Paganism in the Roman Empire

Paganism in the
Roman Empire

RAMSAY MAcMULLEN

New Haven and London

Yale University Press
Copyright@ 1981 by Yale University. All rights reserved. This book may not be repro·
duced, in whole or in part, in any form (beyond thatcopyingpermiued by Sections 107
and 108 of the U.S. Copyright Law and except by reviewers (or the public press),
without wriuen permission from the publishers.
Designed by Nancy Ovedovitz and set in Baskerville type.
Printed in the United States of America

Ubrary of Congress Cataloging in Publication Data

MacMullen, Ramsay, 1928-
Paganism in the Roman Empire.

Bibliography: p.
Includes index.
I. Rome-Religion. I. Tide.
BL802.MS2 200' .9S7 80-54222
AACR2
ISBN 13: 978-0-300-02984-0
 To friends in libraries:
CML; REB, ]G, SS, GV, CHC,
and (in general) SML
 Contents

Illustrations ix
Preface Xl

Chapter 1: Perceptible 1
1. Finding order in chaos.
The shapeless confusion of paganism, partly product of its toler-
ance, partly of reverence for tradition (pp. 1-5). How can the con-
fusion be grasped in the mind? Ranking of the Top Fifteen (Jupiter/
Zeus included) from inscriptions. Does it show the preferences
only of the upper classes? Of the latter, the views more easily dis-
covered. Their many written forms, especially inscribed texts: cult
laws, sermons, hymns, and religious music; prose hymns (5-18).
2. Attracting crowds. 18
Cult theaters, cult dances and miming (18-25). Advertisement of
performances. Festivals and fairs. Parades. Pilgrims. Religious
debates and lectures, and displays through art (25-32). Testimonials,
written and symbolic, set up to divine power; dedicated gifts (32-
34).
3. Displays and accommodations at temples. 34
Sacred fish and birds. Gardens and parks. Picnic and dining facili-
ties. Sacrificial meats and their uses. Parasitic populations (M-42).
4. Routine staff and administration. 42
Temple scenes, daily liturgy, care of idols (42-46). Stratonicea's
nearby shrines and religious practices as a summary of all the
foregoing (46-48).

Chapter II: Debatable 49

1. Needs and answers. 49
The wants to which paganism responded: physical and mental
illness. Exorcism and magic (49-51). Prayers for harvests, wealth,
success; for immortality. Beliefs in afterlife, and the role of initia-
viii Contents

tory experiences (51-57). Religion as a policeman. Divine ven-

geance; divine support for the community (57-59). Means of com-
munication with the gods: via idols, dreams, and oracles (59-62).
2. The vitality of paganism. 62
Denial of the gods, atheists-contradicted by proofs of intense
piety. How can religiosity be measured? Differences by period? By
class? An individual matter? (62-67). Problems of evidence: Is that
from the upper classes different in quality? Religious beliefs seen
in various forms of written evidence. The dominance of the ob-
vious: Homer and the like set against Gnosticism and the like. The
false dominance of the more intellectual, contrasted with super-
stition and irrationalism (67-73).
3. How the divine world was envisioned. 73
Thought-out conceptions of the divine: divine goodness, remote-
ness, etc., in literary sources-not representative of all paganism
(73-77). Mass beliefs "purified" through interpretation; through
postulating degrees of divinity in different layers, of gods and
"demons"; the pyramid of the divine world (77-83). The top;
monotheism, solar and other (83-89). One god under many names.
Syncretism (90-94).
4. Conversion. 94
How to tell a "real'' god; testing of claims of divinity. Evangelizing
(94-99). Control and unity of the message (99-102). Control 'by
emperor, township, or priesthood ( 102-107). Changing fortunes of
shrines, in which private individuals (or families) are the chief
factor (including officials, but not as such) (107-112).
5. The dynamic cults. 112
Beliefs spreading over long distances. Patterns of loyalty, by ethnic
origin, sex, condition-especially among "Oriental" cults (112-
118). Mithraism (118-126). Signs of debility in later paganism?
Rise of non Greco-Roman cults (126-130).

Epilogue: The Manner of Death of Paganism. 131

Abbreviations 139
Notes 141
Bibliography of Secondary Works Cited 207
Index 235
 Illustrations

I. Cult Relief from Ariccia 22

II. Wallpainting from Dura 80
III. Wallpainting from Dura (restored) 81
IV. Marble Statuary from Constanza (Ancient Tomis) 120
V. A Snake God from the Constanza Hoard 121
 Preface

It was a proper melting pot. If we imagined the British Empire of a hundred

years ago all in one piece, all of its parts touching each other, so that one
could travel (not very fast, of course, in the days before the automobile)
from Rangoon to Belfast without the interposition of any ocean, and if we
could thus sense as one whole an almost limitless diversity of tongues,
cults, traditions, and levels of education, then the true nature of the Medi-
terranean world in the two centuries of this study would strike our minds.
Then, too, we would acknowledge the inevitability of interchanges to a
degree and with historical significance .not found earlier. Few corners of
that ancient world can have lacked their own quite distinct way of life, as
few can have felt no challenge at all to that way of life from others forced on
their attention by their conquerors, or by their neighbors no longer sealed
off in enmity, or by soldiers, traders, government clerks, or inquisitive
tourists and their slaves who knew shorthand.
The tidy mind, abhorring so confused a scene, divides it for better com-
prehension into several social strata, provinces, language areas, tax dis-
tricts, and so forth. The Romans themselves treated what they had con-
quered in this manner; in this manner its historians treat it today. But
something is thereby lost, as if one tried to comprehend a good rich Irish
stew by eating in succession, no matter with what speed, each one of its
ordinary ingredients.
In this book I try to repair that loss. It is on the system as a whole and
interchange within it that I focus. The constituent parts, the individual
cults and their derivation and internal nature, are not my concern.
I begin by surveying the variety so as to subordinate the lesser to the more
important cults. From there I move on to the question, Who cared, and how
much, about religion? From that to the categories of written evidence that
can be used in answer- themselves so many forms of advertisement for one
or another cult. From that to the concourse attracted by cults in their daily
routine or at special festivals. And after this group of topics, all having to
do with the outer face of religion, I address a second group, dealing first
with the inner vitality of paganism as a whole and its roots in urgent
human needs, and next with the vitality of certain of its parts and the extent

xi
xii Preface

to which one can discern a dynamic in paganism across the two hundred
years chosen for study. It is, after all, change and movement that historians
are expected to talk about.
In handling such topics, naturally I aim at broad, clear statements. They
are not easy to frame. I have found myself continually having to distin-
guish, for accuracy's sake, between the views and practices common in one
kind of source but not in all or in one class of person but not in some other.
The need for accuracy sets me at odds with my own aim of easy generaliza-
tion. It also sets me at odds with many of my predecessors, including "Franz
Cumont, whose unique authority in the field is universally acknowledged." •
So far as possible, I have left argument to the endnotes; but some sense of
controversy inevitably appears in, and complicates, the text of my discus-
sion.
In the religious history of the Roman Empire, the chief event was of
course the triumph of Christianity. It would be good if the pages that fol-
low helped to explain how the one religion could succeed universally
across the endless variety of cults surrounding it. It seems fair to infer from
that universality, however, some set of broadly shared religious ideas and
practices within the boundaries of which Christianity could make its con-
verts in Gaul as in Syria; or to put it in other words, the thing so arrogantly
called paganism, being in fact all the many hundreds of the Empire's reli-
gions save one, must really have shared certain widespread characteristics.
It remains to determine what they were within the spans of time and space
that saw the triumph of Christianity. The area may be roughly marked out
within the two centuries before Nicaea (or the two centuries of the Apolo-
gists, Quadratus up to certain works of Athanasius, let us say), and within
the oblong of territory having as its corners Alexandria and Origen, Car-
thage and Tertullian, Lyon and lrenaeus, and Pontus and Gregory the
Wonder-worker (not an Apologist, granted, but a great missionary and
saint). These are the names and reasons which set the boundaries around
my curiosity.
The epilogue is no more than that. Had I been competent to handle
Judaeo-Christian texts, early Church history, art and epigraphy, then I
might have attempted something more. In explaining the triumph of
Christianity, however, the first need that I see is to sponge out of the picture
of paganism those false outlines and colors that have been painted in over

•Marcel Simon (himself no mean authority), speaking in praise of Cumont's Religions

orientales, in Religious Studies 9 (1973), 392.
 Preface xiii

the course of the present century. Thereafter, other hands may complete the
picture by adding the Church. But not too fast.

