175 1739 2495 al
latt), C <a> co have δυνατόν. ‘The gloss οι ναυται, which in 920 is
attached to εβουλευοντο, appears in gig vg. codd. sah cod. P pesh as
subject of “knew not” ’ (Ropes, Begs. 3.247).
40. This verse describes the steps taken by the sailors to achieve
the goal planned in v. 39. περιελόντες is a surprising word (it is used
at ν. 20 and in yet another sense at 28.13); it means to strip off, to
take off, here it presumably means that the anchors were detached
from the ropes that held them and allowed (εἴων) to drop into the sea.
At the same time (for ἄμα with participle see BDR § 425.2, n. 2) they
(the sailors) let go the fastenings of the rudders. Fastenings translates
τὰς ζευκτηρίας. According to LS 754 ζευκτηρία is equivalent to
ζεύγλη, which (LS 753) means, in addition to a fastening on a yoke
for draught animals, the ‘cross-bar of the double rudder’. This is the
meaning it will have here, used in connection with the steering-
paddles, of which Greek ships normally had two. The action
described by Luke is reasonably clear: the anchors were let go, the
steering apparatus was dismantled. The crew had given up any
thought of steering the ship; the wind would drive it where they
wished to go.
To accelerate their progress they raised the foresail (τόν
ἀρτέμωνα) to the breeze. τῇ πνεούσῃ is simply that which was
blowing. Unless with P74 we read τή πνοή, there is probably (see
BDR § 241.6, n. 8; Μ. 3.17) an ellipse of τή αὔρα; cf. Arrian,
Epistola ad Traianum 3.2 (3), ταῖς αὔραις ταῖς ἐκ τῶν ποταμών
πνεούσαις. αύρα is a breeze, but sometimes in particular an off-sea
breeze; this is evidently what was in mind here. With this behind
them they held on (κατεῖχον; cf. Herodotus 6.101.1; 7.59.3, ές
τοῦτον τὸν αἰγιαλὸν κατασχόντες τὰς νέας) to the beach. As
appears from the quotation it is necessary to understand τήν ναῦν as
object of κατεῖχον.
Page (262) takes εἴων to mean, They left [the anchors] in the sea.
He says that this accounts for the use of εις, but it surely does so less
well than ‘they let them slip into the sea’.
Stählin (321) understands the reference to the rudders in
the opposite way: ‘man löst die Riemen der zwei während des
Ankerns hochgezogenen Steuerruder und macht sie so wieder
gebrauchsfertig.’
41. περιπίπτειν is (among other things—for an example see ND
3.141)—a nautical term, used of ships falling foul of one another,
also of ships being wrecked. Page (262f.) points out its suggestion of
the unexpected; they were making for the beach when their course
was unexpectedly interrupted, as they fell upon a τόπος διθάλασσος.
 62. THE SEA VOYAGE. 27.1—44. 1213

The adjective is used of a double sea, that is, a sea divided into two
parts (as the Euxine, Strabo 2.5.22). LS 427 give for the present
passage, ‘between two seas, where two seas meet, as is often the case
off a headland’. This suggests a headland jutting out between two
distinguishable stretches of water; Page (263) similarly thinks of a
ridge or reef running out to sea. Begs. 4.339, though unwilling in
translation to commit themselves further than ‘of two seas’, evidently
favour the meaning shoal, and this certainly makes good sense of the
narrative. If the ship ran on to land it is hard to see why the travellers
should need to swim or otherwise take to the water (vv. 42-44).
Even if they were at the stem of the ship, this did not break up at once
(see below) and they could have made their way to the bow and
jumped to the ground (ships often carried a ladder that could have
been used). If however the bow was stuck in a shoal and the stem
was exposed to the force of the waves all is clear. Everyone would be
obliged to swim ashore, or find some other means of getting through
the water. The meaning shoal is adopted by BA 392 (‘eine dem
eigentl. Strand vorgelagerte Sandbank’), but it must be admitted that
it rests only upon a consideration of what appears to fit the present
context. It is not inconsistent with the etymology of διθάλασσος.
There was sea before and behind the ship.
ἐπέκειλαν is a nautical term (ἐπικέλλειν), apparently not used by
prose writers: to run a ship on land. It is used by Homer: Odyssey
9.148, πρὶν νῇας ἐϋσσέλμους ἐπικέλσαι; 9.546, νῆα ... ἐκέλσαμεν
ἐν ψαμάθοισιν. Β2 <a> have the prose word, ἐπώκειλαν, used by
Herodotus (see LS 675; M.2.243). There is no difference in meaning.
Each verb takes τὴν ναῦν as object. This word (ναῦς) was becoming
obsolete; did Luke take it from Homer? Begs. 4.339 think the
suggestion ‘very attractive’. ‘If Luke was acquainted with Aratus and
Epimenides, his knowledge of Homer is easily credible.’ Cf. Μ. 2.9;
Bruce (2.474, 493); Hemer (151). BDR § 47.4d are content to
describe the sentence as literary.
The bow (πρώρα) of the ship, striking ground first, stuck fast,
became fixed (ἐρείοασα); there it remained, unshaken by what was
left of the storm. The stem (πρύμνα; like πρώρα, sometimes written
with –η) was in a different position, exposed to the violence of the
waves.
Surprisingly, Μ. 3.65 describes the imperfect ἐλύετο as conative
(or inchoative), and translates ‘the surf seemed to be trying to break
up the prow’. In the Greek however the subject of the passive verb is
πρύμνα (stem), and this was not trying to break up anything, or even
to be broken up. The imperfect is clearly durative: bit by bit the stem
was being broken up. BDR § 326 also put the verb under the heading
‘Imperfekt de conatu’ but inconsistently translate, ‘ “fingen sich zu
lösen’’ oder “löste sich mehr und mehr’’ ’. Blass (282) correctly:
‘ἐλύετο dissolvi coepit, cf. [v.] 20, Vergil, Aeneid 10.303-305’:
1214 COMMENTARY ON ACTS

namque inflicta vadi dorso dum pendet iniquo

anceps sustentata diu fluctusque fatigat,
solvitur atque viros mediis exponit in undis.
Preuschen (154) adds Aeneid 5.206, inlisaque proraque pependit.
The last words appear in various forms, τής βίας τῶν κυμάτων is
the reading of P74 C <a> sy; A B have the first two words only,
Ψ 2464 have the third and fourth only; gig vg have vi maris. It was
the unessential sort of narrative note with which copyists felt free.
‘The curtness of υπο της βιας led to various expansions’ (Ropes,
Begs. 3.249).
Haenchen (677) takes ὑπό τής βίας to mean ‘unter der Gewalt des
Rammstosses’. One would have expected this to affect the bow (the
point of impact) rather than the stem.
42. The soldiers’ plan was sensible enough. They were responsible
for the safe custody of the prisoners (v. 1); see on 12.19; 16.27. For
them it was better that the prisoners should die than that they should
escape. Where they might swim off (ἐκκολυμβᾶν, compound; cf. v.
43) to is not clear, but no doubt some would have felt the risk of
drowning to be worth taking. See Hemer (152).
βουλὴ ἐγένετο followed by ἵνα is described by Μ. 3.139 as a
Hebraistic figure of speech; we are not told, and it is not clear, what
Hebrew might be responsible for it. To BDR § 393.5 it is a ‘freiere
Wendung’.
The reading of gig (cf. v. 41) is tunc cogitaverunt milites ut omnes
custodios ...
43. The centurion was of another mind. From early in the voyage
he had treated Paul with consideration (v. 3), though he had failed to
take his advice (v. 11). If διασῶσαι is anything more than a
strengthened form of the simple verb it will mean that Julius wished
to bring Paul safely through the present danger; v. 44; 28.1, 4. There
may be a special application (as these verses suggest) to coming
safely through water; cf. Josephus, Apion 1.130, Νῶχος ... διεσώθη.
There may also have been a desire to distinguish rescue from
shipwreck from being saved in a Christian, religious sense.
ἐκώλυσεν is a perfective aorist (Μ. 3.72; similarly Zerwick §
252); Julius succeeded in preventing the soldiers from carrying out
their intention. For κωλύειν with accusative of the person prevented
from doing something or having something done, cf. 8.36; 11.17. He
could not however save Paul without saving all, or at least giving all
the opportunity of saving themselves. Some could swim (from 2 Cor.
11.25 one guesses that Paul could); they were to throw themselves
overboard—άπορίπτειν, normally a transitive verb, is here used
intransitively, or reflexively (without the reflexive pronoun). They
were to begin (πρώτους) the evacuation of the wreck and get to land
 62. THE SEA VOYAGE. 27.1-44. 1215

(ἐπί τὴν γην; Zerwick § 123 contrasts the use of ἐπί here with the
accusative with that in v. 44 where dative and genitive are used).
The reading of gig is: centuno autem prohibuit hoc fieri praecipue
propter paulum ut salvum illum faceret. et jussit illos qui possent
enatare primos exire ad terram ...

44. The non-swimmers would need help. Two groups are dis-
tinguished by οὓς μὲν ... οὓς δέ ..., where τοὺς μὲν ... τούς δέ ...
would have been better. Of Luke’s construction BDR § 293.3b, n. 11
say, ‘Bei den Attikern kaum belegt (K[ühner]-G[ert] II 228), häufiger
in hell. Zeit (Mayser II 1,57).’ The first group went ἐπί σανίσιν.
σανίς is a plank, board, almost any piece of timber, and ἐπί
presumably means on. The second group went ἐπί τινων τῶν ἐπὸ
τοῦ πλοίου. Here ἐπί is constructed with the genitive, which also
would normally mean on (it is not clear how Zerwick § 123 is
distinguishing between in tabulis and super ea quae de navi erant).
But what were the σανίδες if they were not τὰ ἐπὸ τοῦ πλοίου,
broken pieces of timber coming from the ship? This raises the
question whether τών should be regarded as masculine, οἱ ἀπὰ τοῦ
πλοίου. Those who could not swim were assisted by men from the
ship who could. Μ. 3.272 regards ἐπί with the dative and ἐπί with the
genitive as here interchangeable; similarly BDR § 235.1, n. 1. It is
not however easy to see why Luke should have repeated himself:
some on planks or pieces of timber (which must have come from the
ship) and others on some of the things that came from the ship
(which unless they were pieces of wood would not have been much
use as floats). The use of μέν ... δέ indicates a real distinction and
this can only have been in the manner of coming ashore. We may add
by way of support Hanson (251): τινων without a noun is ‘more
usually applied to persons than things’; Delebecque (135); ‘τών
(ἀπό) ne peut être un neutre, car l’expression οἱ ἀπό est courante en
grec pour désigner des “hommes de”, ou “venant de”.’ Cf.
Testament of Naphtali 6.6: ὁ δὲ ’Ιωσήφ ἐπί ἀκατίω ἐπορεύθη. ἡμεῖς
δὲ διεχωρίσθημεν ἐπί σανίδων έννέα.
However the transit from wreck to shore was made it was safely
effected: all came safely (on διασῷζειν see v. 43) to land (ἐπί and the
accusative as in v. 43, except in 614, which has ἐπὶ τῆς γῆς). Cf.
Thucydides 3.108.3, χαλεπῶς διεσώζοντο ἐς τὰς Όλπας.
 63. FROM MALTA TO ROME 28.1-16

(1) Having got safely through we then recognized that the island was called
Malta. (2) The local inhabitants showed us no ordinary kindness, for because
of the rain that came on and the cold they lit a fire and brought us to it. (3)
When Paul gathered a bundle of sticks and put them on the fire a viper came
out because of the heat and fastened on his hand. (4) When the inhabitants
saw the creature hanging from his hand they said to one another, ‘No doubt
this man is a murderer; though he escaped from the sea Justice has not
permitted him to live.’ (5) He however shook off the creature into the fire
and suffered no harm. (6) They were expecting that he would swell up or
suddenly fall down dead. But as they went on waiting and saw that nothing
amiss was happening to him they changed their minds and said that he was a
god.
(7) In the neighbourhood of that place there were domains belonging to
the chief man of the island, Publius by name, who received and entertained
us with kindly hospitality for three days. (8) It happened that Publius’ father
was sick and confined to bed with fevers and dysentery. Paul went in to him
and prayed, laid his hands on him and cured him. (9) When this happened the
others in the island who had illnesses approached and were healed. (10)
They1 bestowed upon us many honours, and when we left put on board the
things that we needed.
(11) After three months we set sail in a ship that had wintered on the
island, a ship of Alexandria, ship’s sign, The Dioscuri. (12) We landed at
Syracuse and stayed there three days. (13) 2Casting off from there we
reached Rhegium. After one day a south wind arose and we came on the
second day to Puteoli. (14) There we found brothers and3 were invited to stay
with them seven days; and4 in this way we made our journey to Rome. (15)
From there the brothers, having heard of our affairs, came to meet us as far as
Appius’ Market and Three Taverns. When Paul saw them he gave thanks to
God and took heart.
(16) When we entered Rome Paul was permitted to stay5 on his own with
the soldier who guarded him.

Bibliography
A. Acworth, as in (62).
H. J. Cadbury, JBL 44 (1925), 223-7.

1RSV, presented many gifts to us.

2RSV, from there we made a circuit; NEB, then we sailed round; NJB, from there
we followed the coast.
3NIB, had the great encouragement of staying a week.
4RSV, so we came to Rome; NEB, and so to Rome.
5NJB, in lodgings of his own.
1216
 63. FROM MALTA TO ROME. 28.1-16. 1217

J. Coppens, as in (47).
J. Dupont, in Kremer, Actes, 359-404 (= Nouvelles Études 457-511).
C. J. Hemer, as in (62).
N. Heutger, as in (62).
W. Kirschschläger, in Kremer, Actes, 509-21.
D. Ladouceur, as in (62).
G. B. Miles and G. Trompf, as in (62).
S. M. Praeder, as in (62).
J. Smith, as in (62).
A. Suhl, as in (62).
A. Suhl, BZ 36 (1992), 220-6.
G. W. Trompf, as in (62).
H. Warnecke, Die tatsächliche Romfahrt des Apostels Paulus (1987).
J. Wehnert, as in (62).

Commentary
The vigorous sea story of ch. 27 winds down quietly into something
like domesticity. Paul joins in the necessary task of finding firewood;
men are cold and wet and warmth is necessary. In the process he is
bitten by a snake. The availability of firewood and of snakes raises
the question of the identity of the island on which the travellers have
been cast. Can it have been Malta, or was it perhaps Meleda, an
island in the Adriatic? A commentary is no place for detailed
discussion of winds and tides, flora and fauna, local inhabitants and
their rulers. Some essential points are considered in the detailed
notes. Here the only point which it is worthwhile to observe is that,
whatever the snake population of either island may be, such a story
as we have in vv. 3-6 cannot settle the identification. No expert
naturalist was at hand to describe the snake and to confirm or dispute
the islanders’ belief that Paul was in danger of death; moreover, the
story is not one of a kind to evoke instant acceptance. If the promise
of Mk 16.18 was not drawn from the story in Acts but was traditional
it could have given rise to fulfilment narratives.
The paragraph proceeds with the natural reaction to Paul’s escape:
the ignorant islanders suppose that he is a god, an opinion that Luke
does not share (see on v. 6). Paul heals his host’s father’s sickness
and finds himself thronged by sick people who need healing. His
response leads to favourable treatment and a generous send-off from
the island—at a date earlier than one would have thought desirable
(see on v. 11).
Details of the journey from Malta to Rome are given. The seven
days’ wait in Puteoli is surprising, and so is the double reference to
1218 COMMENTARY ON ACTS

Rome in vv. 14 and 16. To be confident that there can be no

explanation of the seven day delay is as illogical as to insist on the
historicity of any particular explanation. It may have been the
author’s invention or not; we do not know. The two references to
Rome may very well be due to Luke’s desire to accommodate the
two missions, to Appii Forum and to Three Taverns; these may
themselves arise from different sources.
In this paragraph the use of We, dropped towards the close of ch. 27,
is resumed. It appears in vv. 1,2,7,10,11,12,13,14,15,16, so that it
is very natural to think of a travel narrative that identified the island
and the host who showed hospitality to Paul (and his friends—but not
necessarily to the ship’s complement of (2)76 persons) but did not
contain the wonders of vv. 2-6,8,9. It then took Paul to Rome, but did
not contain the references to Appii Forum and Three Taverns. So e.g.
Pesch (2.296), who thinks that vv. 1,2,10 continue the We-narrative,
while vv. 2b-6,7-9 are Luke’s rewriting of tradition (which however
was probably from the beginning handed down in this context).
Weiser (667) thinks that Luke himself formed the paragraph vv. 1-10
on the basis of two traditional episodes. Roloff (365; cf. Conzelmann
147) thinks that Luke freely created the two miracles, Schille (457)
that they were (though not originally) located in Malta, and that this
was why Malta was chosen as the place of the shipwreck. Cf. also
Kirschschläger in Kremer, Actes, 509-21. Schmithals (234) says that
the miracles show that Paul ‘ist in Wort und Tat ein vollgütiger Zeuge
Jesu, aber er ist dies wie die Zwölf Apostel und durch sie’. This is a
questionable assessment. The episode of the snake adds nothing to
e.g. a miraculous escape from prison by earthquake in Philippi; curing
a case of dysentery is poor stuff in comparison with raising up a dead
young man in Troas. And there is no reference to the Twelve, through
whom, according to Schmithals, Paul acts. It is probable that the
miracles, whatever we make of their historical trustworthiness, were
to be found, and were found by Luke, in the tradition, possibly in local
tradition. Luke’s supposed θεῖος ἀνήρ motif, made much of by e.g.
Haenchen (684) and Conzelmann (147), has already, and rightly, been
questioned by Weiser (669), Roloff (367), Pesch (2.299)—not to
mention Calvin (2.303).
It is true that Luke is here underlining some features of his
description of Paul. He is capable of working miracles; and he comes
through all kinds of dangers and opposition to his goal, Rome. It is
hardly correct to say (Roloff 368), ‘Lukas hat seine Vorlage, die V.
11-13, 14b, 16b umfasste, durch die Hinzufügung von V. 14a, 15,
16a so ausgestaltet, dass aus einem nüchternen Bericht von der
letzten Reiseetappe die Schilderung eines feierlichen Triumphzuges
wurde’, though Schmithals (236) could be right in thinking that these
verses originally followed ν. 1, and in arguing that since in 28.16-31
we hear no more of the Roman Christians ‘Lukas in v. 11-15 eine
 63. FROM MALTA TO ROME. 28.1-16. 1219

Vorlage zugrundelag’ (cf. Pesch 2.301). The delegations from Rome

are not described by Luke in triumphalist vein; their effect is to cheer
Paul and to give him courage for what awaits him.
It would have been easy to omit all reference to the brothers who
came out from Rome to meet Paul at Appii Forum and Three
Taverns, and this Luke would no doubt have done if it had been his
aim to suggest (contrary to fact) that Paul was virtually the founder
of the Roman church (so e.g. Haenchen 688). The view may well be
correct that we hear little of the Roman Christians because their
welcome was not as warm as Luke (and Paul) would have desired;
Philippians was probably written from Rome and bears witness to
those who preach Christ ‘out of envy and strife’ (Phil. 1.15,17). That
the word εκκλησία is not used is not significant; cf. e.g. 15.36 with
14.23. Luke’s silence in 28.17-31 is due to his desire (see pp. 1235,
1237, 1239) finally and decisively to show that Paul is a good and
loyal Jew; that Jewish rejection of the Gospel is the fulfilment of
prophecy; and that the church is committed to the Gentile mission.

1. Ch. 27 ends with the picture of the seafarers struggling through

the water to dry land, which all of them safely reach. Their safe
arrival is picked up here by διασωθέντες, for which see 27.43f. The
next thing (τότε) was recognition (ἐπιγινώσκειν) of the place in
which they found themselves. The island was called Μελίτη. How
the stranded travellers were able to recognize the land Luke does not
say. The local inhabitants are mentioned in the next verse; perhaps it
may be assumed that they had already appeared and been questioned,
though if this were so we learned, or we were informed, would have
been more suitable than we recognized. It may be, of course, that,
once ashore, experienced travellers would know where they were.
It is usually supposed that Μελίτη (in B* lat syh bo, Μελιτήνη; see
Μ. 2.359—an adjective agreeing with νήσος (understood)? or
dittography within Μελίτη ἡ νήσος (Metzger 500)?) refers to the
island now known as Malta, where the traditional site of the wreck is
known as St Paul’s Bay. Early writers seem to have made little or no
attempt to identify the scene of the wreck, but in the 8th century
Constantinus Porphyrogenitus (De Administrando Imperio 36)
thought not of Malta but of an island in the Adriatic now known as
Meleda, Melite, or Mljet, in antiquity as Cephallenia. This belief, that
Paul was shipwrecked in the Adriatic, has been argued at length by
H. Warnecke, Die tatsächliche Romfahrt des Apostels Paulus, SB
127 (Stuttgart, 1987), on the basis of detailed nautical, topographical,
and meteorological observations, which could be pursued here only
by throwing the commentary out of proportion and could be assessed
only by a commentator with adequate knowledge of the sciences
involved. Some relevant facts will be mentioned as they arise in the
following verses; here that a gale that threatened to drive the ship on
1220 COMMENTARY ON ACTS

to the Syrtes (27.17) must have been a north easter, and it is hard to
see how it could have led to a wreck on the east side of the
Adriatic.
Malta had been a Roman island since it was captured from the
Carthaginians in 218 BC (Livy 21.51). On its government and some
other features see on v. 7 (and the bibliography in Schneider 2.400, n.
1). It has been said that the island was appropriately named, since in
Phoenician mělêṭā = refuge, escape. There is no indication that Luke
was aware of this.
Bauemfeind (276) thinks that ή νήσος (with the article) refers to
the island which Paul had foretold (27.26); more probably it is
simply the one on which we have just landed.
In the present verse ‘we’ (ἐπέγνωμεν) is left undefined; as the
narrative proceeds it ceases to refer (as it had done on occasion in ch.
27) to the ship’s company as a whole and means Paul and his
companions.
After the excitement of the voyage the story at this point seems to
settle down. Davies (Land 282) remarks that there is no urgency in
getting to Rome; there is ‘almost a “tourist” air about the voyage’. Is
this because Luke has here a travel handbook source, different from
the theologically motivated source that has at least some relation
with Paul?
Weiser (666) quotes Heutger, who, though agreeing that the wreck
took place on Malta, disagrees with the traditional location in St
Paul’s Bay; ‘... treffe der Ausdruck dithalassos viel besser zur
nächsten, westlich gelegenen Mellieha-Bucht, wo heute noch ausser
der Bai selbst ein zweite See zu sehen sei, der sich jeweils im Winter
landeinwärts bildet’.