July 1980

The courtesy of the publishers allows me, for the second printing, to add a
half-dozen supporting references in my notes, to correct various typo-
graphical errors and one of fact, and to make my thought dearer on several
pages. I may also mention an essay I have written (appearing in Vigiliae
Christianae 1982) that shows, more fully than pp. 88f. and 94f., how I
would explain conversion to Christianity in the period before Constantine.

September 1982
 NORICUM

"•' 4\ 'liot> • Vlrunum

~'\,'to olmono
~((.,~ • Aqullolo
• Milan

0 Paut.ll• •
scaptop~~ra

• Stobl
•

0
• Lambaesls
•Apulum

• Comana

• c•••• ,••
• Coman•

•Eme18

Labraunda

~
•Ttberlaa .,.
~
•C••••r•• ~
•o•r••• A~
Paganism in the Roman Empire
 I
Perceptible

I. FINDING ORDER IN CHAOS

To move about in the Roman Empire at all, or to make the hastiest survey
of its religious variety, brings home the pullulation of beliefs. The standard
Roman city, if such a thing could have been discovered within its bound-
aries, would need room for temples to the Capitoline Triad (Jupiter, Juno,
and Minerva), plus Mercury, Isis and Sarapis, Apollo, Liber Pater, Hercu-
les, Mars, Venus, Vulcan, and Ceres. So Vitruvius recommends, the archi-
tect writing about 27 B.c. and speaking (On Architecture 1.7.1) of Italy in
general. But he indicates cities with gymnasia, hence he is including the
more Hellenized southern parts of the peninsula. At the other extreme, in
the secluded little Val di Non among the Alpine foothills, an even more
generous pantheon appears in dedications set up to Jupiter, chief, and
Minerva, Apollo, Saturn, Mercury, Mars, Venus, Diana, Luna, Hercule.s,
Mithra, an~ Isis, with only the one true native, Ducavavius. 1 Inscriptions
from Philippi in Macedonia give us a Greco-Roman mix of Jupiter or Zeus,
Juno or Hera, Minerva or Athena, the properly Italian Vertumnus and the
properly regional Souregethes, Myndrytus or Etepancus, to the number of
two dozen or more deities;2 just as the coins of Nicomedia, farther east,
advertise more than forty that receive worship in that city, Sabazius ancil Isis
and Apis and Sarapis, Helios, Cybele and Men (all extraneous), plus dozens
of more familiar Greek names. 3 Or we can read the satirist Lucian, in mid-
second century imagining a congress in heaven well attended by Bendis
(from Thrace), Anubis (Egypt), Mithras (Persia), and Attis and Men from
different parts of Asia Minor, not to mention another half-dozen more often
seen on Olympus;• and Eusebius a few generations later reels off the wor-
ship in different provinces of Horus, Isis and Osiris, or Melkart and
Ousoros, or Dusaris and Obdos, or Zalmoxis, or Mopsos, depending on
whether you found yourself in Arabia, Egypt, or some other place. 5
That last testimony points to an obvious explanation for so many gods:
not that the Thracians and the rest, to say nothing of Greeks and Romans,

1
2 Perceptible

were especially given to the picturing of new unseen powers about them,
but that the political unit here under survey had been assembled out of
many peoples, each of which had its own system of faith. In the course of
time, conflict had brought into being successively fewer but larger states
that drew strength from the absorption of divine, as of human, resources.
To rehearse that whole story would be to rehearse the whole of ancient his-
tory. The process was now over. Rome's Empire under our gaze was com-
plete, and completely tolerant, in heaven as on earth.
Perhaps not quite completely: Jews off and on, Christians off and on,
Druids for good and all, fell under ban, in the first century of the era. So did
human sacrifice. In the second century, laws forbade mutilation, even vol-
untary and for divine service. But humanitarian views were the cause, not
bigotry. For laws against soothsayers, the cause was fear of popular unrest,
not any hostility to preaching in itself. 6 Even a Christian could declare that
everyone's ancestral beliefs were permitted-ta patria.7
That was only half the story. The other half was a widespread feeling that
to slight the gods, plural, was wrong. I leave aside the active persecution of
Jews and Christians, except to note that monotheists rated as atheists: to
have one's own god counted for nothing if one denied everybody else's. 8 I
leave aside also, for later discussion, the existence of real atheism. But there
was very little doubt in people's minds that the religious practices of one
generation should be cherished without change by the next, whether within
one's own community or another's. To be pious in any sense, to be respect-
able and decent, required the perpetuation of cult, even if one's judges
themselves worshiped quite other gods. So the Jews could be excused be-
cause they were at least loyal to their own inherited error, as one of them can
declare to a pagan audience. He declares, of course, only a reason that he is
sure will be acceptable. It duly appears unprompted in the mouths of
pagans. 9 Entangled in exactly the same presuppositions, Christians were
continually on the defensive as deserters from what was seen to be their true
spiritual home, namely Judaism. They replied by appealing to considera-
tions that might weigh against ancestral piety, for example, "philosophy"
and "reason" against "superstition" and ta patria. Whatever their answer,
the debate demonstrated polemically one of the bedrock beliefs of both the
Greek- and Latin-speaking halves of the Empire in the second century and
on into the fourth, a belief presented, moreover, out in the public arena and
intended to win approval from a wide audience. 10 As Eusebius explains it,
pagans "have thoroughly persuaded themselves that they act rightly in
honoring the deities, and that we are guilty of the greatest impiety in
making no account of powers so manifest and so beneficent, but directly
break the laws, which require everyone to reverence ancestral custom, ta
 1. Finding Order in Chaos J

patria, and not to disturb what should be inviolable, but to walk orderly in
following the religion of our forefathers, and not to be meddlesome through
love of innovation." 11 Less often because less stimulated by opposition,
pagans thinking only of other pagans affirm exactly similar views. "Such is
the chief fruit of piety," says Porphyry, "to hono~ the divinity according to
one's ancestral custom." 12
Porphyry indicates one reason anyway for saying what he does: the
impious man wrongs his own forebears as well as the deity. 1 ~ Human at-
tachment to parents or grandparents inspires the perpetuation of whatever
they held dear. Plutarch's father reveals a little more, through the warmth
with which he asserts, "The ancestral, ancient faith suffices in itself, than
which no demonstration can be declared or discovered more palpable ... ";
and in Plutarch's own generation, among his circle of friends, various
members express very similar and very lovingly espoused opinions about
religion. 14 All these passages do no more than testify, however, to what
anyone would assume to have existed throughout the Roman world, that
is, a strong emotional element in religious conservatism. What is more
particular and useful is the hint in the quotation from Eusebius above, that
innovation in itself was not only painful but bad. "New'' was a term of dis-
approval, used in that sense both in and beyond the debates between pagans
and Christians; 15 "old" was good. In the apologies written by Christians,
the reader is struck by the emphasis, through position near the front of the
works or through length and frequency of discussion, accorded to proving
the religion of the jews, Moses, and the Pentateuch,older than Hellenism,
Homer, and the Iliad, or to proving the priority of Jewish ethical positions
over Platonic. The writers agreed, then, with Hesiod's mode of reasoning,
which Porphyry appeals to: "ancient law is best law." "With you pagans,
too," says Tertullian, "it is almost a religion to demand belief on the basis
of age." We could add many other statements to the same effect. 16 All that
needs to be marked out in them is the common ground occupied by authors
of divergent origins or aims and their audiences, providing a ready respect
for all cults. Truth as well as venerability had been established through the
number of persons approving, whose approval was in turn implied by the
long survival of an idea or custom. The result was seen as a sort of com-
munity decision formed across time, slowly fixed, and therefore even more
slowly to be set aside. On that view, more than any other, rested toleration
of beliefs boasting a long past.
"Ancestral" is a term ordinarily applied to deities in inscriptions in one
of two ways. In the first, it salutes "him who watches over us here at home,"
and other adjectives may be added to make the feeling clearer: "guard-
ian," "patron." 17 The individual or the community asserts some special
4 Perceptible