2. The local inhabitants are described as οἱ βάρβαροι. In Acts the

word is used only here and at v. 4; cf. Rom. 1.14; 1 Cor. 14.11; Col.
3.11. Its primary meaning related to language and was presumably
onomatopoeic. The βάρβαρος was one who did not speak Greek and
whose words therefore sounded (to a Greek) like a meaningless ba-
ba-ba. He was someone other than a Greek; the word was used
‘specially of the Medes and Persians’; after the Persian war it came
to mean ‘brutal, rude’ (LS 306). This verse suggests that the latter
meaning is not in mind; these barbarians were not brutal and rude,
but showed courtesy and kindness. The word therefore probably
retains its linguistic reference (as it does in 1 Cor. 14.11). This is
used by Warnecke (111-18) as an argument in favour of Cephallenia
rather than Malta as the scene of the shipwreck. Before the first
century AD Malta had been both hellenized and romanized. Inscrip-
tions in both Greek and Latin are found, together with ruins of
Roman villas, theatres, and baths. Julius Caesar settled some of his
veterans in Malta, and a little later the island received the citizenship,
 63. FROM MALTA TO ROME. 28.1-16. 1221

as is confirmed by coins (see also Strabo 7.833). On the other hand,

Zahn (842) (quoted by Haenchen 681) wrote that it is proved by
inscriptions (details conveniently in Hemer 152) that ‘der gemeine
Mann, die Fabrikarbeiter und Packträger, die Handelsgärtner und
Hundezüchter auf Melita nur ihre punische Muttersprache verstehen,
sprechen und lesen konnten.’ Similarly Page (264): ‘Diod[orus]
Sic[ulus] v. 12, ἔστι δὲ ἡ νῆσος αὕτη Φοινίκων ἄποικος. Their
language therefore was probably Punic.’ Cf. Cicero, In Verrem 2.4.46
(103), Itaque in his inscriptum litteris punicis fuit ... It may be
assumed that both in Malta and Meleda there were uneducated people
who could not, and educated people who could, speak Greek (or Latin
or both), and that the former would qualify for the term βαρβαροί.
The barbarians certainly greeted the shipwrecked men in a friendly
fashion. For παρεῖχον the middle might have been expected (Μ.
3.56); for φιλανθρωπία cf. 27.3 (φιλανθρώπως). For οὐ τήν
τυχοῦσαν cf. 19.11; the word is characteristic of the second part of
Acts. There is an excellent linguistic parallel (though the meaning
reverses that of the present clause) in BGU 1.36 [= 436].9f. (a
Fayyum papyrus of the 2nd or 3rd century AD), ὕβριν οὐ τήν
τυχοῦσαν συνετελεσαντο.
Their kindness was shown in that they lit (ἅψαντες; <a> has
ἀνάψαντες with no difference in meaning) a fire and welcomed, or
received, us all, or simply brought us to it. Action of this kind was
desirable because of the rain and the cold. One might have expected a
reference to the wetting in the sea that all had experienced. The rain
was ἐφεστῶτα; perhaps, that came on suddenly, unexpectedly; cf. 1
Thess. 5.3; also Polybius 18.20.7 (Blass 284), διά τόν ἐφεστῶτα
ζόφον. Warnecke (100-2) observes that in Malta the average rainfall
in October is 83.3 mm, whereas in Melida, off the Dalmatian coast,
the rainfall is heavy. He quotes the geographer Partsch (Kephallenia
33f.), ‘... stärker und anhaltender setzen sie [Gewitter] erst im
Oktober ein. Noch pflegt ein kleiner Nachsommer zu folgen ehe der
November und Dezember ihre gewaltigen Regenmengen ausschüt-
ten. In diesen Monaten rauschen die zahlreichsten und kräftigsten
Niederschläge nieder.’ As for temperature (Warnecke 102-4), the
average minima and maxima are for October 17.2°C and 27.8°C, for
November, 12.2 and 23.3. The lowest of these would certainly not
strike someone from Northern Europe as cold. Cephallenia is much
colder. Two points, perhaps, to Warnecke, but men in wet clothes,
with a good breeze blowing, would not be sorry to see a fire.
Kindness to strangers is well illustrated by Betz (94, n. 6) with a
quotation from Lucian, Verae Historiae 1.29, οὐδὲν ἡμᾶς ἠδίκουν,
ἀλλὰ καί ἐπὶ ξενίᾳ ἐκάλουν. Cf. Verae Historiae 2.46.
For προσελάβοντο, Ψ 614 2495 pc lat have προσανελάμβανον
(reficiebant), refreshed us, which Bruce 1.470 regards as the better
reading.
1222 COMMENTARY ON ACTS

3. συστρέφειν is used in a variety of ways: of animals gathering

themselves to spring; of a man screwing up his eyes; of soldiers
rallying, forming a compact body; it is not easy to find a parallel to
what is evidently the meaning here. Paul gathered and twisted
together a bundle of twigs with a view to feeding the fire. πλήθος
elsewhere in Acts refers to a company of people, here to a quantity
(colloquially in English, a lot) of sticks, φρύγανα, for the fire. So
Xenophon, Anabasis 4.3.11, φρύγανα συλλέγοντες ὡς ἐπί πύρ.
The genitive absolute (συστράψαντος ... τοῦ Παύλου) is com-
bined with another genitive participle, ἐπιθέντος. Paul put the bundle
of sticks on the fire. Escaping the heat (ἐπὸ τής θέρμης), a snake
(ἔχιδνα) came out (ἐξελθούσα; H L P Ψ 049 323 614 945 1241 <a>
have διεξελθοῦσα—came out through the flames, through the
sticks?). It appears in the following verses that the snake did Paul no
harm. Luke plainly regards this as a miracle, and therefore under-
stood the word ἔχιδνα in its proper sense; he also represents the
native inhabitants of the island as sharing his view—a man bitten by
one of ‘their’ snakes should swell up and fall down dead. (v. 6). The
occurrence of snakes in Malta and Cephallenia is discussed by
Warnecke (108-10). There were, it seems, no poisonous snakes on
Malta. ‘Auf Malta gibt es weder Sandvipem noch überhaupt eine
Vipernart, ja, nicht einmal irgendwelche Giftschlange: “Die drei auf
Malta lebenden Schlangenarten sind ungiftig” ’ (Warnecke 108,
quoting H. Egger, Malta 159). Cephallenia on the other hand has
about twenty kinds of snake, including the Vipera ammodytus, that
is, the ἔχιδνα. This seems to be a point in favour of Cephallenia as
the scene of the incident; alternatively, it means that the story is
fictitious (or possibly, belongs to another setting), and was written by
one who did not know what kinds of snake were to be found on
Malta. A further possibility is that in the first century Malta was
richer in snakes than it is now, and possessed poisonous snakes.
The snake fastened on Paul’s hand; for καθῆψε, the middle
καθήψατο might have been expected. This is read by C 36 453 614
1891 2495 pm, but this is plainly copyists’ ‘improvement’. Begs.
4.341 quotes Epictetus 3.20.10, ὁ μὲν τοῦ τραχήλου καθάπτων; this
however is in wrestling.
For ἀπὸ τῆς θέρμης cf. Lucian, Dialogi Marini, 11.2, ἡ θέρμη δέ,
ὡς φῆς, ἀπὸ τοῦ πυρός.
Luke himself would probably be surprised by the comment
(Calvin 2.299), ‘According to His secret purpose, the Lord directed it
[the snake] to bite Paul because He saw that it would be for the glory
of His Gospel.’

4. The βάρβαροι (v. 2) saw what was happening; the creature

(now τό θήριον, used of snakes in a variety of places: BA 733f. cite
Diodorus Siculus 20.42 2; Polyaenus 2.3.15; Aretaeus 159.8; 163.2;
 63. FROM MALTA TO ROME. 28.1-16. 1223

Justin, 1 Apology 60.2; Galen 4.779k) was hanging from his hand.
They drew a natural conclusion. Paul has deserved death; having
escaped drowning he is to be punished by the snake.
For διασώζειν see on 27.43f. In the NT ἐᾶν is almost exclusively a
Lucan word. In the NT πάντως is used only by Luke and Paul. Begs.
4.341 translate it perhaps, referring to H. J. Cadbury, JBL 44 (1925),
223-7. LS 1301 do not have this meaning; BA 1232 mention
Cadbury’s interpretation without comment. It is not convincing.
It might be better to spell δίκη with a capital Δ. The word occurs at
2 Thess. 1.9; Jude 7; nowhere else in the NT (unless at Acts 25.15,
where Ε Ψ <a> have δίκην in place of καταδίκην. The barbarians
may well have personified Justice as a divine being. Cf. Sophocles,
Antigone 538f., ἀλλ’ οὐκ ἐάσει τοῦτὸ γ’ ἡ δίκη σ’, ἐπεὶ οὐτ’
ἠθέλησας οὐτ’ ἐγώ ’κοινωσάμην; Oedipus at Colonus 1381f.; Hes-
iod, Theogony 902; Plutarch, De Exilio 5 (601B); Alexander 52
(695Α), τὴν Δίκην ἔχει πάρεδρον ό Ζευς καί τήν Θέμιν.
The thought that sins are not left unpunished (and that a snake bite
may be used) is illustrated in StrB (2.772) by a long quotation from
Sanhedrin 37b.
For a very close parallel to Luke’s story see the Anthologia
Palatina 7.290—a shipwrecked sailor is killed by a viper.
5. ὁ μέν οὖν: he, however. For the adversative use of μὲν οὖν cf.
14.3; 17.17; 25.4 (Moule, IB 163).
ἀποτινάξας: Lk. 9.5. ἐκτινάξασθαι: Acts 13.51: 18.6. Cf. Eur-
ipides, Bacchae 253, οὐκ αποτινάξεις κισσόν;
For ἔπαθεν οὐδὲν κακόν cf. Plutarch, Pericles 34 (171Α), πολλά
μὲν δρῶντες κακά τούς Άθεναίους, πολλὰ δὲ πάσχοντες ὑπ’
έκείνων.
For the snake, cf. Lk. 10.19 as well as Mk 16.17f.; in the
background of both, Ps. 90 (91).13. In Berakoth 33a there is a fine
story of the snake that perished as a result of its attempt to bite R.
Hanina b. Dosa (StrB 2.772).
6. The tenses in this verse are to be noted. προσεδόκων (imper-
fect, they were expecting) that he would begin to swell up
(πίμπρασθαι, present) or fall down (καταπίπτειν, present). But
while they went on for a long time expecting (προσδοκώντων,
present) and observing (θεωρούντων, present) that nothing amiss
was happening (γινόμενον, present) to him, they changed their
minds (μεταβαλόμενοι, aorist—but 648 <a> have –αλλ-, present)
and began to say (ἔλεγον, imperfect) that he was a god. Luke’s use of
tenses is not always so impressive. Cf. 27.7; see Moule, IB 101. For
μή with the participle (γινόμενον) in oratio obliqua see Μ. 1.239.
For the form ἐμπίπρασθαι, which occurs in 323 945 pc see BDR
§ 101, n. 69 (dissimilation). For μέλλειν πίμπρασθαι as a substitute
for the future infinitive see BDR § 356.3, n. 4. For the very hard
1224 COMMENTARY ON ACTS

construction in which the genitive in the genitive absolute (αυτών

προσδοκώντων) becomes the subject of the main clause see BDR §
423.3, n. 9, but cf. Delebecque (136), ‘le génitif absolu peut, en bon
grec, se rapporter à un nominatif sujet.’
Neither άτοπος nor πίμπρασθαι is to be regarded as a medical
term. Doctors are not alone in being aware of accident and illness,
and their symptoms. Cf. Josephus, Ant. 11.134, μηδέν κατά τήν
ὁδὸν παθεῖν ἄτοπον; 3.271,... καί τήν γαστέρα πρησθεῖσαν ...;
Num. 5.21.
The incident does not support the view that Luke intended to
represent Paul as a θεῖος ἀνήρ. Weiser (669) observing that the
impression that Paul was a god is not corrected here as it is at 14.15
adds that Luke ‘lässt sie als heidnischen Irrtum stehen im Vertrauen
darauf, dass die Leser Paulus und der Gott, dem er dient, zu
unterscheiden wissen’. One can go further; the erroneous opinion
was held by barbarians, whom a Greek could not possibly think to be
right. See on v. 2. Luke and his readers would no doubt agree with
Calvin (2.301): ‘If it was necessary to choose one or the other, it was
better to be regarded (sic) a murderer than a god.’

7. It is hardly necessary to ask what noun should be supplied with

ἐν δὲ τοῖς. The phrase τά περί τόν τόπον ἐκεῖνον means the
neighbourhood of that place (the place of the wreck). It is possible
however that we should think of τοῖς χωρίοις: Among the estates
surrounding that place were estates belonging to ... The general
meaning is clear. ‘Le singulier χωρίον signifie “champ”; le pluriel
doit désigner un ensemble de champs, donc une “campagne”, où un
“domaine”, selon l’importance du terrain’ (Delebecque 136). Dele-
becque adds that the meaning here will be domaine since the estate is
the property of the chief man of the island. See 1.18. for ὑπῆρχεν
with the dative as a way of expressing possession cf. 3.6; 4.37; see
BDR § 189.1, n. 1. The owner was Publius; the variant spelling
Πουπλίῳ (P74vid 81 104 945 1739 pc) underlines the connection with
the Roman name Publius (BDR § 41.1, n. 3). He is described as the
πρώτος τής νήσου. The use of this title in Malta is confirmed by
epigraphic evidence; see especially IG 14.601: Λ. Καστρίκιος
Κυρ(είνα) Προύδηνς ἱππεὺς 'Ρομ(αίων), πρώτος Μελιταίων καί
πάτρων, ἄρξας καί ἀμφιπολείσας θεῷ Αὐγούστφ. Cf. CIL
10.7495, municipi Melitensium primus omnium: on this however see
Hemer (153). It seems that in this inscription primus is not a title but
claims that the person named was the first to make certain benefac-
tions. It has often been understood that the inscription(s) referred to
the chief Roman administrator of the island, which had long been
joined with and governed by the procurator of Sicily. Malta however
was separated from Sicily in the early years of Augustus, and CIL
10.7494 (proc. insularum Melti. et Gaul.) indicates that Malta and the
 63. FROM MALTA TO ROME. 28.1-16. 1225

neighbouring small island of Gaulos (modern Gozo) were governed

by a procurator. It is now usual to see in the Protos a local native
officer. Cephallenia was a civitas libera (Pliny, Natural History
4.54), in which Roman interests were looked after by a legatus of the
governor of Achaea. This office might well have borne the title of
primus; Warnecke (123) draws attention to one holder of it, at about
the time that must be assigned to Paul’s journey, whose name was
Publius Alf. (= Alfius or Alfessus) Primus. This must be taken with
the fact, which also has often been observed, that it is unusual for a
Roman official to be referred to by his praenomen alone, as Publius.
Warnecke suggests that the solution of two problems—the use of the
name Publius and the title primus (πρώτος)—may lie here. Luke
gives a confused recollection of the name Publius Primus. The
suggestion is not convincing. The title πρώτος was used in Malta,
whether of a Roman or a local official. Lüdemann (272) thinks that
Publius and πρώτος were both drawn from tradition.
Publius received and for three days entertained us (ἡμᾶς), now in
all probability no longer the whole company from the ship but Paul
and his companions. There can be no doubt that the statement,
correctly or incorrectly, represents the author of the narrative as a
close personal associate of Paul’s.
φιλοφρόνως, here only in the NT, is not uncommon; e.g. Jose-
phus, War 6.115, Καῖσαρ δ’αὐτοὺς τά τε ἀλλὰ φιλοφρόνως
ἐδέξατο, with kindly hospitality. It is not said that Publius (or anyone
else on the island) became a Christian.

8. For ἐγένετο with accusative and infinitive see Introduction, p.

xlvi.
συνέχεσθαι has many meanings; to be afflicted by illness is one
(e.g. Lk. 4.38; Josephus, Ant. 13.398, τεταρταίῳ πυρετῷ
συσχεθείς). κατακεῖσθαι means that Publius’ father was laid up in
bed (e.g. Herodotus 7.229.1). πυρετός is used in the plural by
medical writers; BA 1462 cites Galen, De Diff. Febr. 1.1 (7.275) and
Alexander of Aphrodisias, De Febribus Libell. 31. This proves
nothing regarding the identity of the writer; the plural was used also
by non-medical writers, e.g. Aristophanes, Wasps 1038; Plato,
Timaeus 86a (dysentery is mentioned in the same context). P74 has
the singular πυρετῷ. δυσεντέριον is a late form of δυσεντερία (Μ.
2.125,342).
Paul went in to the sick man (that is, presumably, into his sick
room) and cured him by prayer and the laying on of hands. In πρὸς
ὃv ... αὐτφ Delebecque (136) sees the Hebraism of a relative
followed by a pronoun—wrongly. Daube, NTRJ 234f., thinks that
behind ἐπιθείς there will lie either or but not (see on
6.6). This is not based on the use of ἐπιτιθέναι and faces the
formidable contrary evidence of 1QapGen 20.28, 29, one of the rare
1226 COMMENTARY ON ACTS

Jewish examples of healing by the imposition of hands. The passage

runs: I prayed for [...] and laid my hands upon his head; the
plague was removed from him ... On this passage see Fitzmyer,
Aramaean 96f. Cf. James 5.13f.
Hemer (153f.) suggests that ‘the cause of fever may well have
been that associated with this island, “Malta fever”, discovered in
1887 to be caused by an endemic microorganism Micrococcus
Melitensis, which infected the milk of the Maltese goats’. A point in
favour of Malta against Cephallenia, but a small one; there are other
causes of fever.
Paul prayed that the cure might be effected. ‘By praying Paul
makes it plain that he is not the one responsible for die miracle, but
only the minister, so that God may not be defrauded of His glory’
(Calvin 2.303). Later writers make the same observation and relate it
to the view that Paul is represented in Acts as a θεῖος ἀνήρ. So
Roloff (367); Pesch (2.299); also Weiser (670): ‘Darin ist ein
wichtiges Korrektiv gegenüber der verbreiteten hellenistischen Auf-
fassung von Wundertätern als “göttlichen Menschen” (theioi
andres) zu sehen.’
9. The cure of Publius’ father naturally kindled hope in other
inhabitants of the island who were ill (ἔχοντες ασθενείας); these
came to Paul for healing and were cured. There is no ground for the
suggestion that while Paul (miraculously) cured Publius’ father
(ἰάσατο) the other inhabitants were given non-miraculous attention
(ἐθεραπεύοντο) by Luke and his medical team. Cf. ἐθεραπεύοντο at
5.16.
H, taking τούτου to be masculine and to refer to the sick man,
adds ὑγίους. καί (before οἱ λοιποί) is omitted by B gig vgcl.
10. The possibility cannot be excluded that πολλαῖς τιμαῖς
ἐτίμησαν means paid us many fees (for medical and other services),
but it is much more probable that it means honoured us with (i.e.
bestowed upon us) many honours. The expression was current (see
BA 1628-1630; e.g. Josephus, Ant. 20.68, ταύταις μὲν δὴ ταῖς
τιμαῖς ό Ίζάτης ὑπὸ τοῦ τῶν Πάρθων βασιλέως ἐτιμήθη), Luke
has not suggested that Paul and his colleagues were working in the
ordinary sense as doctors, and material needs are seen to in the next
clause. For the alternative interpretation see Sirach 38.1, τίμα ιατρόν
πρὸς τὰς χρείας τιμαῖς αύτού.
For ἀνάγεσθαι see on 13.13; 16.11. The grateful inhabitants of the
island (in Luke’s source, not the whole population but Publius and
his household—so Pesch 2.299) saw to it that the Christian travellers
should not be in need during the rest of their journey. No conversions
are mentioned.
ἐπέθεντο could mean they bestowed on us or they put on board.
syh* by adding in the ship chooses the latter alternative.
 63. FROM MALTA TO ROME. 28.1-16. 1227

This story seems to be independent of the sea story of ch. 27; that
Paul was a prisoner on his way to trial is not hinted at.

11. How many of the (2)76 who had set out on the voyage
remained on the island for three months cannot be determined; the
first person plural of ἀνήχθημεν (ἤχθημεν in H 049 6 326 1891 2495
pm; on the compound verb see v. 10) may refer only to Paul and his
companions (including of course the military escort). Ships were
commonly laid up during the winter (this had been the intention of
all on Paul’s ship; the only disagreement had been in the choice of a
port to winter in: 27.11f.), and the centurion would probably have
little difficulty in finding a ship that had spent the winter
(παρακεχειμακότι) in an island port and had room enough to take on
the essential passengers.
τρεῖς μήνας gives rise to some difficulty. As a period of delay on a
sea voyage it can be paralleled (Josephus, War 2.203, συνέβη
χειμασθῆναι τρεις μῆνας ἐν τή θαλάσση), but Paul’s voyage began
after the Day of Atonement (27.9) and presumably not very long
afterwards or there would have been no point in using the Day as a
date; allowing for all the time references given in the narrative, one
would suppose that Malta was reached by the end of October. Three
months would elapse by the end of January, a very early date for
sailing to resume. See the passages cited on 27.9 and cf. Apuleius,
Metamorphoses 11.5: On the day of Isis (5 March) it could be said
sedatis hybemis tempestatibus, et lenitis maris procellosis fluctibus,
navigabili jam pelago. Jam does not mean that sailing began on this
day but it is implied that it had not been in progress long. There were
however reasons why a grain ship (see below) might be in a hurry
(see Hemer 154), and the crossing to Rhegium was one that might
have been made early (Preuschen 156). ‘Three months’ is in any case
to be regarded as an approximation, though Bauemfeind (278) insists
that the journey recommenced in January.
The ship was an Alexandrian. ’Αλεξανδρῖνoς has a Latin adjec-
tival termination, borrowed into Greek; cf. the use of Άλεξανδρεύς
(of persons) at 6.9; 18.24 (see BDR § 5.2, n. 11). The ship was in all
probability one of the grain fleet, which regularly brought com from
Egypt to Rome. It is unlikely that such a ship should winter in
Cephallenia.
The sense of the last two words in the verse is clear but their
construction is not. Ships in antiquity often bore an image, or images,
of a god, whose name provided the name of the ship (Lucian,
Navigium 5,... τήν ἐπώνυμοντῆς νεὼςθεὸν ἐχοῦσατὴν 'Ισιν ...);
this one carried and was known by the sign of the Dioscuri.
παρασήμῳ and Διοσκούροις are both in the dative case. παρασήμῳ
may be an adjective; if so, it would agree with πλοίῳ: a boat marked
with, indicated by, the Dioscuri. This, if admissible, makes good
1228 COMMENTARY ON ACTS

sense; it seems to be preferred by BA 1257. But the adjective

παράσημος seems almost always to have an unfavourable meaning.
The meanings listed in LS 1323 are 'marked amiss or falsely,
counterfeit ... falsely stamped ... of words and phrases, false,
incorrect... eccentricity of style ... marked by the side, noted ...
marked, notorious for ... remarked as ... conspicuousness ...
indicative'. In nearly all these meanings the effect of the com-
pounded παρά is evident. It seems more probable that Luke is using
the noun παράσημον, for which LS give as a general meaning
distinguishing mark, with examples including ensign of a ship, or of
a city, or of patricians and plebeians, insignia praetoria (παράσημα
στρατηγικά), birthmarks (παράσημα σωματικά), password. If this
way of taking παρασήμῳ is adopted one is left with the problem of
the datives. Turner (Μ. 3.243) thinks that this may be an example of
a dative absolute, mentioning Mt. 14.6 and Mk 9.28 (P45) as other NT
examples (of which the latter must be considered very doubtful). He
refers to Moule (IB 45), but Moule regards the Matthean and Marcan
passages as the only NT examples and would presumably therefore
not consider Acts 28.11 one. There is a long note in BDR § 198.7, n.
11: ‘Als Dat. soz. “mit den Dioskuren als Schiffszeichen” (Ramsay,
Luke 36f. als korrekt nach Inschriftengebrauch); vielfach als Dat.
instr. aufgefasst: “gekennzeichnet durch die Dioskuren” (Bauer sv
παράσημος Haenchen zSt; vgl Plut. mor. 823B [= Praecepta
Politica 31] ἐπιφθόνοις παράσημος “durch Gehässigkeiten sich
bemerkbar machend”); vielleicht aber auch nur mechanische Dekli-
nation eines registraturmässigne (πλοῖον) παράσημον Διοσκούροι
“ein Schiff, Schiffszeichen die D.”’ The last suggestion is perhaps
the best. Similarly Page (266), quoting an inscription,... gubematore
navis parasemo Isopharia. The inscription is given in full in Smith
(269). See also Plutarch, Septem Sapientium Convivium 18 (162A);
Mulierum Virtutes 9 (247f.).
Διοσκούροις is the reading of most MSS; Ρ74 Ρ* Ψ 81e 104 326
453 2464 al have Διοσκόροις. This is the Attic spelling; –ου- is the
Hellenistic, though it occurs in Plato and Thucydides (BDR § 30.3, n.
6). ‘Διόσκουροι, ὀρθότερον Διόσκοροι. γελάσε οὖν τούς σὺν τῷ υ
λεγοντας' (Phrynichus 212; see Lobeck quoted in Rutherford, 310f.).
See also Μ. 2.88. The minuscules that have Διοσκόροις ‘are mainly
of the I-groups, and it may well have stood in the “Western” text’
(Ropes, Begs. 3.251).
The Dioscuri, Castor and Polydeuces (Latinized as Pollux), were a
natural choice as patrons of the ship; they were called upon by sailors
as helpers in time of need. Epictetus 2.18.29, τού θεού μέμνησο,
ἐκεῖνον ἐπικαλοῦ βοηθόν καί παραστάτην ὡς τοὺς Διοσκόρους
ἐν χειμῶνι οἱ πλέοντες.

12. For κατάγεσθαι (the opposite of ἀνάγεσθαι in ν. 11) see 27.3,

 63. FROM MALTA TO ROME. 28.1-16. 1229

elsewhere in the NT it is used (in a different sense) only at Rom.

10.6. We landed in Syracuse and stayed there three days (ημέρας
τρεις; B has the unexplained dative ἡμέραις τρίσιν). No reason is
given for the delay; it was early in the sailing season and an
unexpected spell of bad weather is as good a guess as any.
Syracuse, originally (in the 8th century BC) a Corinthian colony,
was an old-established Greek city on the east coast of Sicily. It is
perhaps twice as far from Cephallenia as from Malta, but neither
journey would have been impossible.