claim and devotion. For whatever reason, such rich language is far easier to
find in epigraphy of the eastern provinces than of the western. The second
way in which the word "ancestral" is most often attached to a deity is for
remembrance's sake and from a sense of being a stranger abroad. A letter of
the Tyrians settled in Puteoli, addressed to their original home, uses the
term in appealing for financial help to maintain traditional cults. 18 Anum-
ber of other texts from Italy but more often from less central regions record
the lonely and fervent thoughts of strangers turned back to their homes.
The particularly possessive assertion of loyalty to local gods 'can be
noticed in still a third way, though much less frequently: on the occasion of
prayer for the imperial family, in the form "to the ancestral god Such-and-
Such for the well-being of our emperor." 19 The latterfor his part repaid the
courtesy, when he visited a city, by offering some cult object in precious
metal, a statuette or the like, or simply a prayer to the god of that place. So
Hadrian, being in the vicinity, climbed to the sanctuary of Zeus Casius;
again, made an offering to Aglibol and Malakbel of Palmyra. Caracalla
offered a sacrifice to Asclepius at Pergamon, an act advertised on coins; and
Diocletian dedicated an offering to Minerva at Ilium, to Zeus and Leto at
Didyma, to the sun god at Comum, to Mithra at Carnuntum, and to Bele-
n us at Aquileia.2o Evenhanded respect shown by the emperors to the famil-
iar pantheon was shown also to divine figures of only regional fame.
Regional or topical names: Tars giving his name to Tarsus, Kakasbos of
Telmessos, Vintius on the northern Rhone. 21 Of such, there is no end and
no story. A step up in prominence are those names of localities (really at-
testing purely local gods) that are tacked on to the better known, like Zeus
Thamaneitanos in Lebanon, Artemis of Perge, Mother of Zizima-not the
great Mother Cybele. 22 In the eastern provinces it is Zeus around whom, or
hyphenated with whom, the richest ferment of devotion gathers-there
exist some dozens of local versions in Caria alone, for example-and in the
western provinces, Mars. 2 ~ These transformations, substitutions, elisions,
conjunctions, impostures, whatever they may be called and of whatever
origin or end, produced an approach to a defined pantheon for the whole
Empire much closer in titulature than in reality. Labels had been attached
to the outside of hundreds of separate worships that were in fact very little
changed from their form as it had been before Greek or Roman conquest.
We will return to the matter later.
For the moment it is enough to recall those lists of independent and by no
means entirely Greco-Roman gods and goddesses cited above, from writers
like Vitruvius, Lucian, or Eusebius; to which we can add more, from
Minucius Felix-"among the Eleusinians, Ceres; among the Phrygians,
the Mother; among the Epidaurians, Aesculapius; among the Chaldaeans,
 1. Finding Orckr in Chaos

Ba'al; Astarte, for the Syrians; Diana, for the Taurians; among the Gauls,
Mercury" -or from his contemporary and fellow-African, Tertullian. Ter-
tullian has a wider range, taking in Astarte for the Syrians but also Dusares
for the Arabians, Belenus in Noricum, Caelestis in Africa.u A little later,
the lawyer Ulpian offers quite a different selection of deities recognized by
the government as entitled to re.ceive bequests. Of the eight, five are to be
found in western Asia Minor. 25 He has apparently picked out the more
familiar names from some much longer list; or perhaps only more familiar
ones stood much chance of gaining grants of money; but in any case he
excludes the exotic ones that Apologists fortheir own faith might choose,
for polemical reasons, and he excludes the conservative Italian and long-
naturalized onesthat Vitruvius must think of when reviewing the likely
needs of Roman cities.
The purpose pursued thus far, to find some grip, some orderly way by
which the mind can fasten on the shapeless profusion of polytheism, can be
advanced a little further through the epigraphic evidence. We have at our
disposal a quarter of a million published inscriptions. A sizable share have
something to do with religion. Those from the Latin-speaking parts of our
field of survey are well indexed (not the Greek inscriptions) so that some
impression may be easily gained of what deities appear most often. The bar
graph below shows the relative popularity of the fourteen most mentioned,
so far as inscriptions can be trusted, and excluding the most popular of all
in each district: that is, Jupiter everywhere except Africa. In Africa, Saturn
is added to Jupiter to make a more trustworthy total. 26 In the four districts,
the number of all inscriptions of all sorts naturally differs; the customs
dictating how commonly religion received mention on stone at all may
have differed, too; so for each of the four districts the total of the numerically
dominant god has been determined and divided into the number of men-
tions of the other fourteen gods, to yield a percentage. The percentage is
what determines the length of the bar for the fourteen.
The results must be discussed again in the second chapter, but they need
some qualification immediately. First, a goddess like Nemesis, much fa-
vored in the eastern provinces, has little following in the western and by
average therefore fails to make tbe select company of those tabulated; 27
similarly Caelestis in Africa and Belenus in Venetia, though both appear in
Tertullian's list. Next, allowance must be made for the gods not being what
they appear, as, for instance, Apollo sometimes hiding a Gallic god of
oracles, Hercules sometimes hiding a Phoenician or Punic god, and so
forth. This produces one distortion in particular: Mercury served as the
chief translation into Latin for the dominant Celtic deity Teutates, 28 and
one cannot ordinarily be sure when that translation has been made or not.
 FREQUENCY OF MENTIONS OF DEITIES RELATIVE TO JUPITER
IN INDICES TO THE CORPUS OF LATVII INSCRIPTIONS

AE~APJJS

APOLLO

DIANA

FORTIMA

HERCULES
~
..."" """ """'
"' '//// """

.....
ISIS-SARAPtS

m:mmmmJ I : ~-YOL.3=N:lRTH-CENTRAL
PR<MNCES
- lt<CJ.L,VQS. 5,9,f0,81o4(EXCLIXliNG
LIBER OSTIA 8 ROME) • MOST Of ITALY

- :II• <;!!.VOL 8 • NORTH AfRICA

MARS - llr' f:!:b:VOLS.I2813' GAl.l.8GERMANY

CYBELE

loiTMRA-SOL

SILVANUS

VENUS

012 3 4 s 6 7 8 910 n 12 13 14 15 16 17 18 19 20

NOTE: See above, p. 5, for explanation of bar length.

 1. Finding Order in Chaos 7

Finally, certain localities rich in evidence beyond their actual size of popu-
lation are disproportionately represented, though the gods they favor may
not be favored more widely. This inflates the importance of Jupiter Doli-
chenus especially, in Africa, through the disproportionate number of texts
from the military camp and nearby civilian city of Lambaesis. The frontier
zone has the same effect again on Dolichenus in the total of texts from Gaul
and Germany. And exclusion of Ostia has the reverse effect of deflating cer-
tain numbers-of eastern deities like Isis, Cybele, and Mithra-Sol, but also
of Venus and Silvanus-in the total from ltaly. 29 For what it may be worth,
private worship in Pompeii favored Fortuna first, then Vesta, and Bacchus·
Liber, Venus, Jupiter, and Mercury in that order. All figure in the group of
fourteen-plus-Jupiter used for the bar graph, with the exception of Vesta,
who naturally bulks larger than life in domestic worships. so The sizable
presence of Jupiter, even here, and of course overwhelmingly in public
inscriptions, contradicts the view that worship of the supreme deity was
mere empty ceremony.
Because of the unlucky way they have been published, to say nothing of
their far smaller numbers and arbitrary clustering in particular sites, Greek
inscriptions cannot be used to provide even a general impression of the
hierarchy of worships in the east. But, in a standard collection for Asia
Minor, if one sets aside Zeus, invoked two and a half times as often as any
other god, the rest offer no surprises: Apollo, Athena, Dionysus ( =Liber),
Artemis (=Diana), Hera (=Juno), Aphrodite (=Venus), Asdepius, Tyche
(=Fortuna), Hercules, and the Great Mother (=Cybele), in that order. 31
Overlap with the picture in the north and west is striking, but good Greek
gods and goddesses, Athena and Hera, more than take the place of the
oriental Mithra, Jupiter Do lichen us, Isis, and the rest. Even Cybele enjoys
little prominence.
The many gods of the second and third centuries can be arranged in some
sort of ranking, then, with ten or fifteen rather easily identifiable ones ,atop.a
mass beyond ordering. There is a clear summit to this pyramid, Zeus!:
Jupiter; there are much less clear planes of cleavage in the pyramid accord·
ing to regions rather uncertainly defined. But lastly, the gods may bear·
ranged according to the social position of their worshipers.
That the human population of the Empire formed a pyramid like the
divine, anyone would guess. Perhaps its strange proportions, however,
need to be recalled, with the senators, governors, legionary commanders,
and in short almost every single name to be found in the ancient historians
lying at the absolute tip, that is, within the upper I percent, while a similar
figure circumscribes the bare names (lacking anything else we would like to
know about them) that appear in municipal senates; so the extreme steep·
8 Perceptible