13. περιελόντες ( Β Ψ (gig)) is so surprising a reading that the

variant περιελθόντες (P74 A 066 <a> lat sy) is easily explained. But
περιελόντες may be too difficult to be accepted. Its usual meaning is
that suggested by its etymology; it means to take away from round
something. No object is expressed, but BA 1301 (and others) explain
it as referring to the raising of anchors. See Hort, Introduction 226f.
(’the elliptic employment of transitive verbs being common in Greek
nautical language as in English’). See 27.40. Metzger (500f.) refers
to Hort, but also observes that περιελόντες could easily be derived
from περιελθόντες through the simple dropping of Θ before O.
Given the habit noted by Hort (cf. Delebecque 137) it seems best to
take the participle to mean ‘weighing anchor’, perhaps ‘casting off’.
It is in any case clear that the ship headed northwards for the strait of
Messina and came (καταντάν as at 27.12; the word occurs 9 times in
Acts, 4 times in the rest of the NT) to Rhegium. After one day, to
which no incident is assigned, a south wind arose—exactly what was
needed for a voyage up the west coast of Italy. The name of the wind,
νότος (it ought perhaps to be printed Νότος; cf. 27.14), has no
article; see Μ. 3.172; BDR § 253.5, n. 7.
With the aid of this wind δευτεραῖοι ἤλθομεν, we came as
‘second day’ men. See Μ. 3.225; δευτεραῖοι is an adverbial
adjective. This construction became increasingly popular in post-
classical Greek, and ‘eventually became the regular way of forming
adverbs in M[odem] Gr[eek]’. Cf BDR § 243.1. We came to Puteoli
(Greek name Δικαιαρχία; modern, Pozzuoli); for the formation of
Ποτίολοι from the Latin name see BDR § 41.1, n. 2.
Puteoli was founded by Ionians, taken by the Romans in the
Second Punic War, and made a Roman sea colony in 194 BC.
Travellers to Rome often landed at Puteoli and made the rest of the
journey (about 130 miles) by road (Cicero, Pro Piando 26 (65) ...
decidens e provincia, Puteolos forte venisse ...; Josephus, Ant.
17.328; 18.248, προσέπλευσαν ἀμφότεροι Δικαιαρχεία). ‘Puteoli
was the port where passengers were set ashore, though the cargo of
grain was taken up to Portus, the new harbour built by Claudius at
Ostia, by the mouth of the Tiber (cf. Sen[eca], Ep. Mor. 77.1)'
(Hemer 154f.). There had been a Jewish colony at Puteoli since at
1230 COMMENTARY ON ACTS

least 4 BC (Josephus, War 2.101-110; Ant. 17.324-38). See further

NS 3.81 (and 3.110f. for a Tyrian colony, bearing witness to contact
with the east).
14. For the first time since they reached the west the travelling
party encountered αδελφοί, that is, Christians. The party (in the first
person plural) were invited, or encouraged (παρεκλήθημεν; a perfec-
tive aorist, prevailed upon, according to Μ. 3.72; cf. Zerwick § 252)
to stay seven days. Presumably this suited the ship’s programme of
unloading and loading (but see below). Haenchen (687) thinks that
Luke added this first part of the verse so that news could be brought
to Rome and make possible the deputation of v. 15; it does not seem
probable that Luke would think of the necessity of making this kind
of connection.
For ἐπιμεῖναι Η Ψ 049 326 614 2464 2495 have ἐπιμείναντες. If
this is read, παρεκλήθημεν must mean we were encouraged (or
something like this) rather than we were asked: We were encouraged,
having stayed with them seven days. This alleviates the difficulty of
the prisoners’ being invited as if they were free men (Metzger 501).
This may well have been the origin of the reading. The delay is the
more surprising in that the journey was continued by land. Pesch
(2.303) comments, ‘Auch die Soldaten waren als Schiffsbrüchige in
Puteoli angekommen und mussten sich für die Landreise nach Rom
neu ausrüsten usw.’ They had had three months in Malta to recover,
but presumably they had lost much of their equipment in the wreck.
The words οὕτως εἰς τήν 'Ρώμην ἤλθαμεν seem at first sight to
constitute the climax of the book; but in fact the travellers did not
reach their destination till v. 16. It has been suggested that in v. 14
Rome refers to the Ager Romanus, but this is inconsistent with
κάκεῖθεν in v. 15. The best way of taking the words is given by
Marshall (419) (similarly Delebecque 137): And in this way we made
our journey to Rome.
There is no confirmatory evidence for the existence of Christians
at Puteoli at this time. Herculaneum and Pompeii are not far away; at
Herculaneum there are the supposed cross and altar; at Pompeii two
examples of the ROTAS-SATOR square. For a properly sceptical
discussion (with bibliography) of this highly problematical (and in
the end unconvincing) evidence see Hemer (155f.).
15. In the present context the last words of v. 14 (see above)
cannot mean that Paul and his companions had entered the city of
Rome; but they were well on the way; the final stretch from Puteoli
could be covered by a good walker in five days (Schneider 2.407;
Stählin 324). They were also now in contact with the city, since the
(the article oἱ is omitted by B, perhaps because it suggested that all
the Christians of Rome made the journey) brothers (that is, Chris-
tians, as in v. 14) came (ήλθαν; Ψ have ἐξῆλθαν to meet us—the
 63. FROM MALTA TO ROME. 28.1-16. 1231

first person plural is still in use. They came as far as Appius’ Market
and Three Taverns; these were distinct places, and we must suppose
that two groups came from Rome. This may mean only that one (an
advance guard? Pesch 2.303) had more leisure, or could walk better,
than the other; it has been suggested that one consisted of Jewish the
other of Gentile Christians (Rackham 498), but there is nothing in the
text to support this view. They came because they had heard τά περί
ἡμῶν (these words are omitted by gig vg syp), our affairs; about us
would probably serve in English. It is natural to suppose (see above)
that they had heard from Puteoli; the seven days of v. 14 might have
allowed messengers to reach Rome in time to arrange a deputation.
They had presumably had some notice of an intended visit from the
Epistle to the Romans, but this would not tell them when to meet the
writer. Luke is not interested in the details of such arrangements.
On the two places see Hemer (156). Paul would come first to Appii
Forum, 43 Roman miles from Rome: the 43rd milestone is still extant
(CIL 10.6825). Part of the journey is described by Horace (Satires
1.
5.1-4):
Egressum magna me accepit Aricia Roma
hospitio modico ...
... inde Forum Appii,
differtum nautis cauponibus atque malignis.
It thus took Horace two days to get from Rome to Appii Forum,
though the journey should (lines 5, 6) have been done in one—not
however on foot. Cf. Cicero, Ad Atticum 1.13.
Tres Tabemae (deriving its name probably from the institutions
named) was 33 Roman miles from Rome, and thus not far from Appii
Forum. It was on the Via Appia, ‘probably at the point where the
road from Antium crosses it, near the modern Cisterna’ (Knowling
545). Cf. Cicero, Ad Atticum 2.10, Ab Appii Foro, hora quarta.
dederam aliam paulo ante Tribus Tabemis.
How there came to be Christian brothers in Rome we are not
informed. It is possible that they were to be found there as early as
18.2; see the note. Romans makes it clear that the church in Rome
was not founded by Paul. If it had a special relation with Peter one
would expect to find some reference or allusion to this in the epistle;
there is none, but the possibility cannot be excluded; see 1 Clement
5.4; Ignatius, Romans 4.3. For Peter as a travelling missionary see FS
Michel (1963), 1-12. Conzelmann (149) (cf. Haenchen 688; Weiser
674) makes the point that Luke wishes at the same time to show that
the Roman church already existed and welcomed Paul, and that Paul
was a pioneering apostle; hence the speedy disappearance of the
Roman Christians. To say (Haenchen 688) that there were Christians
but not an organized εκκλησία in Rome would be convincing only to
one who thought that a specific church order was necessary for full
1232 COMMENTARY ON ACTS

Christian life. Schille (475f.) thinks that the narrative presupposes an

organized community; Paul is given a triumphal entry (ἀπάντησις).
Paul himself (as Luke depicts him) was not distressed by the
absence of church order (if indeed it was absent and Luke did not
simply leave it out as unimportant). ‘Videbat Christum etiam Romae
esse’ (Bengel 488); and seeing this he gave thanks and took cheer.
Beyer (157) may be right in saying, ‘Sie hatten seinen Brief richtig
verstanden.' For the language cf. Josephus, Ant. 9.55: ... πρὸς τὸ
λαβεῖν αὐτὸν εὔελπι θάρσος. For the spelling θάρσος (not θάρρος)
see BDR § 34.2, n. 7. Cf. 23.11.
For a description of the way see Acta Petri et Pauli 13—21 (L.-B.
1.184-88).

16. Verse 16a seems to repeat v. 14b. See the note on v. 14; but the
possibility should also be considered that the doublet is due to Luke’s
use of more than one source, to the need of an introduction to the
final paragraph. Conzelmann (149) thinks v. 14 to be ‘redaktionell
vorweggenommen’. In contrast with v. 14 'Ρώμην does not have the
article. At least this is so in NA26; the article is read by L Ψ 048
614 1175 2495 al; it is absent from A B 066 <a>. The party would
enter Rome by the Via Appia through the Porta Copena.
Davies (Land 281) notes that as in v. 15 there is no reference to the
military escort (why should there have been?) so here there is no
reference to the Roman authorities, to whom Paul was being
delivered at so much trouble, danger, and expense. The fact is (see
further below) that Luke is allowing the legal proceedings against
Paul to drop out of his narrative. In any case, Davies’ observation
needs some qualification. Paul was permitted (ἐπετράπη) ... But
ἐπετράπη implies someone who ἐπέτρεψε, permitted Paul to take up
the relatively free situation described in this verse, and this must have
been a Roman authority. Paul was permitted to stay καθ’ εαυτόν, on
his own, that is presumably not in a public prison but in private
accommodation (cf. 28.30), in custodia libera, not custodia militaris
(see Tajra 179-181). The concession to legal requirements was that
he should remain with the soldier who was guarding him. Luke’s
expression does not contradict the custom that a prisoner was handed
over to two soldiers; no doubt they would watch in shifts. If this is
correct it must have been at least provisionally decided that Paul was
not a threat to public order.
So far the Old Uncial text has been considered. The Western
text makes two substantial changes. Where P74vid A B Ψ 148vid 066
81 1175 1739 2464 2495 pc vg (syp) bo have ἐπετράπη τφ Παύλῳ
(the Antiochian text apparently following the Western), gig p (syh**)
sa have ὁ εκατόνταρχος παρέδωκεν τούς δεσμίους τφ
στρατοπεδάρχφ, τῷ δὲ Παύλῳ ἐπετράπη. And after εαυτόν, 614
2147 pc it (syh**) add έξω της παρεμβολής.
 63. FROM MALTA TO ROME. 28.1-16. 1233

The στρατοπέδαρχος (some MSS have –άρχης) is the praefectus

castrorum; so LS 1653; whether this is a correct interpretation of the
present passage is disputed. At a later time the official who would
most naturally receive prisoners from abroad would be the prefect of
the Praetorian Guard; thus Pliny, Epistles 10.57.2, where Trajan
writes to Pliny that a person who had been banished ‘vinctus mitti ad
praefectos praetorii mei debet’. The Prefect from AD 51 to 62 was
Afranius Burrus; Sherwin-White (108-11) however, thinks it
unlikely that Paul would be committed to him. More probable is the
view that the prisoners would be handed over to the princeps
peregrinorum; this may be supported by the association of this
princeps with the frumentarii, the officers responsible for exercising
oversight of the com fleet (to which Paul’s ship probably belonged;
see ν. 11) and receiving and distributing the cargoes of com. This
association however had probably not yet come into being, and a still
more probable view is that the prisoners were received and dealt with
by the princeps castrorum, the chief administrative officer of the
Praetorian Guard. It is however to be noted that in this passage gig
has Principi Peregrinorum. p has Prefecto.
έξω τής παρεμβολής (outside the Praetorian camp, or barracks)
interprets καθ’ εαυτού. In the variants we see editors or copyists
filling out a rather thin text on the basis of more or less correct and
complete knowledge of Roman procedures. In addition to Sherwin-
White see a detailed note by Clark (386-8) on the Western text,
princeps peregrinorum, and frumentarii; also Tajra (177-9) and
Hemer (157, 199f.).
 64. PAUL AND THE JEWS IN ROME 28.17-28

(17) After three days Paul called together the leading men among the Jews.
When they assembled he said to them, 'Brothers, although for my part I had
done nothing against the people or our ancestral customs I was handed over
from Jerusalem into the hands of the Romans as a prisoner. (18) They
examined me and wished to release me because they found in me no capital
charge. (19) But because the Jews contradicted this I was compelled to
appeal to Caesar, though not because I had anything of which to accuse my
nation. (20) For this reason therefore I asked to see you and to address you,
for it is for the hope of Israel that I wear this chain.' (21) They said to him,
‘As for us, we have neither received letters about you from Judaea, nor has
any of the brothers who have come here reported or spoken any ill of you.
(22) We should therefore like to hear from you what is in your mind, for as
for this sect it is known to us that it is everywhere spoken against.'
(23) They appointed him a day and came in large numbers to his lodging.
He gave them an exposition, testifying the kingdom of God and persuading
them about Jesus on the basis both of the Law of Moses and of the prophets.
This he did from early morning till evening, (24) and some believed the
things he said, others did not. (25) The gathering broke up without
agreement, with Paul saying one thing. ‘Well did the Holy Spirit speak to
your fathers through Isaiah the prophet, (26) saying, “Go to this people and
say: You will hear and hear and not understand, and you will look and look
and will not see. (27) The heart of this people has been hardened, and they
have heard dully, and they have shut their eyes; lest they should see with
their eyes, and hear with their ears, and understand with their heart, and turn,
and I should heal them.” (28) So let it be known to you that this God-given
salvation has been sent to the Gentiles; they will listen.’

Bibliography
C. K. Barrett, FS Jervell, 1-10.
F. Bovon, ZNW 75 (1984), 226-32.
F. F. Bruce, BJRL 46 (1963/64), 326-45; 50 (1967/68), 262-79.
J. Dupont, as in (63).
K. Haacker, as in (56).
H. J. Hauser, Strukturen der Abschlusserzählung der Apg., AnBib 86,
(1979).
P. W. van der Horst, NovT 17 (1975), 158.
J. Jervell, The Unknown Paul, 13-51.
P. Katz, ThLZ 61 (1936), 284; 83 (1958), 316.
E. Larsson, FS Jervell, 93-105.
1234
 64. PAUL AND THE JEWS IN ROME. 28.17-28. 1235

D. P. Moessner, NTS 34 (1988), 96-104.

G. W. Trompf, as in (62).
W. Wiefel, Judaica 26 (1970), 65-88.

Commentary
Apart from the final verses, 28.30, 31, considered below (pp.
1248-53), this is the end of Luke’s story. It should be considered
first in this light, as a narrative piece which fulfils the goal frequently
alluded to in the course of Acts (1.8; 9.15; 19.21; 23.11; 27.24; cf.
also Lk. 3.6; 24.47). Paul has now reached Rome, and is (notwith-
standing his appeal to the Emperor’s court, before which he must in
due course appear) in circumstances favourable to the proclamation
of the Gospel; Weiser (674) rightly draws attention to the fact that
the present paragraph is framed by 28.16 and 28.30f., which depict a
hopeful situation. Luke is primarily a narrator, and the end of his
book represents the successful achievement of a primary goal, and
thereby the victory of the word of God.
As soon, however, as the contents of this final scene in the play are
examined questions and difficulties appear. Delegations of Christians
had gone out to meet Paul as he approached the city (28.15), but in
the final paragraph there is no further hint of their presence; not only
are they not mentioned, v. 22 suggests their absence. Had there been
Christians in Rome one would have supposed that the Jews of the
city would have known something about them. It is an inadequate
explanation that Luke wished to suggest (contrary to fact) that Paul
was the founder of the church in Rome; had he intended this he must
have omitted 28.15. Why is there no further reference to the Roman
church, to which Paul had written so notable a letter? Again, what is
Luke’s authority for the events he narrates? The first person plural
appears no more after 28.15. The brothers came out from Rome to
meet us, so that we must surely have been present in Rome. Roloff
(371) goes further than this: “Die auffällige Orientierung der Apg
auf Rom hin erklärt sich am besten von der Annahme her, für die
auch sonst vieles spricht, dass Lukas Glied der römischen Gemeinde
war.’ For this there is no direct evidence, but it may provide (and
would thereby find support in) an explanation of the silence about the
local Christians (noted above). Was Paul rejected by the local
Christians, or by an influential number of them, when they had had
the opportunity of learning more about him than was known at the
time of 28.15? When Romans was written there was a majority of
Gentiles in the Roman church, but there were also Jews (Romans 6,
23). Was one party not pleased by the exposition of the Gospel they
had received on paper and now heard from Paul’s lips? Roloff (372)
goes on to cite 1 Clement 5.5, where Paul is said to have suffered διὰ
ζῆλον καί ἔριν; Clement does not say this with special reference to
1236 COMMENTARY ON ACTS

Paul’s experiences in Rome, but he was writing in Rome and his

statement may be based on memories preserved there. This may be
related to a further question, which cannot be fully discussed here
(see Introduction, p. xliii). Why does Acts stop at this point? Why
does Luke not continue with an account of Paul’s trial before the
Emperor, culminating in either his release or his martyrdom? Either
event would have made a most impressive close for the book. A
possible answer to this question is that a true account of the end of
Paul’s career might have been anything but impressive; it might have
had to describe his desertion by those who should have stood by him
(cf. 2 Tim. 4.16). For the present we must be content to discuss the
ending that we have, without speculating why we do not have what is
not there. See pp. 1248-50.
We are on firmer but still difficult ground when we investigate the
assertion made by the Jews that they have had no letter about Paul
from Judaea, and that no travelling Jew has had any evil to say about
him. That the Jews in Rome had heard nothing about Paul from
Jerusalem is ‘historisch schwer denkbar’ (Roloff 373; cf. Stählin
327, but Stählin is assuming on the doubtful basis of 24.27 a long
imprisonment of Paul in the East). The combined improbabilities to
some extent support the view that ‘die Verse [17-28] sind in toto
lukanisch’ (Lüdemann 273), but this judgement does not necessarily
rob them of all historical value. It is not unimportant that Luke
describes two meetings of Paul with the Jews. In vv. 17-22 Paul
sends for the Jews (he is in custody and cannot come to them). When
they arrive he introduces himself as one who had been obliged to
appeal to Caesar not because he wished to accuse his own people (v.
19) but simply as a measure of self-defence. He wished to make it
clear that his imprisonment was for the sake of the ‘hope of Israel’ (v.
20) . His hearers reply simply that they are ignorant of the whole
affair; they have heard nothing, at least nothing evil, of Paul himself.
They have heard of Christianity as a (Jewish) sect, and know that it
has a bad reputation. This does not predispose them to view it
favourably, but they are fair-minded enough to invite Paul to declare
his mind to them (v. 22). This leads to a second invitation and
appointment. The Jews come to Paul’s lodging, as, presumably, they
had done in v. 17. Paul now (v. 23) addresses them in terms that
recall Luke’s customary summaries of the Gospel, and with the
customary result: some believe and some do not. Paul then comments
on the situation by quoting Isa. 6.9f. and asserting, as in 13.46; 18.6,
that the good news of salvation has been sent (he does not say, will
now—from this time—be sent) to the Gentiles, who certainly will
listen. This double encounter between Paul and the Jews recalls to
Weiser (678) the ‘Doppelszene am Anfang der ersten Missionsreise
(13.14-43, 44-48)’. There seems little point however in an artificial
repetition of this double scene; it would have been easy to compress
 64. PAUL AND THE JEWS IN ROME. 28.17-28. 1237

the substance of vv. 17-28 into one event, and there is therefore
some probability that there is some distinct traditional recollection of
what took place, though the language, especially of vv. 23-28, is
undoubtedly Lucan.
What did Luke mean by this encounter between Paul and the Jews,
and especially by his use of an OT passage which plays so large a
part in the NT (cf. Mt. 13.14f,; Mk 4.12; Lk. 8.10; Jn 12.40: Rom.
11.8)? For different views that have been taken with regard to this
question see Introduction, pp. xcviif. Some think that Luke means
that the mission to the Jews is now at an end; individual Jews may
still, perhaps be converted, but it is recognized, by Luke and by the
church at large, that Israel has rejected the Gospel and that hence-
forth the church will be a Gentile church. A paradoxically opposite
view has also been maintained; the mission to the Gentiles has
become possible on the basis of the mission to the Jews. For a full
discussion of these views see Wilson (Gentiles 226-33). Neither of
them is convincing; Luke has throughout his book argued that the
Christian Gospel is based upon (a true reading of) the OT; it is the
true, fulfilled version of Judaism. He has also (and, as we have seen,
in this passage) proclaimed the triumph of the Gospel, not its defeat
by a recalcitrant people. This he still maintains, using the explicit
quotation of Isaiah 6 and the allusion contained in v. 28 (see the
notes) to show that even aspects of the completed story that seem less
than triumphant are present in fulfilment of Scripture; the whole
story is in God’s hand. Cf. Dupont, Nouvelles Études 510f.

17. For Luke’s use of ἐγένετο with the accusative and infinitive
see Introduction, p. xlvi. After only three days Paul’s first step was to
call together (the active and middle of συγκαλεῖν seem to be used
indifferently; see BDR § 316.2, n. 3) the leading men among the
Jews. From this point we hear no more of the Christians in Rome.
Luke offers no explanation of this and no conjecture (see e.g.
Lüdemann 273f.) is wholly satisfying. It may well be that the
reception of Paul by the Roman Christians was anything but whole-
hearted. 1 Clement 5.5 ascribes his martyrdom to ζῆλος καὶ ἔρις;
these may have been found within the Christian community itself.
That Luke wished to make a final clarification of the position of
church and Gospel over against Judaism may be true but it is not a
sufficient reason for omitting all further reference to the church.
πρώτος is used here as at 25.2. Μ. 1.228 refers to Ramsay [Church
52] as quoting J. A. Robinson to the effect that ὁ ὤν ‘introduces
some technical phrase, or some term which it marks out as having a
technical sense (cf. 5.17; 13.1; 28.17), and is almost equivalent to
τού ὀνομαζομένου ’. Here it could mean, ‘those who bore the title of
...’. The expression recalls the δέκα πρώτοι of Tiberias (see
Josephus, Life 69, 168, 296; War 2.639; and cf. NS 2.180) also the
1238 COMMENTARY ON ACTS

the heads of a synagogue community. In Rome, the Jews

(who have evidently returned after the expulsion of 18.2) were
gathered in a number of synagogues. Eleven are known by name (NS
3.95-8). For their άρχοντες (who might well be styled πρώτοι) see
NS 3.98-100. It may be inferred that Paul, as a prisoner, though well
treated, was not at liberty to visit the synagogues; if contact was to be
made their leaders must come to him.
The next clause is badly constructed; the genitive pronoun in a
genitive absolute (συνελθόντων—the result of συγκαλέσασθαι—
αυτών) is followed by an accusative (πρὸς αυτούς).
ἔλεγεν: ‘impf. quia expectatur responsum (ut ἠρώτα [3.3] al.)’
(Blass 288).
For Paul’s address to the Jews, ἄνδρες αδελφοί, see e.g. 2.29;
22.1. In good Attic style, the vocative stands close to the beginning
of the sentence. Paul continues (unusually in the NT) by using
οὐ[δέν] with the participle. The participial clause is concessive:
Although I had nothing ... For ἔθεσι πατρῷοις see 15.1; 21.21; 22.3;
26.3. P. W. van der Horst (NovT 17 (1975), 158) quotes Hierocles the
Stoic 52.9 (= Stobaeus Florilegium 3.39.36), τά ἔθη φυλακτέον τά
γε όντως πάτρια. It is not clear whether ἐξ Ιεροσολύμων is to be
connected with δέσμιος or with παρεδόθην. From Jerusalem prob-
ably represents the ‘authority’ under which Paul was held before the
Romans took charge, but the wording here calls attention to the
discrepancy between this brief summary of events and the more
detailed account of the preceding chapters. Paul was certainly
thought to have acted in a manner contrary to the interests of the
People and the Law, though of course he denied this. The first
expression of this charge was the mob violence that broke out in the
Temple. As a result of this Paul was handed over (παρεδόθην) to the
Romans; the Romans took him by force out of the hands of the Jews
in order to prevent his being lynched; this at least was the effect of
their action. Its initial intention may have been to prevent the
development of a dangerous riot. From this point onwards however
Paul was in what may equally be described as Roman custody and
Roman protection. The Jews would have liked to try him in their own
court and, if we may accept Luke’s narrative, there can be little doubt
that such a trial would have resulted in his death. He refused to be
handed over for Jewish trial and the case was transferred to the
Governor’s court in Caesarea. The Romans could see nothing more
serious in the matter than a Jewish theological controversy and the
bizarre assertion that a dead man, Jesus, was now alive. Eventually,
in order to remain in Roman hands and secure a fair (that is, a non-
Jewish) trial—perhaps also in order to win a passage to Rome—Paul
used his Roman citizenship in an appeal to Caesar. This summary
account is so radically abbreviated by Paul (Luke) as to suggest that
Paul’s position as a Roman prisoner was due to Jewish legal action. It
 64. PAUL AND THE JEWS IN ROME. 28.17-28. 1239

may be correct to see (with Conzelmann 149) in the expression εἰς

τὰς χεῖρας τῶν 'Ρωμαίων the influence of the story of Jesus;
Conzelmann refers to Lk. 9.22; 24.7; one could add Acts 2.23. The
passion of the servant of Christ is described in terms drawn from
Christ’s own. The conformity with Christ’s death that Paul could
claim as a theological proposition (most clearly in Phil. 3.10) Luke
has expressed in historical—perhaps we should say, quasi-histor-
ical—terms.

18. The parallel with the story of Jesus continues. Pilate found
Jesus innocent (Lk. 23.4, 15, 22); the Roman court found Paul
innocent (23.29; 25.18, 25; 26.31f.).
οἵτινες is here little if anything more than the equivalent of oἵ,
perhaps there is a hint of, Fair-minded people like them ... wished to
release me. In 614 pc syh** the relative is followed by πολλά. This is
probably used adverbially: The Romans examined (on ἀνακρίνειν
see on 24.8) me strictly. Metzger (501), however, translates the
longer text, ‘when they had examined me concerning many things
[or, after a long examination]’.
As a result of their examination the Romans wished to release
Paul, because (διά with the articular infinitive and accusative) no
charge requiring the death penalty existed in me. For αιτία see 23.28;
25.18, 27.

19. Paul does not conceal the fact that the Jews took a different
view of the loyalty to Judaism that he professed. According to the
Old Uncial text they simply denied what he said. In the Western text
(represented here by 614 2147 pc syh**) they went further: καί
ἐπικραζόντων αἶρε τὸν ἐχθρόν ἡμῶν, which, though it recalls
22.22, is not simply a verbal assimilation to that verse. It was their
attitude that compelled (ἠναγκάσθην) Paul to appeal to Caesar
(25.11). This was (he asserts) simply a matter of self-defence, a step
taken in order to secure a fair trial and security till the trial was held.
He denies the intention of bringing any accusation against his own
race. He has in fact in the preceding sentences accused them of
making an unjustified and lethal attack upon himself; it is in Caesar’s
court that he will not accuse them. Paul’s true intention is made
explicit in a further Western addition, after κατηγορεῖν
(κατηγορῆσαι in Ψ <a>): άλλ’ ἵνα λυτρώσωμαι τήν ψυχήν μου ἐκ
θανάτου (614 2147 pc gig p vgmss syh**). Here the Western text is
not so much introducing an anti-Jewish element into the text as
filling out connections that the shorter text implies but does not state,
and at the same time sharpening the narrative. Delebecque (138)
accepts the Western reading and defends it at length.
‘The Jews’ stand over against Paul, who distinguishes himself
from them; yet he does not accuse them. The distinction and the lack
1240 COMMENTARY ON ACTS

of accusation are both elements in Luke’s attitude to Judaism. See

Pesch (2.308).
At first sight it may appear that this verse contains another
example (cf. v. 17) of the use of ού with a participle; this however
may not be so. According to BDR § 430.2, n. 4 οὐχ negatives not the
participle but the main verb: I did not appeal because I had ... This
however is not quite satisfactory. Paul did in fact appeal. I appealed,
but not because I had ..., or, though I did not have ... Cf. BDR §
425.3, n. 3, where οὐχ ὡς is rightly rendered ‘nicht, als ob’.

20. διὰ ταύτην τὴν αἰτίαν, for this reason (αιτία not as in v. 18). It
is not clear to what reason Paul refers. The phrase may look forward
(For this reason, namely because (γάρ) it is for the sake of Israel’s
hope that I ...), or backward, to Paul’s explanation (v. 19) of his
appeal to Caesar. The prospective reference scarcely seems adequate,
and it is best to give the ‘reason’ a somewhat wider explanation than
the words themselves demand. It is because my appeal to Caesar puts
me in an ambiguous position, in which I am at the same time
defending myself against an unwarranted attack from the Jewish side
while I am in truth maintaining all that is true and valuable in
Judaism, that I must seize, or create, an opportunity of making clear
to you exactly what the facts are. I have been more than sufficiently
misunderstood in Jerusalem, and I hope by seeing you and speaking
with you to be understood by you and to be on good terms with
you.
Μ. 2.319 takes παρακαλεῖν to have here its simplest, literal sense,
to call to one. BDR § 409.5, n. 7 more cautiously allow the two
possibilities, ich habe euch gebeten euch sehen zu dürfen, and ich
habe euch herbeigerufen, euch zu sehen. Call to one would be a very
unusual meaning for παρακαλεῖν in the NT, and there is no reason
why the simple meaning ask should not be adopted: I have asked to
see you and address you. Bruce (1.477) takes ὑμᾶς as the object both
of παρεκάλεσα and of ἰδεῖν καὶ προσλαλῆσαι. It is not really
needed with παρεκάλεσα (as in English: I asked to see you).
Elsewhere Paul has insisted that the real point at issue between
himself and his Jewish adversaries is the resurrection: 23.6; 24.15,
and it may be that this is the meaning here of the hope of Israel. At
26.6 however (and perhaps at 24.15 also) the meaning of hope is
different, and it is probable that here the hope of Israel is the
realization of the promises God has made to his people, that is, the
promise of Messianic salvation. Paul alleges that the hope has been
and will be fulfilled in Jesus. For Paul (whether this is true also for
Luke is a difficult question) this belief is guaranteed and anticipated
by the resurrection of Jesus, but the hope is wider than the personal
resurrection of the crucified Messiah. Understood in this way the
statement represents fairly enough the issue between Christianity and
 64. PAUL AND THE JEWS IN ROME. 28.17-28. 1241

Judaism: Jesus was or was not the one in whom the promises were
fulfilled. Weiser (680) rightly comments that Luke is concerned with
the question of Jewish and Christian identity.
It was this belief that led to Paul’s arrest and to his being now in
custody. His position is vividly expressed: τήν ἅλυσιν ταύτην
περίκειμαι. This verb is used as a passive of περιτιθέναι (Blass 289:
‘Est κεῖσθαι = τιθεῖσθαι’), which takes a dative of the person
wearing and accusative of the thing worn. The dative becomes the
subject of the passive and the accusative remains. See BDR § 159.4,
n. 4. Ts this metaphorical? If not, what was the force of the Lex
Iulia?' (Begs. 4.346). If Paul was not wearing a chain the metaphor
would fall flat. But the wording is due to Luke.
ἔνεκεν stands in NA26 and no variant is given. In fact, εἵνεκεν is
used in a number of MSS, including and A (also B, according to
Ropes, Begs. 3.252). With this variation in mind Blass (289)
comments, ‘εἵνεκεν forma ionica pro ἕνεκεν (ένεκα et εἵνεκα att.)
edam Lc 4.18 A B al it 2 Cor 3.10.’

21. The Jews in Rome cannot comment on Paul’s relations with

the Jews in Jerusalem or on the version of ‘realized Judaism’ that he
presents. They are completely uninformed about Paul and have
barely heard, and that unfavourably, about Christianity (v. 22). They
have received no letters about Paul from Judaea. This ‘assumes that
there was regular correspondence and personal contact between the
Jews in Rome and the Jewish supreme authority in Jerusalem, the
Sanhedrin’ (J. Jeremias, Jerusalem 64). The correspondence had
contained no reference to Paul. That is surprising, if the Jewish
authorities had in fact determined (as the earlier chapters in Acts
suggest) to destroy Paul; even if they thought it unlikely (as indeed it
was) that local Jewish opinion could affect the proceedings in the
Emperor’s court one would expect them to solicit any help that might
possibly be available (the Emperor might conceivably wish to sound
Jewish opinion) and to put the Roman Jewish communities on their
guard against a disruptive and perverting presence.
The latter part of the verse is ambiguous, and its meaning depends
on the stress laid upon πονηρόν. It may mean that no report at all
concerning Paul and therefore no evil—or good—report had reached
Rome, or that those who had spoken of him had had no evil to say.
On the whole it seems that the Roman Jews are professing complete
ignorance of Paul and are minimizing their knowledge, which they
cannot deny altogether, of the Christian movement.
ἀπήγγειλεν ἢ ἐλάλησεν ‘reported (officially) or spoken (unoffi-
cially)’—Bruce (1.478).
This verse suggests an almost complete cleavage between Jews
and Christians in Rome, and between Jews in Rome and Jews in
Jerusalem. Neither of these seems probable. The effect is to represent
1242 COMMENTARY ON ACTS

Paul as not only a pioneer missionary but as the spokesman of

Christianity to the Jews. This at least seems to be Luke’s intention.
The suggestion that the Jerusalem Jews had decided to withdraw
from the prosecution since there was no hope of obtaining a verdict
in the Emperor’s court and the penalties for accusers who failed to
win their case were heavy is dealt with by Sherwin-White (112-19)
and Hemer (157f.). It is not strong. ‘The possibility that Paul’s
accusers never intended to pursue their prosecution requires much
thought. Roman justice at this period was also severe on default, and
A. N. Sherwin-White (112—19) shows that the thrust of contempo-
rary legislation was rather to enforce prosecution than to favour
release of the unindicted’ (Hemer 157).
22. Having heard nothing about Paul himself and only unfavoura-
ble reports of the faith he proclaimed, the Jews desired to hear what
he had to say, to know his views, what was in his mind—α φρονείς.
It was the same desire that was expressed by the Areopagites in
17.19.
Christianity is described as a αἵρεσις; for the word see on 5.17;
24.5, 14. It suggests a party (within Judaism). If pressed, it would
mean that the Roman Jews still thought of Christianity as one sect,
with its own interpretation of Jewish principles, within Judaism; this
indeed is the way in which it is on the whole presented in Acts,
though it is presented as a form of Judaism with open access to
Gentiles. All that can be said is that it is everywhere (that is, among
all Jews) spoken against, rejected and resisted.
μέν (after περί) is omitted by 104 pc, possibly because there is no
answering δέ (for μέν solitarium in Acts see 1.1; 3.21; 27.21, and cf.
BDR § 447.2c, n. 15). It may be that a δέ clause is to be understood
here: We know that this form of religion is everywhere disliked,
nevertheless, or but for that very reason, we should like to hear what
you have to say. This is more or less the suggestion of Blass (289):
‘μὲν sine δὲ ut 3.21 al.; supplendum fere ut oppositum idem quod
modo praecessit (ἀξιοῦμεν κτέ.).’
The Jews’ ignorance of Paul and their knowledge of Christianity
only by repute fit badly with the view that the expulsion of the Jews
under Claudius (see on 18.2) had been caused by Jewish-Christian
riots (impulsore Chresto). This may be explained by the fact that
Luke is here writing up material available to him in such a way as to
suggest that Paul was virtually the founder of the church in Rome
(see above, p. 1235); more probably however Luke wishes to
represent Paul as starting a clean sheet in relation with the Roman
Jews.
23. The middle ταξάμενος is correctly used (cf. e.g. 2 Macc. 3.14;
14.21) for making an arrangement (cf. 15.2, where the active is
used); here for appointing a day for meeting. On the appointed day
 64. PAUL AND THE JEWS IN ROME. 28.17-28. 1243

the Jews came (ἦλθον; but Ψ <a> have ἦκον, and the less common
word could be right) to Paul; he, though living καθ’ εαυτόν (28.16)
but with a guard, was not free to go where he chose. The usual
meaning of ξενία is hospitality, but here ‘perh. lodging' (LS 1188).
Begs. 4.346 prefers hospitality, but BA 1109 prefers the concrete
sense of the word; so also Haenchen (690). At Philemon 22, the only
other occurrence in the NT, it is similarly ambiguous. Cf. the use of
μίσθωμα in 28.30. It is not possible to reach a clear decision on this
question, and not necessary: if the Jews came to Paul’s lodging he
was their host (the word ξενία does not necessarily imply elaborate
entertainment) and if he entertained them at all it must have been at
his lodging.
πλείονες is probably elative, they came in considerable numbers
(rather than comparative, more came than on the previous occasion,
ν. 17). For the form of the word see BDR § 47.2, n. 2. πλείονες is
used here and at 27.12, πλείονας at 27.20, but πλείους at 13.31;
19.32; 21.10; 2343,21; 24.11; 25.6,14. This might suggest a special
source for chs. 27, 28.
Luke continues the sentence, as often in Acts, with a relative (οἷς).
For the use of ἐκτίθεσθαι cf. 11.4; 18.26. It occurs nowhere else in
the NT (at 7.21 in a different sense), but the meaning to set forth is
clear. P74vid A*vid pc have the aorist ἐξέθετο. The imperfect is clearly
suitable for a process lasting all day, but the aorist can be understood
as constative: on this occasion, viewed as a whole, Paul set forth
the Gospel. The setting forth is amplified by the participle
διαμαρτυρόμενος (διαμαρτυρόμενος, ; παρατιθέμενος, A).
This fulfils the promise of 23.11. Exactly what testifying the
kingdom of God means is not clear, but elsewhere (e.g. 28.31) Luke
appears to use the term kingdom of God as a summary of the
Christian message (so also Schmithals 239) and it should be so taken
here, with the next clause bringing out the Christological content and
base of the message. Paul sets out to persuade his hearers about
Jesus, namely that he fulfils the meaning and especially the promises
of the Law and the prophets. Stählin (327), noting the juxtaposition
of kingdom and Jesus, aptly refers to Origen’s word, αὐτοβασίλεια.
The Law is here understood as a repository of messianic predictions
and promises rather than as a body of commandments, though Luke
still wishes to present Paul as an observant Jew, in no way an enemy
of the Law. Both the Law and the prophets are used: τε stands after
the preposition. This is classical; see BDR § 444.5, n. 6. Another
classical feature is the omission of the article before a word denoting
time (ἕως ἑσπέρας); Μ. 3.179; BDR § 255.3, n. 5. Paul’s interview
with the Jewish representatives lasted all day; if in the end any were
unbelieving (v. 24) the fault was not his. Of πείθων, Knowling says
that the tense is not simply de conatu; it refers to the persuasive
power of Paul’s words but does not claim that they resulted in
1244 COMMENTARY ON ACTS

conviction. But surely the present participle matches ἐξετίθετο, and

the imperfect is durative; the process went on all day, ἀπὸ πρωῖ ἕως
εσπέρας.
Μωϋσέως is omitted by P74, either by accident or because it was
felt to be unnecessary to specify what law was intended.
24. The outcome was mixed; it is expressed by the use of the
classical oἱ μὲν ... oἱ δέ ... construction, cf. 17.32, where a similar
consequence of Paul’s preaching is described, πείθεσθαι is the
opposite of ἀπιστεῖν, that is, it means to believe. This is against
Bruce’s (2.506) rendering of the imperfect, ‘were on the way to
being persuaded’.
It is important, in view of the following verses, to note that Luke is
at pains not to represent all Jews as unbelieving. He refers first to
those who were persuaded by Paul’s message.
25. The Jews were divided, were in disagreement, ἀσύμφωνοι.
The word occurs nowhere else in the NT, but elsewhere is not
uncommon in both literal (unharmonious) and metaphorical senses;
cf. Plato, Gorgias 482c, where ἀσύμφωνοι είναι includes εναντία
λέγειν.
There was no agreement and the gathering broke up, when Paul
had fired one Parthian shot (the aorist participle εἰπόντος refers to
something said before ἀπελύοντο), quoting a passage (Isa. 6.9f.)
used in a number of notable places in the NT (see Mt. 13.14f. = Mk
4.12 = Lk. 8.10; Jn 12.40; Rom. 11.8; cf. Justin, Trypho 12.2; 33.1).
It was addressed to your fathers (υμών, P74 A Β Ψ 049 33 81 323
945 1175 1739 2464 al p s syp Cyril of Jerusalem; <a> gig vg have
ἡμῶν; syh omits); this has the effect of dissociating Paul from his
hearers. ‘Melius ὑμῶν quam ἡμῶν; neque enim P. sibi partem huius
quasi hereditatis vindicat’ (Blass 290).
Belief in the inspiration of the OT is plainly expressed. The words
were given through Isaiah but the actual speaker was the Holy Spirit.
Fitzmyer, (Essays 301), cites CD 4.13, 14, As God spoke through
(literally, by the hand of) Isaiah the prophet, the
son of Amoz, saying ...
26. This verse gives the words of Isa. 6.9 substantially as they
appear in the LXX. In the opening clause there is variation in order,
and where Luke has πρὸς τὸν λαὸν τοῦτον, the LXX has (after
εἰπόν) τῷ λαῷ τούτῳ. On the accentuation of ειπον see below. After
this the agreement is complete (though there are small variations in
the MSS of the LXX).
ειπον is undoubtedly imperative—no other understanding of it
(first person singular or third person plural aorist indicative) makes
sense, but it may be written εἰπόν (said to be Syracusan Greek) or
εἶπον (apparently Attic). The former is used in NA26 and in most
 64. PAUL AND THE JEWS IN ROME. 28.17-28. 1245

printed editions. See Μ. 2.58; BDR § 81.1, n. 1; also BA 456; P.

Katz, ThLZ 61 (1936), 284; 83 (1958), 316; earlier, WM 23, 58, 103;
WS § 6.7d.
πορεύθητι (aorist) is an emphatic, once-for-all instruction: Go, go
now; there is no need to deliberate.
The cognate dative ἀκοῇ represents the Hebrew infinitive abso-
lute, which has the effect of intensifying the meaning of the verb; the
participle βλέποντες serves the same purpose. See Zerwick (§ 61);
BDR § 198.6, n. 9. In the quotation (and in agreement with the LXX)
the future of ἀκούειν has the active form ακούσετε; in v. 28, where
Luke writes without the constraint of quotation, he uses the classical
middle deponent future, ἀκούσονται. See Zerwick (§ 226); BDR §
77.3, n. 2. For details of the quotation see G. J. Steyn, Septuagint
Quotations in the Context of the Petrine and Pauline Speeches of the
Acta Apostolorum (Kampen 1995), 219-29.
The prophet is sent to his people with the message that there is no
possibility of their understanding what they hear or seeing what they
look at. The built-in failure of the message is the content of it. The
unbelief of Israel is not an unhappy accident but part of God’s
intention. The theme is continued in the next verse; see on that verse
and on v. 28 for the use Luke makes of his quotation. Calvin (2.312)
draws the lines of interpretation too sharply for Lucan thought: ‘This
cause, mentioned here [that the reprobate are to blame for their
rejection], does not prevent the secret election of God from distin-
guishing between men, so that those, who have been ordained to life,
believe, but the others continue to be senseless.’

27. The quotation of Isaiah 6 continues. In NA26 there are again

small differences. In τοῖς ὦσιν αυτών Luke drops the unnecessary
αυτών. It is worth noting that for ἐπαχύνθη, (gig) have
ἐβαρύνθη (without LXX parallel); for ἐπιστρέψωσιν, A Ε Ψ 048 81
pc vgmss have ἐπιστρέψουσιν (in agreement with LXX ), and for
ἰάσομαι, E 33 81 2464 pm gig vg have ἰάσωμαι (in agreement with
LXX V). There would no doubt be a tendency to assimilate the text
of Acts to whatever text of Isaiah was known and approved, but the
opposite tendency—to assimilate texts of the LXX written and used
by Christians to the text of Acts—would also take effect. The word
καμμύειν is rejected by Phrynichus (319): δέον ὡς οἱ ἄριστοι τών
αρχαίων καταμύειν.
The verse is probably intended by Luke (and was probably
intended by Isaiah) as a description of people who had made up their
minds not to understand, not to hear, not to see. The voices of the
verbs are probably not significant, unless ἐπαχύνθη is a passive of
divine action: God has hardened the hearts of lus people. But there is
nothing to indicate that this is what Luke intends, βαρέως ἢκουσαν
simply states a fact; ἐκάμμυσαν means that the people closed their
1246 COMMENTARY ON ACTS

eyes and were therefore themselves to blame for the fact that they did
not see. Like v. 26, this verse states that they did not understand,
hear, or see; and that this fact did not stand outside God’s purpose.
The immediate historical consequence of this fact and of this
disclosure of God’s purpose was that the Gospel was taken to the
Gentiles; Luke is not dealing with the more remote consequence
of the ultimate destiny of Israel. He does not see into the situation
as acutely as Paul does in Rom. 9—11, and he does not recognize
(at least, he does not state) the underlying truth that it is only out
of disobedience that men can in faith discover mercy (Rom.
11.30-32).

28. This verse makes explicit the intention that has already been
given implicitly in the quotation in vv. 26,27. As Bornkamm (4.157)
points out, there is here a new, Lucan rather than Pauline, basis for
the Gentile mission. If the Jews vacate their place, the Gentiles will
take it. What the Jews refuse to hear, the Gentiles will listen to (cf.
Schmithals 240). For Luke this is not so much a theological
proposition as a statement of fact; he could not fail to observe that the
Greeks were pressing into the church, the Jews were not. Jewish
Christians were becoming a minority, even if, in Jervell’s term, a
‘mighty minority’. Many take the view that Luke (cf. 13.46; 18.6)
considers the mission to the Jewish people as a whole to be at an end;
individual Jews may still be converted, but the main direction of the
church’s work is now towards the Gentiles (see e.g. Roloff 375;
Weiser 683). This seems to be too simple an analysis of Acts; this has
been shown especially by J. Jervell (especially The Unknown Paul
(1984), 13-51; cf. FS Jervell 1-10; 93-105 (E. Larsson)). It is
indeed probable that Luke could see which way things were moving
at the time at which he wrote, but the chief Lucan motif which
emerges here is one which runs through his work as a whole. Nothing
can or will prevent the spread of the Gospel. Preachers may be
persecuted, imprisoned, even killed, but the word of God is not
bound. The most favoured nation, who of all races should have
welcomed the fulfilment of their hope most gladly, may reject God’s
offer of salvation; others will take it up. You refuse to listen, but they
will hear. Luke has his own kind of triumphalism, but it is the proper
triumphalism of the word.
γνωστὸν εἶναι (γίνεσθαι) is characteristic of Acts: 1.19; 4.10;
9.42; 13.38; 19.17; 28.22. For the possibility of Semitism see Wilcox
90f., 161f. τοῖς ἔθνεσιν is emphatic in its position, but does not
imply that salvation is no longer available to Jews, σωτήριον (neuter
of the adjective σωτήριος) is preferred to the noun σωτηρία here
only in Acts. See BDR § 263.2, n. 5, and cf. Lk. 2.30; 3.6, remarked
on by Dupont (Nouvelles Études 509); but the use of σωτήριον may
be due to recollection of Ps. 67(66).3, τού γνῶναι ἐν τῇ γῇ τήν ὁδόν
 64. PAUL AND THE JEWS IN ROME. 28.17-28. 1247

σου, ἐν πᾶσιν ἔθνεσιν τό σωτήριόν σου. Stählin (328) points out

with reference to this, the end of Paul’s mission as recorded in Acts,
the way it begins: υιοί γένους ’Αβραάμ ... ἡμῖν ό λόγος τής
σωτηρίας ταύτης ἐξαπεστάλη (13.26).
αυτοί καί: For καί used to strengthen the pronoun see BDR §
442.8b, n. 24. Delebecque (140) makes essentially the same point
when he says that καί strengthens the verb: ‘Eux, oui, ils écouteront.’
Luke has just pointed out in his OT quotation the inability or
unwillingness of the Jews to hear the message directed to them; they,
on the other hand, the Gentiles, will hear, ἀκούσονται is the correct
middle deponent form of the future; contrast v. 26.
 χνπ
CONCLUSION
28.(29)—31

65. CONCLUSION 28.(29), 30, 31

(29)1 When he had said this the Jews went away, holding much debate
among themselves. (30) He stayed a full two years2 in his own hired
house, and received all who came to him, (31) preaching the kingdom of God
and teaching the truths concerning the Lord Jesus Christ with all boldness
and without hindrance.