ness of that upward slope from the masses to the local leadership is the
prime fact to be noted in society of the times. 32 But it is important further to
determine if the social fact had anything to do with religious feelings.
Where digging unearths the objects offered in sacrifice, as in some of the
sanctuaries west or south of Carthage or in eastern Gaul, they vary a lot in
cost of manufacture, indicating that rich and poor alike worshiped in the
same place and the same way. And in eastern sanctuaries great crowds,
plainly of all conditions, gathered in holy days to perform the same rites in
honor of the same divinity. Against such evidence, on the other hand, there
are to be set, first, more than ample testimony to a gulf separating rich from
poor, in contexts having nothing to do with religion but suggesting broad
habits of thought; and second, explicit statements that emphasize the gulf.
These deserve consideration.
What is likely first to attract notice are the blunt words of contempt and
disapproval with which the lettered aristocracy, in talking about religious
views, belabor the simple, unthinking, ordinary folk, the unlearned.3 3It is
a pagan, but it could almost equally well be a Christian, who deplores any
sort of theological speculation by the "crude untaught raw yokels, to whom
it is not even granted to understand citizen affairs, let alone to discuss the
divine," 34 just as it is a pagan who tries to place Christianity among the
"goatherds and shepherds, led astray by yokel tricks," "the stupid and ser-
vile," "wool-workers, cobblers, laundry-workers, the most illiterate and
bucolic yokels," only to be answered by a Christian's description of detrac-
tors as "the scum of the marketplace, a godless chorus of superstition." 35
Each polemist assumes that his words will strike a sympathetic chord in his
audience-readers in both camps, that is, and the more than literate. A
public speaker on the other hand must go about it differently. He must not
think to offer profound poems on religious subjects "to the crowd and the
populace, for they will be thought pretty unpersuasive and ridiculous by
the masses." 36
We may note in passing the categories that are held up for derision: in
cities, petty artisans and people of no fixed place of business, the unproper-
tied; in the countryside, peasants; also, everywhere, slaves, children, and
women. 37 What we do not find in anyone is a sense of distance from some
particular ethnic enclave, nor even from life in the backcountry. The line
of cleavage that counts is a matter not of place, then, but of cultivation, or,
we may say, of culture. That line would correspond very considerably with
one of class; but among those managing to attain a higher education, there
were always some not born to it. Whatever their origin, the educated were
exposed to ideas beyond the horizon of the unschooled. Thus they some-
 1. Finding Order in Chaos 9

times grew away from the most broadly held beliefs. It is common to find
specific distinction made between a theology or point of creed proper for
the masses and another reserved for the learned, for initiates, or for believers
specially capable of deeper understanding.ss
Anyone in the Empire who ever picked up a book must have been aware
of the two kinds of know ledge, inner and outer, in the philosophic schools,
mystery religions, or other pagan or Judaeo-Christian religious circles. It
might be treated lightly and with common sense, as a fact of life calling for
a simpler and selective presentation of doctrine to the unlearned (so Cle-
ment), or it might be treated heavily and with dark exaggeration, to arouse
awe of secrets about to be revealed (so in magical papyri). In all its forms,
however, it reminds us of something important for the forming of a just
acquaintance with ancient polytheism: the testimony that reaches us on
paper-papyrus, parchment-comes from witnesses of a particular kind,
not only a tiny minority within the whole population but one whose mem-
bers generally felt themselves to be different from the majority, different in
culture, sometimes aggressively so. When they speak to us about religion as
they observed it in the rest of their world, their report must be heard criti-
cally. When they speak about their own ideas and feelings, though they are
much more to be trusted, their report cannot be used as the basis for any
generalization about the whole. The point might seem obvious but is some-
times forgotten-with the less reason, given the varieties of countervailing
evidence.
In truth, even if we define "literary" evidence very broadly to include
fragments of philosophical treatises or collections of laws, it holds disap-
pointingly little on our subject: the essay on the Syrian goddess ascribed to
Lucian, the dosing chapters of Apuleius's novel, the spiritual diaries of
Marcus Aurelius (but that is really philosophy), and Aelius Aristides (and
that is hypochondria), all amounting to a couple of hundred pages quite ill
suited to the demands of our curiosity. However enlarged by oddments
from the satirists or historians, the dark theologists and seers (to whom we
must return in the next chapter), this whole body of writings suggests
(wrongly, as will appear) that the inhabitants of the Empire were simply
not much interested in the variety of worships around them. Had that been
true, of course the vitality of interchange would have been correspondingly
low, and our subject would bear a very different character.
When one wanted to refer in broad terms to the library of available infor-
mation about religion, the headings first offered were "poets and philoso-
phers." The pair are a fixture in the Apologists. 59 By "poets" were meant
Homer above all but Hesiod also, and others. Their antiquity in itself sane-
10 Perceptible

tified them. Moreover, all poesy was divinely inspired, according to its
practitioners; 40 and the great ones, at least, were so widely known in the
Greek world that reference to them would be understood among all the
literate. 41 To be sure, they had no equivalent in the Latin west. Neither did
the philosophers. That fact did not deter Latin writers from using both
members of the pair as touchstones in debate and verification, for they
joined in generally intelligible abbreviation the two main aspects of reli-
gious heritage: myth and theology, the irrational and the rational, or how-
ever it might be phrased. Philosophy interpreted the poets. The interpreta-
tions by Plato and other heads and followers of all the schools circulated
not only in works still surviving but in many others unfortunately lost: by
the Epicureans Philodemus and Diogenianus, by the Stoics Cornutus and
his contemporaries the Elder Seneca and Heraclitus, writing treatises on
theology; by the Cynic Oenomaus in the second century who exposed reli-
gious frauds, as the Elder Seneca had written on superstition; and by a
number of Platonists, notably Numenius and Porphyry in the second and
third centuries, the latter in works some of which survive, some of which
are known only through quotation. 42 There were special treatises that
focused on a single cult, 43 others dealing with particular iiturgies and rules
of sacrifice. 44 Anyone too lazy to refer to them or in a center too small to
supply them might turn to more or less standard handbooks on religion, by
Apollodorus in Greek and Varro in Latin. 45
The object of this rapid survey is the measurement of educated interest in
the subject of our study. True, the evidence that can be brought to bear is
very little for the span of two centuries in question. It weighs little against
the popularity of the sensational-against "the tens of thousands" of essays
on oracles that Eusebius mentions, with whatever exaggeration, 46 or against
those scenes of worship, marvelous signs, or picturesque rites generally
called for in novels and generally supplied. 47 But more intellectual works
could still find a public in the west as in the east.
That public was certainly quite small. On the other hand, we can go on
to show that it shared its enthusiasms in unexpected ways with a consider-
ably wider community. A traveler to Argos in the 170s found a guide there
ready to unfold local myths in poetic version; a novelist or biographer, it is
not clear which, has his hero at a temple in Elis encounter the author of a
just-finished disquisition on Zeus, which he plans to recite in public the
next day; most extraordinary, a citizen from an obscure place inland in Asia
Minor obtains permission from the authorities to inscribe on a wall near
the city center, in columns six feet high and across a distance of well over
fifty yards, it cannot be said how much more, a collection of mainly philo-
 1. Finding OrdeT in Chaos 11