Bibliography
E. Bammel, in R. P. Bauckham, The Book of Acts 4.358-64.
C. K. Barrett, in FS Hengel 3.545-55.
F. F. Bruce, as in (64).
O. Cullmann, RScR 60 (1972), 55-68.
P. Davies, ExpT94 (1983), 334, 335.
G. Delling, NovT 15 (1973), 193-204.
J. Dupont, as in (60), and as in (64).
E. Hansack, BZ 19 (1975), 249-53; 21 (1977), 118-21.
H. J. Hauser, as in (64).
D. L. Mealand, NTS 36 (1990), 583-97.
F. Saum, BZ 20 (1976), 226-9.
G. W. Trompf, as in (62).

Commentary
The contents of these concluding verses are clear enough: the form of
custody in which Paul is kept permits him to receive all who wish to
visit him, and in this setting he continues to preach with freedom and
without restraint. The question that the verses pose is why Luke stops
where he does. What happened at the end of the ‘two years’ (v. 30)?
1RSV, NEB, NJB omit this verse.
2RSV, NEB, at his own expense.
1248
 65. CONCLUSION. 28.(29), 30, 31. 1249

The question has been asked above (p. 1236) but cannot be avoided
here. Since we have nothing but conjecture on which to base an
answer it is not surprising that many different answers have been
given. An admirably clear list is given by Wilson (Gentiles 233-6); it
may be summarized as follows. 1. Luke was writing at the time to
which the verses refer; he wrote no more because nothing more had
happened. This view has of course implications for the date of Acts.
2. Luke planned to take the story further in a third volume which
either was never written or has been lost. 3. Luke did not record
Paul’s martyrdom because he wished to avoid parallels both with the
acts of the pagan martyrs and with the story of Jesus. 4. Luke did not
record Paul’s martyrdom because it would not have interested him or
his readers; they were concerned with theology, not with biography.
5. It was sufficiently indicated by v. 30 that, after two years had
elapsed, Paul would automatically be released. On this suggestion
see the note on v. 30.6. Luke did not recount Paul’s death because he
did not wish to encourage a piety of martyrdom. Wilson has little
difficulty in pointing out objections to all these attempted solutions.
He himself makes two points, (a) Luke did not need to recount Paul’s
trial and its outcome because his readers already knew the facts. This
seems to presuppose (though Wilson does not say so) that Acts was
written in Rome. (b) ‘Acts 28 summarises and rounds off the rest of
his narrative’ (236). The themes of Jewish obduracy (vv. 26-28) and
of the fulfilment of the commission of 1.8 are both brought out.
These are important points but even these are open to objection. Was
Acts written simply for the Roman church? It cannot be assumed
even that it was written in Rome; and if it was Luke would surely
hope for a wider readership; he had contacts, personal or on paper, all
over the Christian world. Roman Christians may have known what
happened, but that would not mean that they had no desire to see the
record on paper, and Christians throughout the Empire would be
deeply interested in the contact between the church, represented by
Paul, and the Emperor. And though Wilson is careful to cite vv.
26-28 as the basis of his reference to the ‘obduracy of the Jews’ this
theme does not occur in the final verses, where Paul welcomes all
who approach him.
If room remains for conjecture—and there is indeed room for little
else—it may be suggested that the end of the story was omitted
because it was not edifying. There are two possibilities which may to
some extent be combined. One is that after two years Paul’s custodia
libera was changed into more severe imprisonment; in other words
he was shut up in a dungeon and left to rot—perhaps to be forgotten;
was it part of Luke’s purpose to revive the memory of him? The
Roman church may have attempted much or little on his behalf; in
the end they gave up. The second form of this hypothesis is
suggested by 2 Tim. 4.16. The provenance of the apparently
1250 COMMENTARY ON ACTS

autobiographical verses in the Pastorals cannot be discussed here,

and it does not matter whether this verse was written by Paul and
incorporated in the pseudonymous epistle or came from some other
source. Its content is too disreputable for it to be described as pure
fiction; there existed a tradition of a desertion of Paul by those who
should have stood by him. When this took place we do not know. It
could have been at his trial in Rome, perhaps (see A. T. Hanson, The
Pastoral Epistles, NCB, 1982; p. 160) at the prima actio; and it could
have been repeated at the later proceedings. One possible explanation
(see pp. 1219, 1235) of the curious fact that after 28.15 there is no
reference to the Roman church is that the church was divided and
lukewarm in its attitude to Paul. If anything like this is true Luke may
well have thought that his brief concluding note would do better than
a description of what happened when the two years of sustained
evangelism ended in discreditable failure on the part of Roman
Christians.
The two concluding verses were written by Luke himself. Verse
30 takes up v. 23: ἐν ἰδίῳ μισθώματι takes up εἰς τήν ξενίαν,
εἰσπορευομένους πρὸς αυτόν takes up ἧλθον πρὸς αυτόν,
κηρύσσων and διδάσκων take up ἐξετίθετο and πείθων, τήν
βασιλείαν τού θεού is repeated, and τά περί τού κυρίου Ιησού
Χριστού takes up περί τού ’Ιησοῦ. The circumstances are the same
and the message is the same; the only difference is that whereas in
vv. 23f. the audience is wholly Jewish v. 30 refers to πάντας, which
will certainly include Gentiles but cannot exclude Jews. This is
important, because it shows how 28.26-28 are to be understood.
These verses mean not that there is no longer to be a mission to the
Jews, but that Jewish unbelief is not due to the preacher’s failure; it is
foretold in Jewish Scripture itself. ‘What Luke was defending he has
successfully concluded: God’s fidelity to his people and to his own
word’ (Johnson 476). Dupont (NTS 6 (1960), 132-55; Kremer,
Actes* 359-404) agrees with Haenchen that each mission, to Jews
and to Gentiles, influences the other. ‘L’entrée des Gentiles dans
l’histoire du salut devient ainsi un des signes permettant de reconnaî-
tre que Jésus est bien le Messie promis à Israël’ (404).

29. This verse is added by <a> it vgcl syh**, as follows: καί ταῦτα
αὐτοῦ εἰπόντος ἀπῆλθον οι ’Ιουδαῖοι πολλήν ἔχοντες ἐν ἔαυτοις
συζήτησιν (ζήτησιν, 104 pc). It is clearly of Western origin but has
been taken up in the Antiochian text. It was probably felt that the
story begun in v. 23 needed some narrative conclusion in addition to
Paul’s pronouncement in 28.28. ‘The addition was probably made
because of the abrupt transition from ver. 28 to ver. 30’ (Metzger
502). It indicates what Luke probably intended to convey—not the
total and final rejection of Judaism but a people divided in their
response to the Gospel.
 65. CONCLUSION. 28.(29), 30, 31. 1251

The verse is not contained in P74 A Β Ε Ψ 048 33 81 1175 1739

2464 pc s vgst syp co.

30. This verse and the next wind up Luke’s story. For the
questions raised by the fact that Acts ends where and as it does see
pp. 1248-50. The verses in themselves are clear and satisfactory; it is
not what is said but all that is not said that gives rise to problems.
The aorist ἐνέμεινεν is used to describe a state of affairs that
extended over two years; these are viewed as a unit (Moule, IB 13;
Μ. 3.72). For διετία (cf. biennium) cf. 24.27; the evidence suggests
that the word was coming into use in late Greek (see MM 160f. for
inscriptions and papyri; Philo, In Flaccum 128; Josephus, Ant. 1.302
has ἐπταετία). ὅλην implies that a full period of 24 months is
intended; otherwise, using the inclusive method of reckoning com-
mon in antiquity, ‘two years’ (that is, parts of two calendar years)
might mean very considerably less. BDR § 332.1 take the aorist
(ἐνέμεινεν, above) to imply that at the end of the two-year period the
state of things described ended—‘dann aber hörte dieser Zustand
auf. This is perhaps a probable rather than a necessary inference
from the tense.
In a long note in Begs. 5.319-38, Cadbury (quoting Lake at
length) argues that the διετία referred to in this verse was significant
in view of the Roman legal provision that if the accusers of a prisoner
failed to appear within two years the prosecution would lapse and the
prisoner be released. A first-century reader would then take the verse
to mean that Paul had been released and that even during the time he
was kept in custody he was able to preach freely. This suggestion
offers what seems in many respects a satisfactory explanation of the
end of Acts. It is however radically criticized by Sherwin-White
(112-19). The most serious point of criticism is Cadbury’s depend-
ence on a Latin papyrus (BGU 628 recto), which Cadbury, following
L. Mitteis, Grundzüge und Chrestomathie der Papyruskunde (1912),
2.1.281, dated in the time of Nero. It seems now to be agreed that this
document belongs to a much later period. Sherwin-White writes,
‘The rule that Cadbury requires does not make its appearance until
AD 529. In texts dated to AD 385 and 409 Theodosius enacts that if
an accuser fails to proceed with his case within a fixed period, finally
defined as two years, he shall be punished. But that the accused
should be acquitted is only laid down in a rule of Justinian dated AD
529’ (115, n. 4). In the end however Sherwin-White comes to a
conclusion not unlike Cadbury’s. ‘A more probable technical solu-
tion than that of Cadbury lies, as so often, in the nature of imperium.
Claudius in his edict about absent prosecutors did not establish a law.
For that he would have used a decree of the Senate, as was his
custom. He simply stated what he himself was going to do ...
Nothing prevented the successor of Claudius from taking a similar
1252 COMMENTARY ON ACTS

line [in dismissing prisoners and the like] if he chose... Perhaps Paul
benefited from the clemency of Nero [encouraged by Seneca], and
secured a merely casual release’ (118f.). Sherwin-White adds, ‘But
there is no necessity to construe Acts to mean that he was released at
all’ (119). The reference however to a specific period does suggest
termination, whether in release or in death; see above on the aorist
ἐνέμεινεν. See also the excellent critical summary in Wilson (Gen-
tiles 233-8), and more recently Hemer (383-7, 398); also Dupont
(Études 544f.), and (still to be studied) Begs. 5.319-38 (above).
Haenchen’s argument (692) that if after two years there had been a
favourable verdict its apologetic value would have been too great to
be omitted is a strong one. Schmithals (241f.) believes that Paul
came to Rome as a free man and survived till Nero’s pogrom. See
also Schille (479) and Weiser (677, 680).
Paul spent two years ἐν ἰδίω μισθώματι. For μίσθωμα LS 1137
give ‘that which is let for hire, hired house', but for this meaning
they give no reference in addition to Acts 28.30. BA 1060 agree with
reference to this meaning that it is ‘anderweitig nicht belegt’. The
verb μισθοῦν undoubtedly has to do with letting out for hire, or (in
the middle) having let to one, hiring, renting. In the passive it means
to be hired for pay, or (of a house) to be let on contract (so LS, loc.
cit.). For the latter meaning see e.g. Demosthenes 28.1 (836), ὁ
πατὴρ οὐκ ἐβούλετο μισθωθῆναι τόν οίκον. The suffix—ωμα is
normally passive in sense, so that there seems no good reason why
μίσθωμα should not mean a hired house (though Hanson (256)
declares ‘in his own hired lodging’ an ‘impossible translation’). This
would correspond well enough with καθ’ ἐαυτόν in 28.16. Schneider
(2.408) translates, in eigener Mietwohnung. Alternatively, and also
in agreement with 28.16, at his own expense (Hemer 158)—and thus
not dependent ‘on the Roman (or any other) Church’ (Tajra 192).
One wonders, however, who paid the rent.
πάντας τοὺς εἰσπορευομένους (εις- would be consistent with a
reference (in μίσθωμα) to a dwelling of some kind) would certainly
not exclude either Jews or Gentiles. πάντας is important to Luke,
and his meaning (though not his text) is rightly given by the Western
insertion at the end of the verse of ’Ιουδαίους τε καί Έλληνας (614
2147 pc (gig p) vgmss syh**). Before this Clark (171, 388) adds (from
gig h) καί διελέγετο πρός.

31. As elsewhere (e.g. 28.23) the kingdom of God serves as a

general summary of the Christian message as preached by the
apostles and others; the phrase that follows, τά περί τού κυρίου
'Ιησού Χριστού (the last word is omitted by 326 614 2147 2495
pc syh, probably through assimilation to 28.23), brings out its
Christological content. The verse emphasizes ‘... Jesus Christus als
den einen Inhalt der paulinischen Verkündigung’ (Stählin 329). ‘The
 65. CONCLUSION. 28.(29), 30, 31. 1253

Kingdom of God is founded on, and consists in, the knowledge of the
redemption procured by Christ’ (Calvin 2.314). κηρύσσων and
διδάσκων are here synonyms; Luke does not seem to be distinguish-
ing between two different kinds of communication.
Instead of τοῦ κυρίου, P74 has τής βασιλείας, putting side by side
the two kingdoms, of God and of Christ. This reading is probably due
to careless repetition; σωτηρίας in place of παρρησίας is probably
another careless error in P74. p omits τά περὶ τού κυρίου ’Ιησοῦ
Χριστού and, joined by vgmss and syh, adds after ἀκωλύτως, dicens
quia hic est Christus Jesus filius dei per quem incipiet totus mundus
iudicari. ‘The artistic literary cadence of the concluding phrase of the
book of Acts and the powerful note of triumph expressed by
ἀκωλύτως are greatly weakened by the pious Western addition after
ἀκωλύτως ...’ (Metzger 503).
For παρρησία see on 4.13. Here it has its primary sense of
freedom of speech, though it is not inconsistent with the exercise of
boldness in other ways also, μετὰ (πάσης) π. is an Acts phrase: 4
times, nowhere else in the NT.
ἀκωλύτως, unhindered, that is, with no one venturing or able to
hinder or prevent. The word is ‘of constant occurrence in legal
documents’ (MM 20, with many examples; ND 3.17 adds POxy
46.3269.13). For an earlier use see Plato, Cratylus 415d. ‘Rome is
depicted as showing tolerance towards Paul’s ministry and towards
the Christian message’ (Tajra 193). The word brings out several
points that Luke here and elsewhere wishes to make. Nothing, not
even imprisonment, was able to put a stop to the spread of the
Gospel, and nothing could deter Paul from doing the work of an
evangelist, with the result that even in Rome the proclamation of the
word was established. ‘Paul himself has every right to glory that the
Word of God was not bound with his chains (2 Tim 2.9)’ (Calvin
2.315). At the end of the verse the word ἀμήν is added by Ψ 36 453
614 1175 2495 al vgww syh, indicating probably that the copyists took
the book to be a liturgical text, read in church services. This is
relatively late attestation; contrast Lk. 24.53, where ἀμήν is added by
A B C3 Θ Ψ 063 f1 f13 lat syp h bomss. The difference probably
bears witness to a time when Lk. was and Acts was not read in
church because it was not canonical Scripture.
Bengel (489) ends his commentary with words that would prob-
ably have had Luke’s approval, ‘παρρησίας, fiducia) intrinsecus.—
ἀκωλύτως, sine impedimento) extrinsecus, post superata tot impedi-
menta. Victoria Verbi Dei. Paulus Romae, apex evangelii, Actorum
finis ... Hierosolymis coepit: Romae desinit ... Habes, Ecclesia,
formam tuam. tuum est, servare earn, et depositum custodire.’
 ACTS 15-28: CONTENTS

It was pointed out in Volume I, p. 57, that it would be ‘idle to attempt

to draw up a neat, formal, balanced analysis of the contents of Acts
1-14’. The second part of the book is different. The great theme of
the first part, the spread of the Gospel into the Gentile world, is
continued, but it is now bound up so closely with the work of Paul
that it gains a measure of unity and continuity. Its sources are
discussed above (pp. xxiv-xxxi). The story, as Luke has composed
it, begins with the Council at which the dogmatic and practical
conditions of the Gentile mission are established. Accounts follow of
straightforward missionary work, which includes both evangelism,
by which new Christian societies are established, and pastoral work,
by which they are developed and sustained. Relations between Paul
and the Jewish authorities explode in a riot in the Temple, and
thereafter the fortunes of Paul at the hands of Jews and Romans are
described until, after an appeal to Caesar, he finds himself in Rome.
This is the climax of the book, a representative fulfilment of the
programme and promise of 1.8.
X. The Council in Jerusalem: the basis of paul’s gentile
MISSION IS QUESTIONED IN ANTIOCH, ESTABLISHED IN JERUSA-
LEM, AND REPORTED TO ANTIOCH.
(38) Dispute in Antioch (15.1-5)
(39) Council in Jerusalem (15.6-29)
(40) Paul and Barnabas return to Antioch (15.30-35)
XI. Paul’s mission breaks new ground: after covering
SOME OF THE GROUND OF THE FIRST MISSION PAUL AND HIS
COMPANIONS CROSS OVER INTO MACEDONIA, REACH ATHENS,
WHERE THE THEOLOGICAL APPROACH TO GENTILES IS SET OUT,
AND MOVE ON TO CORINTH.
(41) Territory of the First Journey revisited (15.36-16.5)
(42) Guided by the Spirit to Troas (16.6-10)
(43) Paul and Silas at Philippi (16.11-40)
(44) From Philippi to Athens (17.1-15)
(45) Paul at Athens (17.16-34)
(46) Paul at Corinth, with return to Palestine (18.1-23)
XII. The mission based on Ephesus: the position of disciples
of John the Baptist is regularized, Paul evangelizes
Ephesus and its region, is overtaken by a riot, returns
cxix
cxx COMMENTARY ON ACTS

TO OTHER PARTS OF GREECE, AND GIVES A PASTORAL ADDRESS

at Miletus.
(47) Apollos and the Twelve Disciples (18.24-19.7)
(48) Paul’s Successful Ministry at Ephesus (19.8-20)
(49) Riot at Ephesus (19.21-40)
(50) Back to Palestine, through Macedonia, Greece, and
Troas (20.1-16)
(51) Paul’s Speech at Miletus (20.17-38)

XIII. Paul returns to Jerusalem: Paul meets James, and a

PLAN FOR HARMONY BETWEEN JEWISH AND GENTILE CHRIS-
TIANS MISFIRES.
(52) Journey to Jerusalem (21.1-14)
(53) Paul and the Church of Jerusalem (21.15-26)
(54) Riot at Jerusalem (21.27-40)
(55) Paul’s Temple Speech and the Sequel (22.1-29)

XIV. Paul and the Jews: Paul appears before the Council
AND IS SAFELY MOVED TO CAESAREA.
(56) Paul before the Council (22.30-23.11)
(57) The Plot: Paul removed to Caesarea (23.12-35)

XV. Paul and the Romans: neither Felix nor Festus nor
Agrippa can find any fault in Paul and he appeals to
Caesar.
(58) Paul and Felix (24.1-27)
(59) Paul appeals to Caesar (25.1-12)
(60) Festus and Agrippa (25.13-22)
(61) Festus, Agrippa, and Paul (25.23-26.32)

XVI. Paul reaches Rome after a long and dangerous voyage

AND HAS MEETINGS WITH THE LOCAL JEWS.
(62) The Sea Voyage (27.1-44)
(63) From Malta to Rome (28.1-16)
(64) Paul and the Jews in Rome (28.17-28)

XVH. Conclusion: Paul spends two years in Rome.

(65) Conclusion (28.(29)-31)
 INDEX

This Index does not include references to the following: any commentaries; Blass,
Debrunner, and Rehkopf, Grammatik des neutestamentlichen Griechisch; Jackson and
Lake (eds), The Beginnings of Christianity; Liddell and Scott, A Greek-English
Lexicon; Moulton, Howard, and Turner, A Grammar of New Testament Greek;
Wilcox, The Semitisms of Acts.
The two volumes are numbered continuously; volume I ends on page 694, the
commentary on the text in volume II begins on page 695. Roman numerals refer to the
Introduction to volume II.

Abbott, E. 392 Aeneas Gazaeus 810 Anacreon 125

Abrahams, I. 340, 605 Aeneas Tacticus 635 Ananias (High Priest) 219,
Achilles Tatius 292, 1205 Aeschines 97, 582, 851, 1058
Achtemeier, P. J. lxi 1157, 1194 Ananus 219
Acta Andreae et Matthiae Aeschylus 65, 70, 197, Annas 224, 225, 282, 286
845, 1059 235, 296, 326, 354, 402, Anthologia Palatina 1096,
Acta Bamabae 583, 596, 502, 592, 677, 679, 786, 1223
611,627, 755 854, 1042, 1158, 1194, Anti-Marcionite Prologue
Acta Isidori 815,1046 1198 xliv
Acta Joannis 527, 907, Aesop 676 Antiochus I 549
1115 Aetius 505, 618 Antiochus Epiphanes 508,
Acta Petri et Andreae 182, Agnew, F. H. 60, cvii 591
527 After (Elisha b. Abuya) Antipho 526
Acta Petri et Pauli 450, 383 Antoninus 449
1045, 1232 Aland, B. 3, 12, 13, 15, 17, Apocalypse of Peter 591
Acts of Carpus, Papylus, 25, 26, 27, 28, xxii, 783; Apocry phon of James 70
and Agathonike 1087 see also BA Apollodorus 767
Acts of Paul and Thecla Aland, K. 2, 9, xix, xxii, Apollonius Dyscolus 537
330, 667, 673, 988, 779, 783; see also BA Apollonius (Grammaticus)
1100, 1115 Albright, W. F. 128 138
Acts of Philip 146, 232, Alciphron 82 Apollonius of Tyana 82,
298, 834 Alexander Jannaeus 142, 96, 580
Acts of Thomas 88, 93, 98, 423, 424 Apollonius Rhodius 969
148, 201, 275, 277, 452, Alexander, L. xliv, lxi Apollos 886-92
527, 580, 869, 892,979 Alexander of Aphrodisias Apostolic Constitutions
Actus Petri cum Simone 1225 506, 844, 983
413, 1115 Alexander (Pseudomantis) Appian 156,322,326,499,
Acworth, A. 1177, 1211, 591 744, 758, 770, 1137
1216 Alexander the Great 424 Applebaum, S. 1070, 1086
Adam 206 Altaner, B. 37, 38, 862 Apuleius 909, 1149, 1227
Adamantius 34 Amaru, B. H. 333 Aqiba (Akiba) 124, 659
Aejmelaeus, L. 962 Ambrose 427 Aquila 618, 953
Aelian 125, 514 Ammonius 73, 813, 867, Aquila (of Corinth)
Aelius Aristides 978 907, 1164 858-63, 877
1255
1256 INDEX