sophie, partly theological texts. 48 The inscription would strike the eye of
small farmers and peddlers as they came to market. Its author sought to
raise their thoughts to a higher plane. With what success, who can say? The
substance of the collection of writings is elevated, the language at points
obscure, but Diogenes the expositor of the whole must know his fellow citi •
zens better than we and thinks to address them all. He makes clear that he is
also in erudite correspondence with students like himself in Athens, Rhodes,
Thebes, and other cities. He is not alone, then, in his passionate interest in
theosophy. And before he is dismissed as unique, his effort to share it with
the general public should be put side by side with something similar at
Phlius, where, in the third century, an observer could study "the so-called
'rites of the Great Mother.' There is a portico in it [the city] and in the por·
tico is depicted to this day an outline of all the doctrines expounded. Many
things are depicted in that portico that Plutarch discussed, too, in his books
On Empedocles." 49
In addition to general public libraries, temple libraries are attested in
both the Greek and Latin halves of the Empire. so There is no special reason
why they should be mentioned at all, so their presence can perhaps be
imagined at most large centers. Public lectures presented in temples also
crop up casually, at Aigeai and elsewhere (we will return to them); and
temples drew people interested in discussion of philosophy and religion-
the man Plutarch met at Delphi, perpetual note-taker and intellectual
tourist, lately returned from Egypt and points east, or the other stranger
back from Britain, all three joining Ammonius the philosopher in the
temple of Apollo, for a good talk.s 1 Such travelers will reappear later.
Temples had resident experts, their theologues, ready to join the discus·
sion.s2 An epitaph from early in our period or a little before praises "the
man. of many books, guardian of all research, who picks out the ancient
page of the poets; lover of wisdom, Gorgos mighty in mind, the servant of
the tripod in Apollonian Claros." A wonder of learning! ss But no doubt
Claros was atypical: a much bigger and better equipped shrine than most,
and as an oracle, for the preparation of responses, specially in need of some·
one like Gorgos.
Great pains might be expended on determining the things most suitably
to be said, offered, or done before a given temple. When a new cult center
was established, all this had to be made clear at the very outset and posted
where strangers could not help but see the regulations they must follow.
Regulations were needed for the worship of a man's son, revealed to be a
demigod or hero; they must be specified for worship of a little-known local
Jupiter-the-Free in an Italian village5• or ad hoc for relief of particular situ·
12 Perceptible

ations.55 Some required re-publication through the fresh cutting of an

inscription blurred by age. 56 Cult associations bound their members to
detailed usages, 57 incidentally offering us a glimpse of what not to do: you
were not to vomit up your wine or carve your initials on the walls of the
temple's banquet rooms. 58 But entrants in general, at whatever sanctuary,
were likely to find notice of what was expected: you were to present yourself
suitably clothed, uncontaminated by recent childbirth, by sexual inter-
course with woman or dog, by sight or touch of a corpse, or by consuming
pork or garlic or milk; 59 you were to offer specified animals in sacrifice to
specified deities, or not animals but wine only, or incense only; and at par-
ticular shrines you were to respect local conditions, not lighting a fire near
the walls, not poaching sacred fish or cutting trees in the sacred grove.
Directions might be given on the whole range of daily service, a subject that
will recur. And the text might be very long-190 lines, at the shrine of
Demeter and Kore in Andania-or very short. At Athens, "by oracle from
Hygieia and Asclepius: the celestial serpent of the gods [directs the setting
up of?] these statues, where should be offered sacrifices unmixed with wine,
on the 5th and lOth of the month at noon. Send a good taste to the good
gods! "&o
"By oracle" was a common way of finding out what the gods wanted. The
learned Gorgos would intervene for that purpose, after consulting the
proper works of reference. They were, as we have seen, readily available to
say what was proper for every sort of divine purpose and why.& 1 Neopy-
thagoreans and Neoplatonists were full of information about the effects of
diet or sexual regimen on worship, or about the reasons for offering sacri-
fices on one rather than on some other day of the year. Research in their
treatises and the like would lie behind the moment when "the god replied
oracularly." 62 Occasionally a god was alleged to have spoken through
some person he had chosen as his voice, his commands being duly recorded
on stone. 63 Sometimes, again, ancestral customs might be invoked to sanc-
tify the rules of worship because, simply by virtue of their great age, they
could only have been learned from the gods themselves. 64
Whatever the means by which they were communicated, cult practices
made up a body of revealed truth rich in content and stimulating to in-
formed piety. Anyone who wanted could read the rules prominently on
display. At Oenoanda, home of the eccentric Diogenes, a city wall bore the
text of one of the answers given by Apollo of Claros to a certain Theophilus
who, in the early third century, inquired about the nature of the divinity.
He was told: "Born from himself, innately wise, without mother, unshake-
able, abiding no name but many-named, living in fire, that is god. But we
 1. Finding Order in Chaos lJ

are particles of god, his messengers. Whatever persons ask god what he is,
he answers, 'Looking upon him, the Aether, the All-seeing god, pray facing
east in the morning.'" At Claros, this too on stone: "When someone asked
Apollo whether the soul remained after death or was dissolved, he answered,
'The soul, so long as it is subject to its bonds with the destructible body,
while being immune to feelings, resembles the pains of that [the body]; but
when it finds freedom after the mortal body dies, it is borne entire to the
aether, being then forever ageless, and abides entirely untroubled; and this
the First-born Divine Providence enjoined.'" Direct glimpses of celestial
verities could be had not only through inscriptions scattered about Miletus
but collected for easier reference in the Oracle Archive in neighboring
Didyma. Before the mid-third century a compilation circulated. Lactantius
used it and, much more fully, the pagan theologian Porphyry, in the
writing of his Philosophy Drawn from Oracles.&s
Before leaving the subject of religious practices as they were described for
contemporaries in epigraphic form, note should be taken of the wide
spread of so-called sacred laws given to cult centers. Certainly more often
drawn (and better edited in modern dress) from the Greek-speaking prov·
inces, they are known also from Palmyra, Africa, Italy, and Dalmatia. 66 A
dedicant C. Domitius Valens puts up a long notice in Salonae to fix the
right procedures for sacrifice to Jupiter, and "the other regulations for this
altar shall be the same as those assigned to the altar of Diana on the Aven~
tine [in Rome]. By these regulations and within these points, as I have de·
dared, this altar I give, assign and dedicate to you, Jupiter best and greatest,
that you may willingly favor me and my colleagues, municipal senators,
citizens and residents of the Julian Martian colony of Salonae, and our
wives and children.''
The text serves as reminder of the particular character borne by the re·
gions within Rome's reach: far more susceptible to cultural challenge than
the eastern half of the Empire, where various ways of life were so much
better able to maintain their differences; far more directly touched by a
civilization, the Roman, that actively favored uniformity and made use of it
in the planting of its outposts. From even a cursory reading of Caesarian
colonial charters, it is easy to sense that Roman inclination and to infer
some corresponding pressure from the center to establish traditional Roman
worships in traditional Roman forms wherever settlements were made. If
we have only the one lex sacra of Diana used for such a purpose, we can at
least see it taken as a model in the establishing of practices at two other
points, in Italy and Gaul, and with the key phrase, "this temple shall have
the same laws as Diana at Rome on the Aventine," all reduced to abbre-
14 Perceptible

viations-obviously from being routinely repeated in records now lost

to us.
On the larger question, what to make of the chance survival of evidence
on stone and the distortions to be feared in the resulting picture, there can
of course be no last word. As a working principle, better too much caution
than too little. Yet the single inscriptions of Valens, of Diogenes, and of
Gorgos each invite generalization outward to whole categories of people
and practices. Of all three men, assessment may and should be made by
asking oneself, What surrounding and supporting modes of behavior must
be assumed to account for their having done, in the face of their whole com-
munity, what each of them plainly did do? Students of ancient religion or
of society or culture in any of its manifestations must be willing to pause
over individual pieces of evidence and make an attendant effort of imagi-
nation, in which the only aid that can be expected will come from a survey
of analogies. That means footnotes. Apology may as well be entered here as
at some other point for all the dense documentation that my purpose seems
to call for. In the broadest terms, it is at least clear that a very great deal of
religious lore was not only in lively motion among educated people but
actively thrust upon their attention from the walls of public buildings in
cities east and west.
The lex sacra is not the only genre of exposition (or however we might
call it) that is hardly known to us save through inscriptions: there is are-
talogy as well, the narrating of wondrous powers at work. Thus an inscrip-
tion from a temple in Stratonicea states "that statues are set up ... of the
aforesaid deities [Zeus of Panamara, Hecate of Lagina, and others-see
below] to make known the wondrous deeds, aretai, of their power,'' through
further inscriptions no doubt displayed on the bases. 67 Narratives of this
sort found an eager audience. Hence professionals, hawking them about-
heard agog by the simple but scorned by the wise. Hence, too, a half-dozen
texts that survive on stone around the Aegean from as far back as the second
or first century B.c. running down to the third century A.D., all in honor of
Isis. Their common source one dedicant places in Memphis's Ptah temple:
a source already in Greek, however, and in a style that would blend easily
into the Greek world. One derived version slips into literature in excerpts;
another, into both verse and prose on stone. The most recent version runs
like this:
To Karpokrates, to Sarapis, to the ears of Isis, to Osiris who gives heed, to
Hestia who rears from childhood [ ... ] I am Karpokrates, son of Sara pis and
Isis, of Demeter and Kore and Dionysus and Iacchus [kin?], brother of Sleep
and Echo. Every good season am I, providing for every time, discoverer of the
 1. Finding Order in Chaos