Arai, S. 318 792, 796, 798, 810, 813, 468,497, 569, 575, 602,
Arator lxxi 815, 830, 836, 848, 851, 617, 622, 623, 624, 625,
Aratus 843, 848, 849 852, 867, 870, 875, 905, 651, 660, 665, 667, 681,
Aretaeus 1222 908,911,912,913, 922, 687, 691, xxii, li, lxi,
Aretas 460,466 926, 952, 955, 968, 972, lxiii, lxxxi, lxxxix, cviii,
Argyle, A. W. 128 975, 978, 991, 996, cix, cx, cxiii, cxvii, 695,
Aristeas 352, 723, 781, 1012, 1039, 1046, 1066, 707, 726, 741, 760, 784,
1057, 1115, 1167 1074, 1076, 1078, 1098, 791, 798, 802, 823, 826,
Aristophanes 86, 98, 181, 1103, 1130, 1147, 1148, 828, 860, 877, 883, 892,
184, 230, 235, 239, 291, 1149, 1150, 1153, 1154, 898, 900,902, 905, 918,
317, 327, 350, 364, 426, 1162, 1166, 1167, 1183, 919, 929, 933, 944, 962,
445, 448, 508, 526, 617, 1190, 1194, 1197, 1199, 964, 974, 975, 981, 994,
638, 652, 742, 759, 789, 1200, 1203, 1204, 1207, 1053, 1090, 1098, 1102,
792, 793, 813, 870, 873, 1213, 1222, 1223, 1225, 1103, 1106, 1108, 1118,
891, 1024, 1059, 1062, 1226, 1228, 1229, 1243, 1143, 1152, 1159, 1177,
1073, 1103, 1107, 1141, 1245, 1252 1179, 1195, 1204, 1208,
1199, 1225 Baarda, T. 751,761 1209, 1234, 1235, 1248
Aristotle 70, 102, 168, Babrius 1113 Barth, G. cviii
254, 349, 353,458, 502, Bacher, W. 729, 810 Barth, K. 74, 83, 120, 138,
556, 741, 853, 873, 953, Bagoas 679 143, 144, 172, 248, 266,
983, 1096, 1130, 1154, Bailey, R. E. 379 288,290, 388,411,432,
1194, 1196,1198 Ballance, Μ. 663,673 449, 524, 585,602, 802,
Arnold, W. T. 626 Bamberger, B. J. 1053 864, 897, 951
Arrian 113, 989, 1212 Bammel, E. cviii, 695, Barthes, R. 490
Artemidorus 67, 618,906 707, 1025, 1248 Bartlett, V. 593
Artemis 922, 923 Bammer, A. 916, 923 Barton, J. lxxviii
Ascension of Isaiah 70 Barag, D. 216 Bartsch, H. W. lxxxi
Aseneth 441, 442 Barbi, A. 187 Basil the Great 7
Assumption of Moses 193, Barclay, J. M. G. cix Bates, W. H. 622
365 Bar-Jesus (Elymas) 609, Baucis 677
Athanasius 19, 33, 34 610, 613, 615, 617, 618 Bauckham, R. 695
Athenaeus 171 Barker, D. C. 784 Bauer, J. B. 707
Athenagoras 537 Barnabas, Epistle of 36, Bauer, W. 979; see also
Atomus 407 37, 39, 70, 173, 194, BA
Attalus II Philadelphus 229, 300, 354, 374, 514, Bauemfeind, O. 187, 622
690 718, 839, 861, 951, 983, Baugh, S. Μ. 916
Auffret, P. 823 1043 Baumbach, G. 280
Augustine 20, 21, 67, 68, Barnard, L. W. 37, 302 Baumgarten, J. Μ. 1017
71,74, 77, 99, 102, 117, Barnard, P. Μ. 18 Baur, F. C. xl, xli, xlii,
118, 147, 150, 196,427, Barnes, T. D. 823, 831, lxiii, lxxiii, lxxiv, cxiii,
449,493, 606, 616, 720, 832 cxviii, 1006
801 Barnett, P. W. 280, 1030 Bavel, T. J. van 106
Augustus 549, 592, 673 Bamikol, E. 419 Beare, F. W. 106, 438,
Aune, D. E. 561 Bams, J. W. B. 280, 286 459,476, 558
BA (W. Bauer, K. and B. Barrett, C. K. 6, 28, 47, Bede 6, 20, 66, 68, 78, 84,
Aland) 325, 352, 363, 175, 187, 191, 205, 224, 99, 112, 116, 137, 150,
367, 391,404,415,435, 229, 283, 286, 296, 309, 152, 179, 266, 277, 288,
483, 537, 557, 581, 582, 316, 323, 324, 333, 339, 292, 296, 312, 346, 350,
583, 588, 603, 640, 660, 381, 385, 395, 396, 398, 351, 375, 385,410, 424,
668, 676, 700, 717, 742, 402,405, 411,412, 413, 434, 449, 525, lxxi,
748, 755, 763, 772, 779, 423, 443, 450,460, 466, 1050
 INDEX 1257

Behm, J. 1095, 1177, 1188 701,713,749, 760, 787, Brock, S. 257, 259
Bell, H. I. 123, 393, 1097 797, 857, 883, 919, 943, Brodie, T. L. 395, 419
Bengel, J. A. 21 950, 955, 999, 1013, Broughton, T. R. S. 499,
Benko, S. 857 1041, 1063, 1090, 1098. 622, 921
Benoit, P. 55, 60, 91, 98, 1153, 1209; see also NS Brown, R. E. 89, 95, lxiii,
147, 158, 160, 165, 170, Blass, F. 18,22,23,24,25, cvii
213, 251,272,273, 535, 28 Brown, S. cviii
558, 568, 577, 587, 594, Blastus 589 Brownlee, W. H. 430
695, 707, 723 Blinzler, J. 568 Bruce, F. F. 206, 333, 622,
Bentley, R. 387 Blomberg, C. L. cix, 707 663, lxi, cvii, 707, 765,
Bentzen, A. 208 Boers, H. W. 128 999, 1070, 1081, 1234,
Berger, K. 187, 900, 909, Boethus 225, 286 1248
999, 1090 Böhlig, A. 568 Bryan, C. 752, 760
Bergmeier, R. 395 Boismard, M. E. 24, 25, Buchanan, E. S. 673
Bernice 574, lvii, cxiv, 28, 318, 333, 379, xix, Buchanan, G. W. 128
1135 xx, xxii, XXX, xxxi, 707, Buckler, W. H. 916,921
Bertram, G. 60, 64, 786 1177 Budesheim, T. L. 962,
Best, E. 60, 128, 598, 600, Boman, T. 707 1030
601 Boor, C. de 575 Büchsel, F. 194, 1063
Betz, H. D. 77,79,81,113, Borgen, P. cx, 707, 751, Bultmann, R. 53, 55, 76,
126, 161, 182, 183, 249, 761 78, 118, 135, 141, 151,
354, 356, 367, 374, 387, Borger, R. 775, 779 169, 170, 207, 231, 250,
405, 427, 434, 452, 613, Bormann, L. 775, 780 307, 314, 396, 400,552,
679, 772, 869, 908, 913, Bornemann, W. 642 564, 594, 597, 599, 600,
925, 936, 970, 996, Bornkamm, G. 325, 761, 604, 682, xxv, xciv,
1036, 1046, 1096, 1168 762, 772, 823, 853, 946, cviii, 702, 707, 710,
Betz, O. 106, 1030 1012, 1246 722, 740, 784, 831, 838,
Beutler, J. 663 Borse, U. 707 839,949, 950, 953, 979,
Beyer, H. W. 962, 975 Bousset, W. 203, 290, 515, 993, 995, 1030, 1036,
Beyer, K. 377 585, 628,660, 722, 736, 1043, 1158
Beyschlag, K. 395 737, 1043 Burchard, C. 80, 274, 275,
Beza, T. 5, 323 Bouwmann, G. 60 427,438, 457, 468, 469,
Bickermann, E. J. 544, lxi, Bovon, F. 216, 490, 532, 490, 544, 594, 629, 663,
1017 lxxxii, cvi, 962, 985, xc, cviii, 775, 796, 798,
Bieder, W. 272, lxxxi 991, 1234 823, 855, 883, 894, 916,
Bieler, L. 354, 358 Bowers, W. P. 765 934, 999, 1011, 1030,
Bietenhard, H. 182 Bowker, J. W. 128, 350, 1035
Bihler, J. 333, 379 622, 624, 781 Burger, J. D. 257
Billerbeck, P. 622, 628 Bowman, J. 208 Burkitt, F. C. 10, 158, 172,
Bion of Soli 425 Brandon, S. G. F. 80, 87, 199, 608, 615, 823
Birdsall, J. N. xx, xxii 103,497, 574, 575, 578, Burrows, Μ. 370, 430,
Bishop, E. F. F. 419, 598 586, xli, 863, 1013 447
Black, Μ. 11, 128, 140, Braun, H. 60,97, 104, 111, Burton, E. de W. 481
148, 172, 179, 180, 196, 140, 148, 163, 168, 208, Bussler, J. Μ. 490
206, 221, 237, 274, 280, 290,311,339, 376, 378, Cadbury, H. J. 72, 106,
288, 296, 307, 317, 326, 448, 633, 843, 954, 962, 257, 259, 261, 264, 476,
327, 366, 377, 392,402, 1017, 1025, 1143, 1161 481, 588, xxxi, xlv,
407, 419, 435, 467, 503, Brawley, R. L. cix xlviii, lxi, lxii, lxxxi,
514,541,561,562, 582, Bréhier, E. 309 916, 921, 999, 1003,
585, 593, 596, 614, 641, Bright, J. 352 1143, 1146, 1154, 1177,
669, 673, 675, 679, 685, Brinkman, J. A. 106 1189, 1216, 1223
1258 INDEX

CAH (Cambridge Ancient Chuza 570 Coggins, R. J. 333, 395

History) 740, 957 Cicero 135, 169, 171, 253, Cohen, S. J. D. 752, 759
Caiaphas 224, 225 255, 283, 341, 344, 466, Colaclides, P. 823
Calder, W. Μ. 767, 769 626, 632, 682, xlviii, Collins, J. J. 582
Caligula (Gaius) 573, 614 749, 759, 789, 801, 803, Colpe, C. 384
Calloud, J. 823 808, 817, 827, 832, 846, Colwell, E. C. 884, 891
Calvin, J. 400, 498, 952 905, 928, 936, 1050, Connolly, A. L. 876
Cambier, J.459 1094, 1111, 1118, 1131, Connolly, R. H. 1030
Campbell, A. cviii, 707 1153, 1187, 1195, 1221, Constantinus Porphyro-
Campbell, J. Y. 158 1229, 1231 genitus 1219
Campbell, L. 392 Clark, A. C. 23,24,67,68, Conzelmann, H. 141, 413,
Campbell, R. A. cviii 73, 99, 196, 229, 245, 672, xciv, xcvi, cviii.
Campeau, L. 280 289, 295f., 449f., 523, 823,943
Campenhausen, H. von 60, 537, 564, 581, 615, Coppens, J. 302, 395, 598,
91, 158, 333, 419, lxiv, 642f., 672, xxiii, 735, 884, 1217
lxxxi, xciv, cvi, cviii, 778, 803, 804, 820, 910, Corbulo 556
975, 999, 1005 955, 1002, 1075, 1079, Corpus Hermeticum 840,
Capper, B. J. 261, 263 1099, 1100, 1141, 1147, 841,842, 1042
Carcopino, J. 905 1153,1182, 1184, 1205 Corssen, P. 18, 23, 985
Carroll, J. T. cvii Clark, D. J. 1177, 1198 Cosgrove, C. H. cvi, 1053
Casey, R. P. 999, 1002 Clark, K. W xli Cramer, J. A. 676, lxxi
Cassidy, R. J. cx Clarke, W. K. L. 84, 103, Cranfield, C. E. B. 35. 77,
Casson, L. 1177, 1187 384,590, 1006 443
Catchpole, D. R. cix, 707 Claudius 527, 556, 559, Creed, J. Μ. 60
Cato 125 563, 564, 573, 574, 614, Crescens 31
Catullus 98 661, 673, lv, cxiv, 862 Crito 156
Causse, A. 106 Claudius Lucius Hermmia- Croesus 633
Cerfaux, L. 60, 106, 257, nus 591 Cullmann, O. 83, 84, 88,
395, xxxi, cviii, 707 Cleanthes 848 91, 95, 151, 152, 165,
Chambers, C. D. 593, 596, 1 Clement 35, 80, 95, 103, 170, 171, 194, 195, 199,
1133, 1135 104, 173, 194, 242, 243, 208, 214, 262, 290, 302,
Chapot, V. 930 244, 342, 542, 576, 578, 307, 318, 333, 339, 384,
Charito of Aphrodisias 931 636, 638, 683, 687, 688, 391, 398, 410, 432, 433,
Charles, R. H. 356, 744 xl, xliii, lxix, xcvi, cxiii, 568, 587, 681, lxviii,
Charlesworth, M. P. 1090, 718, 839, 844,953, 983, 707, 723, 1248
1095 1057, 1152, 1153, 1162, Cumont, F. 107, 1090,
Chase, F. H. 98, 197, 222 1231, 1237 1097
Chevallier, M. A. 107, 2 Clement 35, 37, 197, Cyprian 8, 18, 19, 20, 21,
128 288, 290, cxiii, 912, 27, 78, 118, 335, 646,
Childers, J. W. xxiii 978 1061
Christiansen, E. J. 158, Clementine literature 72, Cyril Lucar, Patriarch 4
cviii 192, 208, 234, 292, 336, Cyril of Alexandria 20, 27,
Chrysippus 200 347, 362, 405, 406,417, 102
Chrysostom, John 20, 21, 445, 785, 851, 1207 Cyril of Jerusalem 19
72, 112, 211, 220, 291, Clement of Alexandria 7, Dahl, N. A. 107, 187, 333,
304, 308,414, 417, 423, 18, 21, 315, 569, 571, 345, cviii, 707, 724
616,672,676, lxxi, 718, 646, 745, 840, 848, 913, Dalman, G. 66, 75, 85, 86,
798, 817, 841, 846, 863, 925, 993 99, 100, 140, 148, 159,
923, 936, 938, 978, 990, Clement of Rome 31, 35, 180, 191, 224, 259, 430,
1002, 1061, 1066, 1169, xliii, lxiv, xcvi 1023, 1028
1182 Cleopatra 549 Damasus 9
 INDEX 1259

Danby, H. 810, 1012, 344, 395, 574, 575, 977, Dio of Prusa 925, 940,
1059 1072 1027
Daniélou, J. 158, 333 Devine, C. F. 962 Dioscondes 618
Danker, F. W 707 Dibelius, M. 69, 101, 109, Dittenberger, W. 196, 699,
Danoff, C. Μ. 817 156, 168, 240, 245, 270, 733, 798, 852, 935, 938,
Dar, S. 1070, 1086 293, 298, 304, 321, 323, 958, 1012, 1020, 1027,
Daube, D. 113, 268, 276, 337, 382, 386, 397, 436, 1081, 1082, 1137, 1148
282, 294, 302, 315, 316, 486,491,496,511,571, Doble, P. 379
428, 431, 449, 454, 639, 585, 587, 589, 609, 625, Dobschütz, E. von 2, 5, 20
640, 752, 759, 1053, 640, li, lxxv, lxxvi, Dockx, S. 379, 382, 598,
1065, 1066, 1225 lxxxi, 710, 737, 765, lxii
Dautzenberg, G. 986 770, 841, 842, 843, 844, Dodd, C. H. 70, 130, 207,
Davies, J. G. 107 846, 848, 853, 854, 947, 212, 234, 390, 657,811,
Davies, P. 1248 956, 963, 1005, 1099, 880, 1165, 1166
Davies, W. D. 72, 75, 76, 1150, 1158, 1160, 1163 Doignon, J. 438
80, 101, 333, 339, 342, Didache 163, 178, 194, Dombrowski, B. W. 158
345, 413, 426,457, 473, 243, 244, 412, 432, Donaldson, T. L. 333
525, 529, 535, 749,772, 452, 561, 562, 602, 604, Donfried, K. P 89, 95,
1012, 1046, 1220, 1232 666, 688, 799, 951, 978, 807, 816
Dehandschutter, B. 389, 983 Dowd, W. A. 158
438 Didascalia, Syriac 506 Downey, G. 549
Deissmann, A. 85,96,257, Dieterich, W. 277, 707, Downing, F. G. 240, 333,
377,407,512, 604,616, 723 622, 663, cix, cx, 823
798, 908, 928, 959, 973, Digest 795, 1113, 1116, Dressier, H. 1106
1107, 1146, 1153 1131,1137, 1147 Drusilla 574
De Lacey, D. R. 459 Dillon, R. J. 128, 187 Dubarle, A. Μ. 823
Delebecque, E. 747, 752, Dinkier, E. 81, 207, 290, Duchesne, L. 606
901,916,943 419, 425, 432, 556, 587, Duensing, H. 986, 1000
Delitzsch, F. J. 508, 606 613, 857, 860, 871, Duncan, G. S. 916
Delling, G. 69, 121, 128, 1043, 1126, 1129 Dunn, J. D. G. 107, 438, c,
142, 155, 158, 171, 197, Dio Cassius 123, 308,424, 695,
213, 244, 245, 248, 289, 425, 564,611,758, 805, Duplacy, J. A. 192
290, 388, 405, 429,490, 830, 862, 873, 937, 958, Dupont, J. 53, 60, 91, 107,
501, 528, 622, cvii, 775, 1049, 1135 128, 149, 158, 187, 216,
783, 798, 977, 1009, Dio Chrysostom 67, 196, 238, 240, 251, 272, 280,
1115, 1248 744, 758, 940, 1026 302, 333, 377,438,459,
Demosthenes 73, 78, 101, Diocletian 628 490, 532, 558, 568, 593,
119, 134,203,204,211, Diodorus Siculus 67, 119, 596, 598, 622, 663, 689,
221, 269, 502, 503, 504, 142, 222, 424, 458, 676, xxxii, lxxiv, 707, 724,
513, 536, 580, 588, 590, 927, 1066, 1086, 1222 823, 962, 1133, 1136,
606, 675, 679, lxxx, Diogenes Laertius 253, 1143, 1217, 1234, 1237,
733, 749, 763, 788, 791, 838,913 1246, 1248, 1250, 1252
801, 830, 831. 833, 834, Diognetus, Epistle to 38 Easton, B. S. 1005
926, 932, 959, 973, 978, Dionysius of Alexandria Eckert, J. 707
990, 1005, 1006, 1046, lxx Edwards, Μ. J. 823
1049, 1057, 1059, 1067, Dionysius of Halicarnas- Egeria 110
1094, 1097, 1100, 1118, sus 81, 82, 142, 285, Ehrhardt, A. 586, 815,
1154, 1155, 1183, 1252 508, 852, 1094, 1141 874, 875, 877, 887, 931,
Demosthenes Ophthalmi- Dionysius the Areopagite 986, 1005, 1088, 1136,
cus (Aetius) 505 855 1146
Derrett, J. D. M. 91, 261, Dionysus 580 Elbogen, I. 781, 782
1260 INDEX

Elderen, B. van 608, 613, 841, 851, 936, 945, Franklin, E. 60, cvi
614, 663, 673 1027, 1100, 1130, 1158, Fredriksen, P. 438
Eleazar 508 1194, 1196, 1223 Fridrichsen, A. 1143, 1170
Eleazar of Modiim 1008 Eusebius 7, 19, 30, 31, 40, Fuller, R. H. 92, 171, 190,
Electra 70 45, 72, 102, 295, 308, 201, 204, 290, 438
Eliezer ben Hyrcanus 431 315, 380, 405,407, 422, Funk, R. W. 558
Elijah 206 564, 569, 570, 581, 583, Fusco, V. cix
Elliott, J. K. xxiii 589, 591, 693, xl, xli, Gabinius 424
Ellis, E. E. 128, 187, 383, lxiv, lxv, lxvi, 723, 736, Gaechter, P. 707
490, 532, 558, 561, 622, 792, 855, 993, 994, Gärtner, B. 140, 328, 663,
xlviii, cix, 986 1117,1183 823, 845, 846, 850,
Ehester, W. 823, 843 Eustathius of Thessalonica 1098
Elymas 262, 609, 613, 830 Gager, J. G. 302,438,490,
615,908 Evans, C. A. 128 501
Emerton, J. A. 60, 91,107, Evans, C. F. 128,187,490, Gaius (Caligula) 382, 466,
128, 158, 187, 216, 240, 622 614
333, 361, 622, 636, Exum, C. 962 Galazzi, C. 4
xxxii Ezekiel (tragic poet) 338, Galba 626
1 Enoch 196, 328, 336, 354, 355, 356 Galen 205,458, 506,1207,
1066, 1194 4 Ezra 113 1223, 1225
2 (Slavonic) Enoch 844 Fadus 293 Galerius 591
Enslin, M. S. 91 Farmer, W. R. 87 Gallio lvi, cxiv, 858, 870,
Ephraem Syrus 10, 597 Fascher, E. 438, 561, 916, 871
Epictetus 103, 166, 237, 919 Gamaliel 292, 293, 294,
300, 355, 393, 522, 729, Feldman, L. H. 615, 1012, 296, 297, 308, liv, cxiv
745, 763, 788, 793, 840, 1025, 1058 Gapp, K. S. 558
875, 879, 890, 905, Felix lvii, cxiv, 1080, Gasque, W. lxx, lxxii,
1036, 1048, 1222,1228 1081, lxxxi
Epimenides 97, 847 Fenton, J. C. 663 Gaventa, B. R. cvi
Epiphanius 19, 646, 840, Ferguson, E. 302, cviii Geagan, D. J. 823
1026, 1098 Festus 236, 445, lvii, cxiv, Geer, T. C. xxiii
Epistula Apostolorum 38, 835,1117, 1123 Geoltrain, P. 302
39,40, 577, xlii, lxiv Feuillet, A. 60 George, A. 216,303, lxxxi,
Epp, E. J. 6, 14, 28, 60, Filson, F. V. xxxiii, lxii, cix
102,187,514, xxiii, 819 737 Georgi, D. 1108
Erasmus 341, 353, 415, Finkelstein, L. 1053, 1063 Georgius Syncellus 351
1110 Finn, T. 490,501 Gerhardsson, B. 251, 253,
Erman, A. 424 Fischer, B. 9,10 313, cviii, 707, 709,
Esler, P. F. lxxvii, lxviii, Fitzmyer, J. A. 60, 72, 89, 713, 728, 729
lxxxi 107, 128, 147, 158, 163, Gerstinger, H. 3
Etymologicon Magnum 164, 171, 209, 210, 255, Gert, B. (K.-G.) 525, 542,
1194 307, 317, 350, 370, 403, 1166
Euler, K. F. 431 448, 566, lxxxii, 707, Giblin, C. H. 107, 128
Eupolis 153 723, 726, 888, 901, 904, Giet, S. 459, 558
Euripides 120, 142, 169, 916, 976, 1013, 1097, Gilchrist, J. Μ. 1177
183, 197, 201, 227, 235, 1152, 1226, 1244 Gildersleeve, B. L. 659,
269, 293, 294, 298, 362, Flusser, D. 216, 240 1189
374, 445, 449, 452, 483, Foakes-Jackson, F. J. 334 Giles, K. N. 472, 962
502, 506, 513, 516, 526, Foerster, W. 298,302,379, Gill, D. W. J. 438, 923
529, 580, 635, 652, 733, 785, 887, 1072, 1148 Gilmour, S. Μ. 107
743, 794, 796, 830, 833, Fränkel, Μ. 604 Glasson, T. F. 208
 INDEX 1261

Glombitza, O. 128, 303, Hansack, E. 1248 1138,1173, 1174, 1177,

318, 334, 379, 622, 765 Hansen, G. W. 769 1182, 1184, 1185, 1187,
Gog and Magog 245 Hanson, A. T. 1250 1190, 1193, 1194, 1195,
Goldsmith, D. 622 Hanson, R. P. C. xxiii, 1198, 1203, 1204, 1210,
Goodman, M. see NS 1177 1212, 1213, 1214, 1217,
Goodwin, W. W. 720 Harlé, P. 1144, 1172 1221, 1224, 1226, 1227,
Goodspeed, E. J. 857, 868 Harnack, A. von 53, 54, 1229, 1230, 1231, 1233,
Gordon, R. P. 622, 632 55, 98, 187, 281, 473, 1242, 1252
Gospel of Peter 482 547, 560, 564, xxiv, Hengel, M. 87, 118, 142,
Goulder, M. D. xli, xlii, xxxii, xliv, lxvii, lxviii, 187, 280, 303, 459, 468,
lxiii, cxiii, 707, 711 lxxxi, xciii, xciv, cviii, 469, 470,473, 487, 566,
Gourgues, M. 107, 395 972, 993 568, 622, cxvii, 765,
Grässer, E. 60, lxxii, lxxvi, Harrer, G. A. 608 766, 767, 775, 781, 786,
lxxxiii, cvii Harris, J. R. 18, 222, 250, 993, 1005, 1015, 1026,
Grant, R. M. 38, 158, lxii, 417,615, 1166 1036, 1078
708,752, 765 Hamson, P. N. 37 Hera 676
Grassi, A. 420 Harrisville, R. A. 107 Heracleides Ponticus 197
Gratus 224 Hart, J. H. A. 884 Heracleon 207
Gray, E. W. 769 Hartman, L. 128, 158, Heraclitus 562
Green, H. B. 622 cviii Hermas 35, 37, 75, 83,
Greeven, H. 598 Harvey, A. E. cviii 103, 343, 435, 456, 576,
Gregory, C. R. 4 Hatch, E. 89, 297, 342, 584, 585, 687, 692,
Gregory Nazianzen 1046 358, 835, 953, 967, cxiii, 876, 949
Gregory the Great 384, 1152 Hermes 676, 677
lxxi Hatch, W. H. P. 11, 13 Hermetica 163, 242, 449
Grelot, P. 622 Haufe, G. 159 Hermippus 382
Gribomont, J. 10 Haulotte, E. 490, 532 Herod Agrippa I 262, 564,
Gnesbach, J. J. 21 Hauser, H. J. 1234, 1248 569, 570, 571, 572, 573,
Grintz, J. M. 1030 Heckel, U. cxvii 574, 575, 576, 577, 579,
Grundmann, W. 303, 395, Hedrick, C. W. 438 582, 588, 589, 590, 591,
544,557,1114 Hegesippus 380, 693, xl, 592, 593, 595, 603,
Güting, E. 107, 122 xli, lxiv, 723, 1117 xxxiv, lv, CX1V
Gundry, R. H. 107, 116 Heitmüller, W. 199 Herod Agrippa II 445,573,
Haacker, K. 60, 107, 438, Helena of Adiabene 563 lvii, cxiv, 1134, 1135
490, 663, cvii, cviii, cix, Heliodorus 441, 635 Herod (Antipas) 241, 246,
857, 883, 893, 1030, Hemer, C. J. 62, 102, 122, 247, 590
1036, 1053, 1063, 1090, 124, 180, 197, 296, 324, Herod (the Great) 146,
1143, 1234 402, 453, 466, 483, 498, 246,436, 446, 573, 591,
Hadrian 156 503, 556, 563, 584, 603, 604
Haenchen, E. 14, xxiii, 614, 615, 628, 672, 676, Herodian 451, 758, 906,
708, 1177 678, xxxii, 740, 778, 936
Hahn, A. 527 779, 782, 786, 808, 809, Herodotus 82, 125, 140,
Hahn, F. 60,187,303, 398, 814, 816, 820, 823, 832, 142, 182, 227, 230, 260,
552, 672, lxxii 855, 857, 861, 862, 881, 278, 298, 301, 311, 356,
Halstead, S. 823 887, 893, 904, 908, 909, 359, 371, 373, 402, 426.
Haltgren, A. J. 438 921, 930, 935, 937, 938, 505, 515, 536, 582, 603,
Hama ben Hanina, 585 943, 947, 948, 952, 953, 619, 633, 667, 668, 677,
Hamm, D. 174 960, 991, 992, 995, 678, 718,780, 792, 799,
Hammond, C. E. 583 1003, 1027, 1058, 1061, 813,818, 830, 840, 851,
Hanina b. Dosa 193 1070, 1081, 1087, 1092, 880, 892, 921, 924, 988,
Hannas 225 1101, 1118, 1133, 1135, 1005, 1015, 1019, 1067,
1262 INDEX

1085, 1118, 1138, 1149, 92, 107, 128, 159, 174, 486, 547, 558, 560, 563,
1152, 1196, 1197, 1212, 187, 216, 240, 251, 272, 564, 566, 583, 604,
1225 357,438, 708, 721, 823, xxxii, xcii, ciii, 760,
Herondas 267 837, 901, 943,954,962, 783, 857, 863, 909, 983,
Hesiod 1193, 1223 984, 1030, 1036, 1234, 984, 1021, 1030, 1035,
Hesychius 70, 86, 285, 1238 1053, 1058, 1063, 1090,
292,451,503,516,675 Hort, F. J. A. 555, 744 1118,1155, 1241
Heutger, N. 1177, 1217 Houlden, J. L. cvi, 695, Jerome 8,9, 20,21,27,32,
Hickling, C. J. A. 1144 Howard, W. F. 593, 596, 34, 36, 51, 78, 152,268,
Hicks, E. L. 916,923,966 1133,1135 351,406, 564,616, xliv,
Hierocles the Stoic 954, Hughes, J. H. 884, 888 lxxi, 837, 994, 1050,
1238 Hull, R. F. xxiii 1078, 1117
Hiers, R. H. cvii Hunkin, J. W. 60, 67, 101, Jervell, J. 438, 490, xli,
Hill, D. 561 523 lxii, cxiii, 1000, 1234,
Hillel 211 Hurd, J. C. 596 1246
Hillier, R. lxxi, lxxxi Hyrcanus 446 Jewett, R. 862,952, 1118
Hipparchus 505 Iamblichus 676, 1183 Joanna 570
Hippocrates 267, 618,991 Ignatius 35, 36, 95, 103, Johanan ben Zakkai 431,
Hippolytus 406,416 104, 144, 173,201,300, 520
Hirsch, E. 438 374, 527, 557, 687, 688, Johanan b. Gudgeda 260
Hiyya bar Abba 202 lxiv, lxix, xcvi, cxiii, Johanan the Sandalmaker,
Hock, R. F. 857 778, 951, 977, 978, 297
Hoehner, H. W. 246 1077, 1166, 1183, 1231 John Hyrcanus 146
Holl, K. xciv Iliffe, J. H. 1017,1020 Johnson, L. T. cix, cx
Holmes, B. T. 568 Innocent 1400 Johnson, S. E. 263, 317
Holofemes 679 Irenaeus 7, 15, 16, 17, 26, Johnston, G. 187, 280
Holtz, T. 334, 438, lxii, 27, 45, 46, 47, 70, 97, Jonathan 65, 225
695, 708, 747 184, 196, 244, 304, 315, Jones, A. H. Μ. 549,935
Holtzmann, O. 174 422,433,450, xliv, lxvi, Joseph 198, 228
Homer 86, 115, 151, 234, lxvii, lxx, lxxix, cxiii, Joseph and Aseneth 330,
248, 298, 317, 388,402, 771 953, 1035
424,452,483,506,511, Isaeus 227,911,932 Josephus 66, 67, 82, 85,
549, 565, 580, 581, 618, Ishmael (ben Phiabi) 286, 86, 87, 88, 97, 98, 102,
626, 653, 657,660, 676, 1124 103, 105, 111, 114, 122,
729, 783, 788, 799, 837, Isho’dad of Merv 847 123, 124, 135, 142, 146,
845, 854, 919, 978, Isocrates 65, 255, 1036 161, 168, 178, 179, 182,
1047, 1100, 1193, 1195, Jacob, E. 162 183, 191,219,221,222,
1203, 1204, 1206, 1213 James, M. R. 38, 577 223, 224, 225,230, 237,
Hommel, H. 823, 847 Jasper, G. 1000 242, 244, 260, 265, 271,
Hondius, J. J. E. 264 Jastrow, Μ. 99, 234, 259, 282, 284, 291, 292, 293,
Honi the Circle-maker 267, 424, 749, 933, 294, 295, 308, 312, 313,
193, 200 1107 314, 326, 327, 328, 336,
Hooker, M. D. 962, 969 Jaubert, A. 92 341, 351, 352, 354, 356,
Horace 580,676,782,909. Jensen, J. 708 374, 378, 380, 382, 390,
978, 988, 1131, 1196, Jeremias, J. 54, 55, 60, 402, 407, 408, 423, 424,
1199, 1231 146, 147, 155, 164, 170, 426, 432, 435,436,446,
Horbury, W. 92, cix 174, 180, 201, 218, 219, 480, 482,483,484, 501,
Horsley, G. H. R. 857, 222, 223, 225, 230, 238, 505, 513, 537, 538, 549,
884,901,916 259, 260, 262, 266, 267, 563, 573, 576, 581, 586,
Horsley, R. A. 280 281,310, 324, 351,355, 588, 589, 590, 591, 603,
Horst, P. W. van der 60, 401, 402, 430, 431,484, 607,611,618,627,630,
 INDEX 1263

631, 634, 635, 648, 652, Justus of Chalcis 102 Koch, D. A. 395
655, 656, 659, 660, 668, Juvenal 362,501,549,630, Kodell, J. 303, 593
xl, xliii, cxiv, 698, 699, 781,877,909, 1135 Kosmala, H. 118, 383,
700, 723, 733, 737,742, Käsemann, E. xciv, xcvi, 392,448, 479, 480, 482,
745, 748, 763, 781, 783, cix, 884, 885, 889 484, 704, 759, 787, 888,
786, 792, 799, 802, 811, Kaestli, J. D. cvii 893, 899, 1098, 1157,
812,813,817,818, 830, Kamith 225 1161
833, 835, 836, 837, 840, Karrer, Μ. cviii, 962 Kraabel, A. T. 490, 501,
841, 845, 853, 862, 873, Karris, R. J. lxxi cix
879, 889, 903, 905, 908, Kathros (Kantheros) 286 Kraeling, C. H. 544
913, 929, 934, 935, 937, Katz, P. 334, 1234, 1245 Kränkl, E. cvii
941, 946, 947, 954, 958, Kaye, B. N. 747, 752 Kraft, R. A. 1177
959, 978, 979, 992, 997, Kearsley, R. A. 930 Kreitzer, L. J. 916
1008, 1012, 1020, 1021, Keck, L. E. 303 Kremer, J. 110, 111, 116,
1022, 1023, 1025, 1034, Kemmler, D. W. 807 785, 807, 1250
1036, 1046, 1057, 1058, Kennard, J. S. 128 Kühner, R. [K.-G.] 525,
1059, 1065, 1066, 1068, Kerigan, A. 128 542, 1166
1072, 1074, 1077, 1078, Kertelge, K. 159 Külling, H. 823
1079, 1080, 1081, 1085, Kilgallen, J. J. 962, 983 Kümmel, W G. 60, 385
1086, 1094, 1096, 1113, Kilpatrick, G. D. 107, 128, xxxviii, xlii, xliv, lviii,
1116, 1117, 1123, 1124, 187, 334, 377, 379, 384, lxxii, lxxiii, cvi, 708
1125, 1130, 1132, 1134, 622, 645, 646, 678, 679, Küng, H. cix
1135, 1136, 1137, 1148, xxii, 708,725, 728,752, Kurz, W.S. lxxxi, 962
1152, 1154, 1156, 1183, 807, 810, 884, 895, 926, Lachmann, K. 200
1185, 1188, 1190, 1193, 986, 992, 995, 1070, Lachs, S. T. 1053, 1066
1196, 1198, 1200, 1202, 1078, 1120 Lactantius 591
1207, 1208, 1210, 1214, Kippenberg, H. G. 1118 Ladouceur, D. 1177, 1217
1224, 1225, 1226, 1227, Kirk, J. A. 92 Lagercrantz, Ο. 663
1229, 1230, 1232, 1237, Kirschschläger, W. 1217, Lagrange, M. J. 24, 285
1251 1218 Lambrecht, J. 962, 966
Joshua b. Hanamah 260 Kittel, G. 1000 Lamouille, A. 24, 25, 28,
Jousseu, A. xxiii Klausner, J. 76, 264, 286, xix, xx, xxii, xxx, xxxi,
Jowett, B. 451 933,1025 1177
Jubilees 111,351,356,734 Klein, G. 75,102, cviii, 912 Lampe, G. W. H. 107,400,
Judah, R. 575 Kleiner, G. 959 978
Judah the Prince, R 211 Klijn, A. F. J. 334, 564, Lampe, P. 857, 901, 916
Judas Galilaeus 294, 295 lxxxi, 708 Larsson, E. 303, lxii, 1234,
Judge, E. A. lxxviii, lxxxi, Klinghardt, Μ. 708 1246
807, 815, 1082 Knauf, E. A. 459 Laud, W. 6, 20
Jülicher, A. xxiii Knibb, Μ. A. 328 Lawlor, H. J. 31, 591
Julius Caesar 314, 446, Knopf, R. 243 L.-B. (Lipsius, R. A. and
549, 1192, 1211 Knox, J. 47, lxvi, lxvii, Bonnet, M.) 93, 95, 146,
Juster, J. 802 lxviii, lxxxi 148, 182, 201, 232, 269,
Justin Martyr 15, 41, 42, Knox, W. L. 54, 118, 149, 275, 277, 298,413,417,
43, 44, 115, 183, 234, 152, 160, 197, 198, 294, 450, 527, 583, 596, 611,
244, 361,374,403,405, 327, 338, 339, 364, 371, 627, 834, 845, 869, 892,
412,415,416, 452, 502, 376, 424, 427, 428, 434, 907, 979, 988, 1045,
638, 646, 683, xlii, xlix, 497, 635, 638, 672, 680, 1101, 1115, 1232
lxiv, lxv, lxvi, 836, 953, 681, 793, 954, 969, 970, L. Cornelius Sulla 499
978, 1009, 1137, 1148, 983, 1025, 1056, 1093, Leary, T. J. 901,907
1166, 1223, 1244 1158, 1167 Lebram, J. C. 823
1264 INDEX

Le Déaut, R. 112, 206, Lucian 21, 77, 79, 83, 100, Mare, W. H. 334
300, 333, 345, 346, 356, 125, 161,181, 204, 248, Marin, L. 490, 532
383, 503, 577, 1160 249, 255, 308, 326, 352, Markland, J. 820
Lee, G. M. 765, 769, 901, 354, 358, 382, 387, 390, Marshall, I. H. 60, 61, 62,
916, 1090, 1113 432, 452, 543, 557, 591, 71, 74, 107, 114, 129,
Leenhardt, F. J. 438 659, 676, 687, 766, 772, 139, 228, 231, 384
Légasse, S. 1017, 1030, 788, 792, 793, 809, 814, Martial 905
1053, 1070, 1090, 1120, 817, 834, 873,908, 913, Martin, R. A. xxxii
1133, 1144 922, 938, 954, 979, 989, Martini, C. M. 187, xx,
Legrand, L. 823 1094, 1113, 1114, 1115, xxiii
Le Moyne, J. 219 1139, 1167, 1168, 1184, Martyrdom of Justin 1087
Lentzen-Deis, F. 379 1185, 1186, 1193, 1194, Martyrdom of Polycarp
Lerle, E. 663 1198, 1201, 1221, 1222, 930, 998, 1023
Leszynsky, R. 219, 1065 1227 Martyrium Andreae Prius
Lewis, N. 123 Lucifer of Cagliari 8, 19 95
Libanius 343 Lucius Caecilius Metellus Marutha of Maipherkat
Lieberman, S. 1024, 1094 446 lxviii
Liechtenhan, R. 459 Lucius Valerius 446 Masson, C. 92,459
Lienhard, M. 303 Lucretius 717, 850 Mastin, B. A. 901, 909,
Lietzmann, H. 38, 165, Lüdemann, G. 395, lxxvi, 916
318, 708 lxxvii, 857 Mattill, A. J. lxxxi, cvii,
Lieu, J. M. cix Lührmann, J. 1053 823, 1090
Lifshitz, B. 544 Lundgren, S. 438, 1030 Mattili, A. J. and M. B.
Lightfoot, J. B. 90, 416, Luther, M. 400, lxxi, lxxii, lxxii, lxxxi
560, 583, xlii, lxiii, 951, 841 Mattingly, H. B. 544, 556
1182 Lyonnet, S. 159, 438, Maurer, C. 1057, 1090,
Lilly, J. L. 438 1030 1106
Lindars, B. 149,943 Lysanias 573 Maximus of Tyre 925,
Lindijer, C. H. 420 Lysias 69, 70, 192, 203, 1039
Linton, O. 107, 459 833 Mayser, E. 283
Livy 82, 105, 237, 341, 4 Maccabees 393, 508, McCasland, S. V. 438
779, 801, 808, 828, 913, 679,698, 1152 McDonald, J. I. H. 708
927, 986, 1050, 1220 Mackenzie, R. S. 490,622, McEleney, N. J. cix
Lobeck, C. A. 799,1228 xxiii, 823 McHugh, J. 90
Löning, K. 490 Mackinnon, D. M. 159 Mealand, D. L. 60, 159,
Lösch, S. 420, 1090, 1095, Maddox, R. 62, 77, 78, 169, lxii, cx, 1248
1100 102, 136, 156, 220, 336, Meecham, H. G. 685
Lövestam, E. 645, 962 383, 387, 457,474, 666, Meeks, W. A. 208, 549,
Loewe, R. 9 686, 825, 1014, 1045, lxxvii, lxxxi, 863
Lohfink, G. 84, 92, 187, 1064, 1097 Men 407, 660,676
438, cix Malherbe, A. J. ex, 1144 Ménard, J. E. 187,240
Lohmeyer, E. 80,473,744 Manaen 570 Menoud, P. H. 28, 60, 62,
Lohse, E. 107, 112, 154, Manek, J. 708 81, 92, 107, 159, 261,
163, 194, 223, 292, 316, Manson, T. W. 327, 530, 263, 438, 459, 622,
370, 396, 583, 605, 606, cviii cviii, 708, 735
1026 Manson. W. 340 Merkel, H. cix
Longenecker, R. N. 196, Marcellus 382 Metatron 84, 383
204 Marcion 46, 47 lxv, lxvi, Metzger, B. Μ. 2, 8, 9, 10,
Longinus 1167 lxvii, lxviii, lxx, cxiii 11, 13, 14, 24, 32, 60,
Longus 1204 Marcus Aurelius 166, 635, 107, 121, 122
Lucan 342, 586, 986 875 Meyer, B. F. cix
 INDEX 1265

Meyer, F. E. 280 474, 510, 514, 517, 530, 1012, 1021, 1024, 1026,
Meyer, R. 604 537, 542, 592, 638, 643, 1028, 1037, 1041, 1049,
Michaelis, J. D. lxxiii 669, 692, xlviii, 736, 1068, 1073, 1074, 1082,
Michaelis, W. 544 749, 768, 771, 778, 788, 1084, 1087, 1096, 1098,
Michel, H. J. 963 816, 817, 842, 876, 891, 1106, 1113, 1123, 1126,
Michel, O. 379, 438 896, 898, 914, 916, 922, 1137,1139, 1140, 1151,
Middleton, J. F. 138 928, 937, 938, 939, 940, 1165, 1191, 1194, 1212,
Miles, G. B. 1177, 1217 963, 964, 973, 978, 1253
Millar, F. see NS 1003, 1025, 1039, 1040, Neirynck, F. 490, 522,
Miltner, F. 901, 916, 928 1045, 1080, 1085, 1096, lxxxi
Minguez, D. 420 1101, 1102, 1106, 1107, Nellessen, E. 92, 663
Minnen, P. van 775 1110, 1115, 1136, 1137, Nero 424, 536, 563,626
Minucius Felix 953 1140, 1141, 1150, 1155, Nestle, Eberhard 272, 280,
Mircea, I. 159 1156, 1163, 1165, 1166, 568, 1153, 1165, 1177
MM (Moulton, J. H. and 1169, 1172, 1185, 1192, Nestle, Erwin 2, 5, 20
Milligan, G.) 100, 142, 1211, 1223, 1228, 1251 New, S. 182
200, 265, 266, 349, 483, Moule, H. W. 245 Neyrey, J. lxxvii, 1030
484, 518, 554, 583, 588, Moulton, W. F. 503 Nibley, H. 60
729, 730, 739, 748, 759, Mullins, T. Y. 598 Nickle, K. F. 559, 1108
785, 799, 850, 852, 867, Munck, J. 438,963, 1006 Nicolas of Damascus 424
904, 908,911,912,913, Mundle, W. 334 Nock, A. D. 227, 373,405,
922, 925, 927, 932, 952, Muratorian Canon xxvii, 605, 823, 833, 925,
973, 982, 993, 996, xliv, lxx 1006, 1191
1026, 1049, 1073, 1074, Murphy-O’Connor, J. 85, Nolland, J. 708, 719, 720
1084, 1086, 1088, 1093, 146, 598, lxii, cix, 708, Noorda, S. J. 251, 257,
1106, 1107, 1111, 1136, 807, 812, 857, 860, 862, 272
1137, 1140, 1145, 1148, 871, 884 Norden, E. 243, 823
1153, 1173, 1190, 1191, Musonius Rufus 721, 984, North, J. L. 708, 713, 857,
1199, 1207, 1251, 1253 1036 875, 1088
Moehring, H. R. 438, Mussner, F. 129, 159, 187, Noth, Μ. 352
1030, 1039 318, 334, 379,490 Noy, D. 775
Moeris 503, 799, 860, Musurillo, H. A. 456, NS (Schürer, E.; revised by
1130 1046 G. Vermes, F. Millar,
Mœssner, D. P. 187, cvii, Myers, J. L. 614 Μ. Goodman and Μ.
1235 Naber, S. A. 1196 Black) 218, 219, 223,
Molland, E. 708 Naiamrn
e, A. 593 224, 260, 292, 323, 324,
Momigliano, A. 790, 862 Nauck, W. 280, 823 325, 382, 402, 407,423,
Mommsen, T. 33, 549, ND (Horsley, G. H. R., et 435, 436, 446, 447, 454,
775, 802, 909, 1030, al.) 138, 151, 156, 204, 482, 515, 549, 550, 566,
1087, 1092 255, 263, 286, 289, 309, 573, 592, 626, 628, 781,
Morgan, R. C. xxxi, 311,349, 378, 384, 387, 802, 817, 828, 860, 861,
lxxviii 407,425, 435,454,480, 872, 878, 903, 908, 959,
Morgenstern, J. 590 483, 504, 525, 527, 579, 988, 989, 1003, 1020,
Mosbech, H. 92 583, 629, 660, 661, 701, 1021, 1023, 1028, 1056,
Moses 208, 355-67 702, 716, 729, 730, 763, 1063, 1065, 1080, 1117,
Moule, C. F. D. 60, 67, 70, 766, 781, 782, 783, 784, 1134, 1135, 1181, 1183,
72, 75, 101, 138, 187, 786, 793, 797, 808, 812, 1230, 1237, 1238
220, 232, 239, 240, 276, 814, 841, 850, 852, 864, Numenius 446
296, 303, 318, 334, 357, 872, 876, 877, 878, 890, Oberwies, Μ. 1177
362, 367, 376, 384, 387, 904, 905, 908, 922, 923, Odes of Solomon 149
415, 428, 451,453, 458, 930, 946, 948, 982, 996, Oepke, A. 205
1266 INDEX

Ogg, G. 663 PGenève (= Geneva et Pauli 413, 417, 450,

Ogilvie, R. Μ. 1177, 1193 Papyri) 530 1045
O’Neill, J. C. cvi, 1090 PHib (= Hibeh Papyri) Pathrapankal, J. 438, cvii
Omas 441 518 Patsch, H. 558, 986
Orchard, B. 695 PHolm (= Papyrus Paullus Fabius Persicus
Oribasius 378 Graecus Holmiensis) 613
Origen 7, 18, 27, 74, 89, 699 Paulus 1050
147, 198, 207, 234, 356, Pland (= Papyn Pausanias 197, 424, 819,
405,406,580,616, lxxi, landanae) 946 836, 837, 923, 927,
735, 794, 854, 908, PKöln (= Kölner 946
1243 Papyri) 276 Payne Smith, J. 274, 863
Orosius 564, 862 PLond (= Papyri in the Perictyone 445
Orphic Fragment 849 British Museum) 1167 Periplus Maris Erythraei
Osborn, E. F. 41, lxv PLond medicus (= 86
Osburn, C. D. xxiii London Medical Perrot, C. 708
Oster, R. 107, 916 Papyrus) 505 Persius 841, 953
O’Toole, R. F. 129, 420, PMacquarie (= Pervo, R. I. xlix, lxii
622, cvii, 823, 1144 Macquarie Papyri, Pesch, R. 379, 383,708
Oulton, J. E. L. 31, 395, Australia) 4 Petersen, T. C. 14, 361,
591, 884 PMagdola (= Papyrus xxiii
Overman, J. A. 490, 501, de Magdola) 1094 Petersen, W. 577
cix PMich (= Michigan Peterson, E. 545,598,602
Ovid 114, 249, 387, 580, Papyri) 2,585,1153 Petronius 432, 588, 1205
677, 1148 PMil Vogl (= Milan Petzer, J. H. xxiii
Owen, Η. P. 379, 823, 840 Papyri) 4, 1073 Pfitzner, V. C. 107
Palmer, D. W. 60 POxy (= Oxyrhynchus Phiabi 225, 286
Panagopoulos, J. cvii Papyri) 2, 142, 204, Philemon 677
Panten, K. E. xix 219, 253, 325, 340, Philip (Herod) 573
Papias 37,93,98,570,583 355, 456, 502, 518, Philo 65, 67, 82, 98, 104,
Pappas 676 527, 577, 661, 787, 109, 111, 113, 117, 121,
Papyn 1026,1047,1066, 122, 123, 125, 169,182,
BGU (= Berliner 1080, 1095, 1113, 197, 205, 207, 208, 242,
Griechische 1148, 1153, 1167, 285, 308, 313, 320, 323,
Urkunden) 85,934, 1168, 1189 338, 339, 341, 342, 343,
1221,1251 PStras (= Strasbourg 350, 354, 355, 356, 374,
PAmh (= Amherst Papyri) 525 386, 393,405,407,480,
Papyri) 692,770 PTebt (= Tebtunis 505, 520, 577,611,618,
PBodmer (= Bodmer Papyri) 285,668, 626, 628, 629, 656, 698,
Papyri) 3 781, 1086, 1111, 732, 736, 777, 781, 786,
PChester Beatty (= 1131 811,817,818, 830, 835,
Chester Beatty Papyri) 3 PVindobG (= Vienna 839, 844, 845, 848, 850,
PColl Youtìe (= Papyri Papyri) 3, 102 861, 975, 1010, 1020,
published in Honor UPZ (= Urkunden der 1057, 1072,1082, 1096,
of H. C. Youtie) 448 Ptolemäerzeit) 1058 1106, 1111, 1125, 1138,
PEg (= Egerton Papyri) Parker, D. C. xix, xxiii 1158, 1185, 1188, 1193,
226 Parker, P. 60, 460, 545, 1196, 1202,1251
PFlor (= Florence 593 Philostratus 82, 134, 322,
Papyri) 782,926 Parker, S. E. xxxii 358,424,434, 577, 580,
PGM (= Papyri Parratt, J. K. 884 783, 837, 1046, 1190
Graecae Magicae) Passio Andreae 269 Photius 227
908,912,913 Passio Apostolorum Pétri Phrynichus 307, 353, 425,
 INDEX 1267