beginning [and end? 1. I am the framer. I was the first to construct shrines and
palaces for the gods; the modes of weighing and counting [ ... 1I devised. I it
was that made Isis' sistrum, and devised the hunting of every sort of beast [ ... 1
and I who established forever in cities their magistrates. I directed the children
as they were brought up. Hymns and choruses of men and women with the
Muses I established. I discovered the mixing of wine and water. Flutes and
pipes [ ... ] I forever attend among those in court, so that no injustice may be
done. I join in Bacchic revels of men and women, I gave free rein to [ ... ]. I
cleansed all the earth, I who live in mountains, seas, and rivers. I spoke from
oracular thrones and stars [ ... I who am] formed as a crescent, and Apollinian,
Highway-guard, [Dionysian] Bassareus, [Jovian] Heights-dwelling, [Diony-
sian] Ind-slayer, brandisher of the thyrsus, Syrian hunter, dream-visitor, vision-
sender [ ... ]. I welcome, while dealing justice to unlawful loves. I hate the
accursed. To physicians for recovery, every medicine [I give?].
Titanius of Epidaurus. Greetings, 0 Chalcis, who bore and raised me.
Liguris

(What the name Liguris signifies at the bottom, and who Titanius is, no
one knows.)
Isiac hymns survive in Latin prose of the second century, in first- and
third-century papyri, and from the pen of the second-century poet Meso-
medes. 68 Apuleius in the second century celebrated Asclepius as he had
learned to do in the god's temple at Athens, in both prose and verse, but
published in North Africa-"published" both through declamation and
circulation as books in Latin. 69 There is a still-surviving Latin hymn to the
Mother of the Gods, and in Greek, on papyrus, an author's unfinished
verses to Dionysus; 1o and there are favorite old hymns now lost71 and other
favorites even more ancient, recovered now from inscriptions: the Paeans to
Asclepius by Ariphron of about 400 B.C. and by Sophocles a generation
earlier,7 2 plus some otherwise unknown verses to the god by Aelius Aris-
tides. 1s These last are somewhat fragmentary; but it is pleasant to think that
in a museum in Turkey a traveler today can read the same stone read by that
gifted, boring, tenacious, pitiable, godly sufferer who dedicated years of
residence to the healing sanctuary at Pergamon, undergoing a thousand
experiences of divine assistance recorded for us in his spiritual diary. It is
pleasant to suppose, too, even if we cannot be quite certain, that Sophocles'
verses were sung before temples of Asclepius every morning in series un-
broken for six hundred years up to (and, as we know, more centuries be-
yond) the date at which their text in his own city was most recently inscribed
on stone.
That particular inscription is on a monument containing several other
similar compositions, not all by very inspired poets. Further epigraphic
texts could be added to them in good number, and from as far afield as
16 Perceptible

Mauretania and Britain. 74 But clearly their ancestry and origin had once
been Greek, and in the lands of that language is found the bulk of evidence
about what hymns were like, who wrote them, for what purpose, and how
they were sung.
So-called Orphic ones survive in a collection perhaps the best known
(though there were other collections in use). 75 It contains twenty-eight
rather sorry exercises six to forty lines long, sometimes addressed to local
deities of Asia Minor (where all the poems were written, most likely in the
second or third century), more often to deities like Artemis, Asclepius, Eros,
and Hermes. For example, the ninth: twelve lines that string together
nothing but thirty-three adjectives, seven adjectival phrases, two impera-
tive verbs, and two nouns ("king" and "maiden"), one of which is very
boldly repeated three times. Sophocles did better! But it is not untypical of
the genre, in apostrophizing the deity named, describing him or her hon-
orifically as descended in thus-and-such a line, capable of thus-and-such
wondrous actions, known in thus-and-such celestial haunts, gracious to
petitioners, Hail! Attend! Occasions needing particular celebration in
verse outnumbered and exhausted the Muses. There was the initiation of
aspirants to a fuller knowledge of some god, 76 plagues or imperial visits to
be called to divine attention, 77 festival parades, and daily services. 78 Quality
suffered. For hymns, simple shouted phrases might be substituted in the
god's honor (for Asclepius at Pergamon, at least sometimes) or the mystic
litany, "Bedu, zaps, chthon, plectron, sphinx, cnaxzbich, thuptes, phlegmo,
drops!" (for Apollo at Didyma).79
Instrumental accompaniment one expected to hear from worshipers of
Cybele, Mother of the Gods, both those itinerant and at fixed sanctuaries.
Typically, they played pipes made with reeds like oboes or bagpipes, ahl
bagpipes of a shrill and carrying note. 80 The cithara belonged especially to
Apollo; 81 to Isis, the rattle (sistrum), along with other instruments. 82 Cen-
sorious writes, "If it (music] were not welcome to the immortal gods,
theater spectacles would not have been instituted to conciliate the gods, the
horn-player would not be used in all sacrifices in sacred temples, nor would
triumphal parades be conducted with the horn or bass-horn player in
honor of Mars, with the cithara for Apollo, with the pipes for the Muses .... " 83
The picture Censorious offers, of accompaniment suited to the god hon-
ored, may be too schematic, but in its broad assertion of the habitual place
of music in worship it is certainly true to life.lt serves further to introduce
scenes of worship that are not Greek. In fact, music could be heard at
shrines of many -we cannot be sure of all-large regions throughout the
Empire. 84 Both at the heart of Greece and in an obscure shrine of Asia, both
 I. Finding Order in Chaos 17

to the great Apollo and to the unknown Sinuri, hymns were inscribed on
their temple's walls with musical notation, presumably to instruct casual
worshipers in the most acceptable tune. 85 Hymners who appear to have
been specialists and at least semiprofessional were sometimes heard; but
they cost more than cities, and obviously more than cult groups, could
easily afford; so either the emperor paid for them to serve the imperial cult
or they served more than one master.86
The word "hymner," hymnodos, may also indicate a sort of master of
musical ceremonies, semihonorific. Choruses made up of amateurs were
the general rule where cult and music were joined, choruses sometimes of
men or women of the upper class, sometimes of their daughters, normally
of their sons, boys before their voices changed. 87 They sang in groups of
twenty-five or thirty, except when they traveled to the oracle of Apollo at
Claros near Colophon. Then they usually numbered nine. As many in-
scriptions from Claros make plain, it was almost obligatory for corporate
inquirers, that is, cities as such, to return thanks for answers given them by
singing a hymn. In the second century, when the oracle enjoyed great fame,
it heard delegations of boys, their trainers, and conductors from many
distant places. 88 lt is odd to think of little groups of children, of course very
strictly chaperoned, traveling a hundred miles or more to the sanctuary "in
obedience to the oracle," as the inscriptions declare. After their performance
was over, they had their names put up on the walls of the shrine, along with
the names of the adults in the party.
From hymnology and aretalogy to my point of departure-general knowl·
edge about religion-! return by way of hymners who were also theolo·
gians, or theologians who were in charge of religious music; 89for in either
combination they suggest a close working relation between the two kinds
of expert. Moreover, hymners had also to pronounce prose hymns, called
"theologies."9° The third-century professor of eloquence, Menander of
Laodicea, wrote a short guide to their composition. They were to dwell "on
the naming of a god, or on valedictory, or on the god's nature or story or
birth or form, or to invoke or deter, or a mixture" -which sounds exactly
like the advice offered by Alexander the Rhetor in the preceding century or
like the epigraphic texts discusse4 some pages past-with the difference
that they were not versified. Menander indicates elsewhere the freedom he
would allow to speeches of praise- here, however, praising men, not gods:
"You should invent dreams and pretend to have heard certain voices and
wish to proclaim them to your listeners, or of dreams, for example, that
Hermes stood by your bed at night, commanding you to announce who was
the best of the magistrates; and, 'Obedient to his commands, I will repeat
18 Perceptible

from the very center of theaters what I heard him say ... .'" 91 That last boast
recalls another passage with the same setting, "when the populace as-
sembles in the theaters and someone enters wrapped in some special robe,
carrying a cithara and singing to it (as he says in the hymn of the Great
Initiation, without knowing what he says), 'Whether offspring of Kronos,
whether blessed of Zeus, whether of great Rhea, hail to thee, mournful
message of Rhea, 0 Attis. The Syrians call you Adonis, thrice-desired, all
Egypt calls you Osiris, Greek wisdom calls you the heavenly crescent of the
month, Samothracians, venerable Adamna; the people of Haemon, Cory-
bas; and the Phrygians, sometimes Papa, sometimes the corpse, the god, the
sterile unharvested, or the goat-herd, the verdant ear of grain gathered in, or
the piper whom fruitful Amygdalos brought forth.' " And so he continues,
quoting next from a hymn to Attis.92

2. ATTRACTING CROWDS
"In the theaters .. .'' Alexander and Menander, who made their living by
speaking to a general audience and, in their handbook~, by writing for a
still wider one, and who must then be supposed to have intended nothing at
odds with prevailing views but everything conventional, acceptable, and
safe, are the very best witnesses a historian could ask for-that is, if he seeks
to define broad cultural phenomena. Of which, first and most obvious to be
noticed is the size of audience implied by the need for a theater at all, a
theater called into use to hear verse or prose hymns about the gods only
because there were very large numbers of people to whom such things were
really important.
In the northwestern provinces, the evidence for crowds of worshipers
gathering around some cult exposition or spectacle is archaeological, un-
assisted by the literary; and the prime site is typically rural, not in some city
center. Adjoining dozens of more or less isolated temples are cult theaters,
sometimes right inside the same sanctuary wall. Functional links between
the two kinds of building are clearly indicated by various facts but are
actually clinched by those cases where the pair lie along a single axis; so
seated spectators could see past the stage (which lacks or breaks the back-
drop or scaenae frons) to the temple's front, beyond. Just what those spec-
tators witnessed, no one knows-no doubt the acting out of scenes or musi-
cal performances illustrative of the beliefs of the local cult. It was not one
god but many that received this tribute, in central and northern France and
the Moselle and upper Rhine valleys. It can have been offered only at festi-
vals, not daily or weekly, to attract the crowds requiring such accommoda-
 2. Attracting Crowds 19

tions. They numbered many thousands where they can be counted, for
example at Sanxay and Grand in southwestern and central France. 1
In the provinces of North Africa, only one similar site is known, a Hadri·
anic temple to Liber having a small stepped structure suitable for seating at
one end of the sanctuary, but lacking a stage. 2 It turns its back on the temple.
Tertullian gives a characteristically high-colored treatment to theatrical
presentations in Carthage with religious subjects. We will return to them.
But they were offered in secular amphitheaters, connected to shrines only
by the inaugural parades that reached them "from temples and altars."'
Moreover, they only followed Roman practices, brought in with everything
else Roman in the course of conquest and settlement. In their land of origin,
Italy itself, such practices are too familiar to need description.•
The Syrian priest and worshipers of Hadad and Atargatis dedicated a
theater on Delos in 108/7 B.c. It lay at one end of a terrace, the other end of
which held the temple; and on holy days the seated idol was paraded out of
ber house, along the terrace, to a marble throne that faced the theater, there
to receive offerings at an altar in front of her. What else was shown the
spectators is not known. Not quite a century later in the Hawran (Auranitis)
at Seeia (Si'), the temple of Ba'al-shamin was equipped with steps that
could be sat on along three sides of the main court, shaded by a portico
behind them, and with the temple front as the fourth side. The whole can be
identified as a theater by an inscription, and compared to closely similar
arrangements at two other sites in the region, Sur of the first century B.c. or
A.D. and Sahr of a slightly later date, with the difference that in the latter
two the steps are actually inside and covered by the portico, and that the
temple at Sahr is also flanked by a small proper theater capable of seating
several hundred people. 5 Moving north, a mile outside Gerasa lies the
sacred pool for the Maioumas festival, connected to the city by a sacred
boulevard and flanked by a Severan theater with fourteen rows of seats.
From the upper ones there was a clear view through the stage (lacking a
backdrop) to the pool. 6 Next, to the obscure sanctuary of Artemis at Amy·
zon in Caria, with a theater; 7 and so on to Dura on the Euphrates. There,
the theater attached to the shrine of Artemis Nanaia was a little smaller
(fourteen meters in diameter), ensconced in a closed building, and therefore
too dark for anything but dancing or singing; and there were four more tiny
theaters in the town tied to shrines of other Syrian deities. 8
Theaters for the ceremonies of Syrian cults, spread over several centuries
and a very wide area, display a good deal of variety, but they have in com·
mon the need to accommodate only limited numbers. The most natural
explanation for that is the use of the facilities for special doings in which a
20 Perceptible

minority of worshipers participated at any given time. That in turn might

be explained by the cult's offering revelations to which all members were
exposed only once in their lives, or rarely. Of this, however, nothing can be
said for certain; and in any case it is not my purpose to discuss more than
the open face and shared aspects of religious practices.
Accordingly, there is no need to do more than mention some of the cult
theaters for initiates in the Greek world proper. Of these, Eleusis is best
known. Clement of Alexandria, for example, speaks of its rites as still ad-
ministered in his own day. 9 The telesterion at Samothraee, with wooden
benches around three sides and a sign at the door, "Noninitiates stay out,"
is familiar, too. 10 More to my purpose because of its more significant
dimensions is the theater facing the temple front of the god Cabiros, a few
miles from Thebes in Boeotia. It had room for one to two thousand specta-
tors.11 And better still, the great Asclepius shrines at Pergamon and Epi-
daurus. The former lay some distance out of the city, at the end of a sacred
boulevard, and included a 3,500-seat theater adjacent to one corner of the
quite immense walled and porticoed sacred enclosure; the latter, Epidaurus,
though its shrine lay far off the beaten track, nevertheless required a theater
to hold 12,000-15,000. That was for crowds attending the festival every four
years. 12 Pausanias in the second century marveled at its proportions.
We cannot assume, however, that the wide and intense interest and en-
joyment in the "spectacular" side of worship (using the word in its root
sense), here at these locations so spectacularly accommodated, did not also
exist where similar facilities do not happen to be demonstrable. People
rather made do with available substitutes. They used wooden bleachers
which have left no trace for the archaeologist, 13 or, as at Altbachtal near
Trier, they placed odd-sized, half-cut stones for seats in a rough half-circle
around the slopes of a natural cup of land-facilities very easily overlooked
in excavation. Or again, they borrowed the city's central square for a few
days 14 -preferably, the city's secular theater or amphitheater. In Italian or
Romanized cities, performances with religious themes so offensive to the
Apologists inevitably took place where, on the next day, there might be
executions or gladiatorial combats. We will touch on those scandalous
dramas very shortly. But in the eastern provinces, in what category, secular
or holy, should we place the theater in which ended the procession of Arte-
mis's worshipers, in Ephesus, and in which the icon was ceremonially en-
throned to oversee, perhaps in its wooden way to enjoy, the musical, dra-
matic, and athletic events? In what category should we place the gladiatorial
troupe hired for exhibits locally by high priests, or (no doubt purely secular)
dancers and musicians paid to perform in the city theater all year long, as a
 2. Attracting Crowds 21