556, 578, 579, 584, 589, 1111, 1125, 1130, 1138, Quesnel, Μ. 411, cviii
659, 682, 887, 1023, 1166, 1183, 1190, 1191, Quintilian 70, 741, 1094,
1095, 1207, 1228, 1245 1193, 1197, 1199, 1200, 1149
Pickering, S. 4 1202, 1223, 1228 Quirinius 294, 295
Pindar 248, 659, 1135, Pohlenz, Μ. 663, 823, 841, Qumran Sect 104, 163,
1158, 1189 843, 847, 848 164, 167, 168, 252, 263,
Places, É. des 823 Pokorny, Ρ. 129, 159, lxii, 271,307,311,338,447,
Plato 125, 156, 163, 166, 1177, 1180 480, lxxxviii, xcix, evi,
168, 183, 197, 200, 205, Polyaenus 1222 738, 793, 954
227, 232, 233, 236, 237, Polybius 67, 155, 156, Qumran Texts
252,297,311,329, 355, 220,435, 455, 473, 499, 1QS 74, 88, 96,
367, 368, 371, 414, 432, 555, 588, 744, 763, 789, 97, 103,
451,483, 507,516, 524, 1072, 1086, 1196, 1211, 104, 105,
579, 584, 590, 626, 638, 1221 111, 124,
648, 672, 683, xlviii, Polycarp 11, 36, 37, 87, 162, 163,
702, 749, 773, 799,811, 98, 144, 325, 330, 456, 167, 171.
813, 829, 830, 834, 840, 542. 687, lxiv, lxix, 172, 208,
845, 846, 853, 879, 905, 981 255, 263.
912,914, 932,968,997, Polycrates of Ephesus 40 311,360,
1003, 1020, 1059, 1094, Pomponius Mela 1195 376, 383,
1096, 1114, 1115, 1118, Pontius Pilate 195, 224, 415, 448,
1130, 1138, 1149, 1167, 241, 246, 247, 382, 641 843, 888,
1168, 1172, 1183, 1189, Porphyry 906 954, 981,
1190, 1191, 1196, 1198, Posidippus 308 1025, 1161
1201, 1225, 1228, 1244, Posidonius 847 Damascus Document
1253 Powell, D. cviii (CD) 85, 96, 97,
Plautus 393, 1158 Praeder, S. M. xxxii, 943, 162, 255,
Pliny Major 277,424, 425, 1177, 1217 369, 370,
480, 549, 661, 809, 958, Pratscher, W. 708, 723, 371,378,
1195 963 415. 448,
Pliny Minor 171, 557, 588, Preisker, H. 884, 1096 566, 659,
861, 927, 951, 1094, Prentice, W. K. 568 726, 811,
1136, 1156, 1167, 1233 Prickard, A. O. 65 889, 975,
Plümacher, E. 67, 96, 134, Prigent, P. 3, lxv 1060, 1244
143, 160, 202, 234, 237, Priscilla 861 1QM 633,843,
258,290, 293,311,353, Prometheus 232 1025
405,414,421,422, 432, Protevangelium of James 1QSa 96,97,171
455, 478, 487, 568, 617, 1077 1QSb 290
664, xxxv, lxii, lxxii, Psalms of Solomon 156, 4QFlor 147,245,
lxxvi, lxxviii, lxxix, 963 645, 1158 290, 726
Plutarch 97, 105, 116, 117, Pseudo-Aristotle 848 4QTest 208,209
196, 200, 233, 237, 327, Pseudo-Philo 341, 346, 1QpHab 376,653,
341, 352, 353, 355, 376, 356 1020
382, 392,425, 451, 515, Pseudo-Phocylides 914 Genesis Apocryphon
522, 591,617,618, 660, Pseudo-Plato 907 (1QapGen) 341, 1225
678, 687, 729, 785, 789, Pseudo-Plutarch 983 1QH 84, 144,
791,793,813, 827, 835, Ptolemy 1192, 1202 328, 811,
836, 840, 850, 905, 913, Ptolemy VIII 446 840, 843,
925, 929,947, 949, 956, Pummer, R. 334, 339, 1160
983, 991, 1008, 1034, 395 4QPs37 147
1073, 1088, 1095, 1096, Quasten, J. lxxi 4QExod.4 350
1268 INDEX

Rabbinic writings Yebamoth M 424; 1020

Mishnah, Tosephta, B 564, 699, Kelim Μ 179,
Talmudim 759 1020
[M = Mishnah; T = Ketuboth M 486; Oholoth Μ 516
Tosephta; B = Talmud B639 Niddah Μ 424
Babli; J = Talmud Yer- Nedarim B 616, 1072
ushalmi] Nazir M 1011 Targumim
Berakoth M 180, 362, Sotah M 292, 573, (Fragment) Genesis 435
483, 682; 628; B 292, (Pseudo-Jonathan)
T 178, 180; 616 Genesis 383
B 178, 180, Gittin M 616, (Onqelos) Genesis 474
202, 245, 1086 (Yerushalmi 1)
646,659, Kiddushin B 486 Exodus 85,
810, 1223; Baba Qamma M 212; Numbers 1167
J85 B292 (Fragment) Numbers
Peah M 223, 310, Baba Metzia T 310; 114
312; T 312 B310 1 Samuel 636
Shebiith J 734 Baba Bathra M 486; (Jonathan) Isaiah 375
Shabbath M180, B 86, 310, Psalms 149
483,486; 312, 520, Ecclesiastes 104, 1197
B 111,223, 1138 Esther 502
407, 505, Sanhedrin M 96, Amos 728
736, 1025; 223, 225,
J211, 1188 234, 386, Midrash Rabbah
Erubin B 85 392, 575, Bereshith (= Genesis)
Pesahim B 111,211, 679, 785, 350, 383,
286, 486, 1065, 1072; 501, 585
505,734, T 1025; Exodus 354
1058 B 203, 223, Leviticus 265
Shekalim M 284; 726, 734, Numbers 115
B810 1025, 1073, Midrash Psalms 145,
Yoma M 104, 285, 1223; J 501, 509
1188; 734 Midrash Qoheleth 501
T179; Makkoth B1060
B 179, 223, Shebuoth B 1059 Other writings
635; Eduyoth M 223, 285; Mekhilta Exodus 78, 85,
J225 T564 529,743,
Sukkah T 564 Abodah Zarah M 992; 781
Betzah J146 B 222, 392, Aboth de R. Nathan 431
Rosh ha-Shanah B 124, 850 Tanhuma 137,501
223,659 Aboth M 96, 234, Derek Eretz Zuta 146
Taanith M 193, 200; 346, 366, Pesiqta Rabbathi 378
B504; 718, 1008, Pirqe R. Eliezer 1197
J501 1036 Rabbuia 10,11
Moed Qatan M 234, Zebahim M 300 Radermacher, L. 355, 417,
392,435; Menahoth T 286; 428, 455, 538, 585, 635,
B 264, 616, B346 674, 773, 794, 795, 802,
659 Kerithoth M 210 876, 892, 960, 975,
Hagigah B 84,407, Tamid M 104 1110, 1137, 1141, 1192,
616; J 114, Middoth M 179, 191, 1200, 1207, 1211
146, 211 223, 286, Rahlfs, A. 537
 INDEX 1269

Ramsay, W. M. lxii, 720, Routh, Μ. J. 37,93, 98 Schulz, F. 1030, 1048

767, 768, 769, 780, 782, Rowland, C. 383 Schwartz, D. R. 61, 272,
789, 794, 819, 855, 928, Rudolph, K. 395 708, 775
958, 1050, 1237 Rüger, H. P. 129, 1030, Schwartz, E. 1019, 1118
Rapske, B. 777, 790, 793, 1038, 1155 Schweizer, E. 92, 162,
795, 801, 1063, 1200 Ruphus of Ephesus 377 205, 250, 296, 316, 536,
Rashi 135 Russell, D. S. 219 542, 559, 602, 606, 622,
Rees, B. R. 185 Rutherford, W. G. 556, 646, cvii, 884, 888,
Reicke, B. 54, 89, 170, 578, 584, 660, 682, 793, 1056, 1062, 1161
171,281,660, cvii, 695, 794, 887, 958, 970, Schwyzer, E. L. 283
708, 738, 799, 921, 950, 1037, 1207 Scobie, C. H. H. 187, 334,
1177 Sabbe, Μ. 379 339, 395
Rengstorf, K. H. 92, 93, Saddok 294 Scrivener, F. H. 404, 984
310, 649, cviii Sahlin, H. 708, 728, 963, Scroggs, R. 303
Renié, J. 92 973, 986, 1018, 1019, Seccombe, D. 159, 303
Rese, Μ. cix 1120, 1129 Seesemann, H. 718, 963,
Reumann, J. 89, 95 Sallust 80 967
Reynolds, J. (Μ.) 500, cx, Sallustius 605 Segal, A. F. 490
1144, 1165 Samuel 210 Segal, M. H. 1028
RGG (Die Religion in Sanders, H. A. 2 Segbroeck, F. van lxxxi
Geschichte und Gegen- Sanders, J. T. cix, cx Seifrid, M. A. cix
wart) lxxii Sandnes, K. O. 823 Seleucus I 549, 610
Richard, E. 334, cx, 695, Sandt, H. van de 708 Selwyn, E. G. 980
708 Saum, F. 1248 Semler, J. S. lxxiii
Richardson, A. 384, 404, Schaeder, H. H. 140, 1098 Seneca 252, 387,424, 841,
556, 604, 788 Schechter, S. 447 845, 846, 850, 983,
Riesenfeld, H. 490, 522, Scheidweiler, F. 261, 267 1229
943, 950 Schenke, Η. Μ. 13 Sententiae Pauli 1131
Riesner, R. 857 Schermann, Τ. 376 Sergius Paulus 262, 609,
Rimaud, R. 240 Schlatter, A. 101, 229, 611,613, 614, 616, 618,
Roberts, C. H., Skeat, 307, 381,467 lv, cxiv
T. C. and Nock, A. D. Schlier, H. 60, 490 Servius 425
373, 786 Schmid, L. 1144, 1158 Sevenster, J. N. 85, 324,
Robertson, A. T. 593, 832, Schmidt, A. 708 498, 1020, 1024, 1085,
836, 1062 Schmiedel, P. W. 99, 615, 1094
Robins, V. K. 1177 739,768, 1171 Sextus Empiricus 308
Robinson, D. F. 460, 558, Schmithals, W. cviii, 963 Sheeley, S. M. lxxix, lxxxi
568 Schmitt, J. 622 Sherwin-White, A. N. 779,
Robinson, J. A. T. 187, Schnackenburg, R. 490, 788, 789, 790, 801, 802,
190, 202, 204 689 814, 815, 816, 868, 872,
Robinson, J. Μ. 542, 983 Schneider, G. 99, cvii, 873, 923, 930, 935, 937,
Rodgers, P. R. 129 1090 938, 940, 1020, 1026,
Rohde, E. 424 Schmder, F. 277 1027, 1048, 1087, 1092,
Ropes, J. H. 111, 112, Schœps, H. J. 347, 732, 1097, 1109, 1131, 1138,
1000, 1002, 1152 736 1174, 1233, 1242, 1251,
Rordorf, W. 943 Schrenk, G. 341 1252
Ross, J. Μ. 689, 857, 880 Schubert, P. 129 Sibylline Oracles 208, 840,
Rostovtzeff, M. I. 123 Schürer, E. see NS 1152
Roth, C. 92 Schürmann, H. 598, cviii, Silva, M. 216, 663
Rougé, J. 1177 963 Simeon b. Shetah 193
Rousseau, A. 15 Schuler, C. 807, 814 Simon, M. 334, 708
1270 INDEX

Simonson, G. 1107 Steck, O. H. 438, 1030 Synesius 852

Simplicius 1203 Stegemann, W. lxii, 775, Tacitus 312, 323, 436,
Skarsaune, Ο. 44 1030, 1051, 1070, 1144 515, 556, 557, 564, 581,
Skeat, T. C. 373 Stegmüller, F. lxx, lxxxi 586, 591, lxxviii, 801,
Sleeper, C. F. 107 Stempvoort, P. A. van 61, 937, 988, 1080, 1081,
Slingerland, D. 857, 862, 81,82 1113, 1117, 1135, 1137,
871 Stendahl, K. 107,263, 307, 1148
Smalley, S. S. 107 317,410,442 Tajra, H. W. 779, 800,
Smallwood, E. M. 216, Stenger, W. 107 805, 814, 862, 871, 872,
382 Stenning, J. F. 375 873, 960, 1020, 1026,
Smith, B. D. T. 884 Stem, M. 1090 1048, 1051, 1087, 1127,
Smith, D. E. 857 Steuemagel, G. 438 1131, 1136, 1232, 1233,
Smith, G. A. 423, 1085 Steyn, G. J. 1245 1252, 1253
Smith, J. 1177, 1178, Stobaeus 270,445, 1238 Talbert, C. H. 558, lxxxi,
1184, 1187, 1192, 1198, Stoops, R. F. 916, 940 cvii, cx, 695, 708, 857,
1202, 1203, 1204, 1210, Stowers, K. 963 962
1217, 1228 Strabo 122, 123, 252, 322, Tannenbaum, R. 500, cx
Smit Sibinga, J. 476 402, 423, 425, 549, 793, Tannehill, R. lxxix, cix
Smothers, E. R. 708 923, 988, 1026, 1068, Tam, W. W. 812
Socrates 156, 234, 237, 1213, 1221 Taylor, B. E. 901
283, 729, 824, 827, 828, Strack, H. L. 264 Taylor, J. 708, 1053, 1066
829, 830, 831, 834 Strange, W. A. xxiii, 884, Taylor, R. O. P. 257
Soden, H. von 8, 18, 408, 894, 901, 916 Tcherikover, V. 123, 324,
xciv Strathmann, H. 604, 1053, 325, 737, 989, 1026,
Sohm, R. xciii, xciv 1057 1086
Sokolowski, F. 916 Strecker, G. 336, 558, 560, Terence 112, 417, 1158,
Sol Invictus 407 601,603 1169
Solon 102 Streeter, B. H. 3, 295, Tertullian 18, 27, 47, 70,
Sophocles 151, 191, 196, xxiii 169, 171, 178, 230, 253,
227, 237, 248, 326, 392, Strobel, A. 61, 303, 558, 400, 405,452, 591, 646,
414,452, 456, 484, 592, 568, 577, 580, 582, 708, lxvi, lxvii, lxx, cxiii,
617, 638, xlvii, 733, 963,964 793, 837, 838, 953, 977,
833, 836, 847, 854, 924, Strom, M. R. 568 1137
1042, 1056, 1104, 1115, Stroumsa, G. G. 1053 Testament of Benjamin
1156, 1160, 1183, 1194, Stuehrenberg, P. F. lxx, 934
1207, 1223 lxxi, lxxxi Testament of Gad 236
Souter, A. 32 Suetonius 323, 556, 557, Testament of Job 994
Sparks, H. F. D. 10, xlvi, 564, 589, 591, 861, 862, Testament of Joseph 793
xlviii, lxii 913, 914, 953, 958, Testament of Judah 109,
Sperber, D. 303 1048, 1080, 1118, 1135, 934
Spicq, C. 339, 545 1148, 1185, 1187 Testament of Levi 910,
Squires, J. T. 974 Suggs, M. J. 807 1167
Staats, R. 943,950 Suhl, A. lxii, 1177, 1217 Testament of Naphtali
Stählin, G. 765, 1183 Sukenik, E. L. 324, 327, 1215
Stanley, D. M. 438 781, 1057 Testament of Reuben 351,
Stanton, G. N. 303, 318, Sulla 591 1167
334, 379 Sullivan, R. D. 1090 Testament of Solomon
Stauffer, E. 155, 174, 180, Swete, H. B. 744 230
228, 291, 384, 646, 681, Sylva, D. D. 61, 334 Testament of Simeon 347
687, 783, 1059, 1105, Symmachus 298,553,565, Thackeray, H. St J. 120,
1161 618,1150 135, 326, 362, 364, 576,
 INDEX 1271

716, 844, 970, 978, Tischendorf, C. 408 Veitch, W. 227

1111 Tissot, Y. 708, 752 Vergil 114, 247, 249, 432,
Theissen, G. lxii, lxxvii, Tödt, H. E. 385 483, 581, 848, 875,
lxxxi Torrey, C. C. 66, 67, 72, 1104, 1192, 1203, 1204,
Theocritus 244 75, 101, 120, 131, 140, 1213, 1214
Theodoret 809, 840 148, 150, 172, 196, 198, Vermes, G. 120, 193, 370,
Theodotion 236, 325, 414, 199,211,233,244,274, 383; see also NS
582, 1041 275,377,508,517,522, Vespasian 244, 740
Theodotus 324 523, 563, 691, 740 Vettenus 324
Theophanes 989 Trajan 527 Vettius Valens 171, 378,
Theophilus 65 Trebilco, P. R. 762, 775, 635, 660, 793, 906
Theophrastus 364, 618, 782, 786, 812, 903, 923, Victor of Rome 40
813,835, 1115 959, 988, 1026, 1186 Vielhauer, P. 132, 321,
Theudas 293, 294, 295 Trites, A. A. 280, 1018, 650, 651, lxxxvii, cvii
Thiede, C. P. 568 1053 Vincent, L. H. 1018
Thiele, W. 10, 89 Trocmé, É. 55, 80, lxii Vitellius 224
Thieme, G. 927 Troeltsch, E. 168 Viviano, B. T. 1053, 1066
Thiering, B. E. 159, cviii, Trompf, G. 1177, 1217, Vögeli, A. 280, 568, 775,
963, 976 1235, 1248 1144, 1158
Thomas, D. W. 113, 354 Trygaeus 184 Vögtle, A. 61, 663, 689
Thomas, Gospel of 328 Trypho 43 Völkel, Μ. 490
Thomas, J. xci, cviii Turner, C. H. 708, 803, Vööbus, A. 11, 159
Thomas of Harkel 12 1090 Vos, C. S. de 775
Thornton, C.-J. xxvii, Turner, N. 220, 266, 408, Voss, G. cvii
xxviii, xxx, xxxii, xliv, 754, 791, 792, 898, 910, Wainwright, A. W. 460,
767, 945 976, 1035, 1116, 1150, 752
Thornton, L. S. 92 1151, 1156 Waitz, H. 395,420
Thornton, T. C. G. 61, 334, TWAT (Theologisches Walaskay, P. W. 1, lxii
375 Wörterbuch zum Alten Walker, N. 303
Throckmorton, B. H. 216 Testament 844 Walker, W. Ο. 61, 752,
Thucydides 163, 197, 214, TWNT (Theologisches 1222
221, 233, 238, 255, 283, Wörterbuch zum Neuen Warnecke, Η. 1217, 1219,
294, 314, 326, 345, 447, Testament) 162, 182, 1220, 1221, 1225
465, 474, 503, 505, 548, 430, 604, 785, 962, Weber, R. 10
553, 563, 576, 580, 592, 963, 967, 975, 1053, Wedderburn, A. J. M. cix,
610,652, 672, 678, 682, 1057, 1106, 1114, 1148, 708
729, 730, 739, 742, 743, 1177 Wehnert, J. 1177, 1217
779, 794, 799, 808, 820, Tyson, J. B. 490, lxxix, Weigandt, P. 14, xxiii,
833, 873, 914, 935, 947, lxxx, lxxxi, cvii, cix, cx 775
948, 958, 983, 989, Ullendorf, E. 420 Weinstock, S. 107
1003, 1056, 1073, 1077, Ulpian 1113, 1131 Weiser, A. 490, 708, 775
1085, 1090, 1097, 1104, Unnik, W. C. van 61, 80, Weiss, B. 895
1105, 1118, 1124, 1125, 147, 233, 395, 420,490, Weiss, K. 557
1137, 1138, 1170, 1181, 608, xxxv, xlviii, li, lxii, Wendt, H. H. 1056
1184, 1193, 1194, 1199, lxix, 775, 790, 1031, Wensinck, A. J. 114, 435
1205, 1208, 1211, 1215, 1034, 1035, 1090 Westcott, B. F. 20, 21,
1228 Valentinus 46, lxvi 32
Thyen, H. 129 Vanhoye, A. cix Wetter, G. P. 438
Tiberius 382 Vaux, R. de 375 Wexler, P. 901,905
Tiberius Alexander 563 Vazakis, A. A. 173 White, H. J. 10
Tiede, D. L. 61 Vegetius 577, 1188 Wiefel, W. 1235
1272 INDEX

Wikenhauser, A. 81, 438, Witherup, R. D. 1031, 1170, 1172, 1183, 1189,

490, 703 1144 1192, 1196, 1202, 1206,
Wikgren, Α. Ί WM (Winer, G. B. and 1222
Wilcken, U. 867 Moulton, W. F.) 503, Xenophon of Ephesus 538,
Wilckens, U. 129, 133, 813, 1245 927, 990, 991
141, 152, 187, 207, 216, Wolter, Μ. 884 Yaure, L. 615
280, 438, 490, 521, 523, Wood, H. G. 439 Young, C. Μ. 81
622, 637, 641, cvii Woodhouse, W. J. 1177, Young, F. Μ. 334, cviii,
Wilcox, Μ. 25, 92, 97, 1192 963
280, 490, 491, 501, 654, Wordsworth, J. 10 Zahn, T. 15,22,23,24,28,
659, 669, xx, xxiii, cx, Workman, W. P. 1177, 583, 906,1221
775, 783 1188 Zeller, D. 491
Wilken, R. L. 549 WS (Winer, G. B. and Zeno 925
Wilkens, W. 395 Schmiedel, P. W.) 99, Zerwick, Μ. 703,739,754,
Wills, L. Μ. cix 813,958, 1204, 1245 762, 769, 771, 809, 817,
Wilson, R. McL. 395 WW (Wordsworth, J., 834, 836, 842, 870, 874,
Wilson, S. G. 61, 69, 70, White, H. J., Sparks, 891, 892, 896, 898,969,
75, 81, 84, 85, 136, 139, H. F. D. and Adams, A. 988, 990, 995, 996,
202, 213, 329, 383, 385, W.) 9, 10, 89, 595, 853, 1003, 1021, 1038, 1042,
391,444,491,515, 524, 864, 936; see also 1043, 1063, 1073, 1085,
535, 537, 547, 601, 650, White; Wordsworth. 1103, 1106, 1107, 1110,
657, 665, 667, lxxxiii, Wycherley, R. E. 191,824, 1135, 1137, 1139, 1141,
cvii, cx, 700, 703, 724, 827, 828, 837 1149, 1154, 1155, 1160,
726, 734,735, 738, 740, Xenophon 67, 98, 140, 1163, 1170, 1191, 1192,
742,745, 763, 825, 866, 156, 183, 195, 233, 1210, 1214, 1215, 1230,
868, 964, 966, 979, 265, 278, 297, 317, 327, 1245
1014, 1015, 1056, 1157, 355, 368,428,435, 449, Zeus 676, 677, 678, 679
1237, 1249, 1252 451,454, 457, 524, 530, Ziesler, J. A. cvii, 1053
Windisch, H. 438 581, 582, 584, 589, 648, Zimmermann, H. 273, 303
Winer, G. B. 99, 503 660,670, 671,687, 705, Zonaras 813
Wingren, G. 439 717, 755, 813, 828, 830, Zuckschwerdt, E. 568,709
Winter, B. 923, 1090 835, 907, 913, 925, 927, Zuntz, G. 709, 735, 943,
Winter, P. 280, 708, 1090, 931, 955, 968, 978, 953
1098 1002, 1005, 1027, 1073, Zwaan, J. de 172
Wiseman, J. 857, 860 1105, 1106, 1116, 1119, Zweck, D. 824
Witherington, B. xxiii 1130, 1137, 1167, 1168,