gift from the ministrants of Hecate? 15 Above all, there are the shows that
took place in the precincts of temples where, as is already clear in shrines of
Syria, steps could serve as seats. As much may be assumed of the hundred
yards of stairway with its eleven steps that stretch around the enclosure at
Lagina in Caria; as much is implied by the like features at Didyma near
Miletus and Eleusis near Athens. 16 Direct testimony to the requirements of
gatherings largely or wholly religious in purpose is enough, surely, to give
some sense of the weight or significance of that purpose in the daily life of
the Empire. But indirect testimony, too, carries weight: for every piece of
evidence must be measured in terms of the likelihood of its survival. By that
measurement, crowds attending in hundreds or thousands must be taken
for granted even in settings where their voices have been cut off by the total
silence of our sources.
Spectators in the numbers described could hardly have been assembled
for a lecture on theosophy. Something else it was that drew them: a varied
bill of entertainment, but still all holy. The Maioumas, notorious set of
rites at Gerasa, brought women out of town to a sacred pool in the suburbs,
there to bathe naked, that is, to be purified and magicked in some way never
really described, before the stare of the townsfolk. 17 But it happened only
every third year. Annually at Patrai, on the other hand, occurred a sort of
roundup of wild animals, a great battue, driving them into a corral dedi·
cated to Artemis, and up a ramp, and so (apparently) right on to a bonfire.
Pausanias describes the amazing scene-perhaps duplicated, however, at
another city in Pontus. 1s In areas and cities not enjoying an inheritance of
customs so dramatic, there were nevertheless very rich entertainments pro·
vided to visitors by the generosity of liturgists, hiring professionals. Com-
panies of dancers made their living through sale of their services to many,
many shrines. They specialized in the forms traditional to particular cults:
the Silens of Dionysus at Pergamon, the kinaedoi or wagtails of Oriental
cults (most often connected with Egypt), the ballatores of Cybele, the akro-
batai or toe dancers of Artemis, the armed Couretes for the same goddess at
Ephesus. 19 What everyone most liked to watch, however, were miming
dancers. Those in Egypt added their motions to the reading aloud of holy
texts by the priest, producing "danced hymns," as they have been well
described. The performers were professionals, men and women popular the
length and breadth of the province. 2o
Clement of Alexandria has a great deal to say, none of it good, about the
depiction of gods and goddesses acted out in initiatory performances. Per-
haps he means plays, not ballets; but his brother Apologists to the west-
Cyprian, Minucius Felix, and the other Africans of the second and third
22 Perceptible

A relief found not far from Rome (Ariccia) shows a religious scene in Egypt, identi-
fied by the ibises in the foreground and icons in a row at the back. To the right, on
the stairs of a temple, worshipers gesticulate with the music. In the temple court-
though C. Picard (1955) 244 would see a cult theater-dancers clack sticks like
castanets and whirl and wiggle frenziedly. Some may be female, others male. The
latter are nicknamed in Greek "wagtails" after the word for swallows, since the
swallow is a bird "qui agite particulierement la partie inferieure du corps" (so
Bernand 2 [1969) 121, discreetly).

centuries-make clear that there was no sharp distinction between actors

and dancers. The same word was used for both, and their means of portray-
ing stories were mingled. In the amphitheaters, "your deities dance, sup-
plying themes and stories to the wicked." Performers can "call up your
tears with make-believe agonies through the empty movements of hand or
head," and Venus leading a troupe of child dancers as Cupids uses her own
glances so expressively that she can be said "to leap with her eyes alone." 21
Venus was a popular subject, Hera also, Zeus or Jupiter, all in erotic scenes
no doubt presented very suggestively. Tertullian (Ad nat. 15) reels off a list
of common favorites on the stage: "Anubis Adulterer, Mr. Moon, Diana
Flogged, Jove's Last Testament, and as a comedy, Three Hungry Hercu-
leses."
In Italy, too, dancers and actors were more or less interchangeable.22
They were of course active in the capital, but in lots of towns up and down
Latium and Campania as well, playing the same parts as their like in
Alexandria or Carthage and just as closely tied to some of the major cults.
 2. Attracting Crowds 2J

They too used the tilt of the head, eyes, eyebrows, hands, and very fingertips
to communicate a great range of thoughts and feelings, much like Oriental
dancers to this day. 23 Presumably they found a responsive knowledge of
sign language among their audience, for much of the repertoire was highly
traditional, and symbolic gestures would soon settle into their own conven-
tions.
Tradition went by region. Each had its own. So one writer speaks of rites
addressed "to Egyptian deities generally by lament, to the Greek for the
most part by choruses, but to the non-Greek by the clangor of cymbalists,
drummers and pipe-accompanists"; or to the same effect, another reports
that "the Bacchic dance is taken especially seriously in Ionia and Pontus,
although it belongs to Satyric drama, and has so taken hold of people there
that, in the appointed [i.e., festival] time, they put aside everything else and
sit the day through, watching corybants, Satyrs and shepherds; and people
of the best lineage and foremost in every city dance, not in the least embar-
rassed but proud of it." 24 Lucian, the author of this latter passage, also cites
a number of other local specialties. "At Delos not even the sacrifices are
offered without dancing .... Boy choruses assembled and, to the pipe and
cithara, some moved about, singing, while the best performed a dance in
accompaniment; and hymns written for such choirs are called dances-for-
accompaniment." 25 It is likely that, wherever cult dancing is found in the
Greek-speaking world, it has something of local color about it.2 6
That takes care of the great public ceremonies on holy days and the reli-
gious use of the really large municipal theaters at places like Pergamon; but
there are the smaller facilities at Samothrace and elsewhere still to be filled.
They served "Mysteries." The word conjures up quite mistaken ideas of
secrets about the gods revealed in darkness to a tiny circle of oath-bound
devotees. At all periods there were indeed such groups and rituals-elabo-
rate ceremonies reserved for the specially devout (or wealthy), or more often
viewed by neighbors with disapproval or alarm (see section 3, n. 29, below),
A mysterion normally meant something more open and unexciting, essen-
tially a lesson in a cult to be learned, perhaps by very large numbers at a
time, as in the Eleusinian rites or in the mystery cult best known in the
period of our study, that of Cybele. Its setting might be in cities of Asia
Minor, where "the populace assembles in the theaters and someone en-
ters ... "-but the description has been quoted already.2 7 We gather from
the context as well as the explicit words that anyone at all could attend,
quite unchallenged, exactly as one would gather also from the facts: first,
that children and professionals-that is, outsiders-were employed in the
choruses to sing the hymns of revelation; 28 second, that theaters for mys-
24 Perceptible

teries were designed to hold, not scores, but hundreds or thousands. Lucian,
in his recollections of his old teacher Demonax (11), reports how th.tt free-
thinker was pilloried in Athens as "the only person of all that-had never
been initiated into the Eleusinian mysteries." No very select company,
those votaries! And the same author, in his essay On Dance (15), tells us
"there are no ancient rites of initiation without dancing," which satisfac·
torily distributes over a very wide area the conclusions forced on us from
other directions: so-called mysteries were in general quite open, come-if·
you-wish ceremonies, to which as large an audience as possible was attracted
by interpretive dancing, singing and music of all sorts.
"Barbarian" or "non-Greek" rites, notably of· Atargatis, Isis, and the
Phrygian Cybele, made use of inspired, mad dancing that produced a great
impression on observers and was also easily and often seen, since its prac·
titioners wandered about in public in search of an audience. To the sound
of rattles, tambours, and shrill pipes, with their heads tipped back or roll-
ing wildly on their shoulders, accompanied by their own howls and yells,
they whirled about and worked themselves into a state of frenzy. Then they
held out the begging bowl. The scene meets us in poems, novels, and fres-
coes-all too familiar to need discussion.29
The entire range of musical instruments known to the world of the
Apologists was called into the service of the gods in one cult or another, so
along with every conceivable style of dance and song, theatrical show, prose
hymn, lecture or tractate philosophizing, popularizing, edifying, and so
forth-in sum, the whole of culture, so it would seem. The same conclu-
sion can be expressed negatively. From the arts of those centuries, remove
everything that was not largely devoted to religion. The heart of culture
then is gone.
But another conclusion emerges: the heart belonged to the cultivated.
Hardly a novel insight! -yet very strikingly demonstrable through the
foregoing pages. Theaters were built and programs and artists chosen and
paid for by the more educated well-to-do, who, themselves, also offered
some of the performances. Of anything that could be called popular cul·
ture, hardly a trace appears. 31 Naturally, that fact reflects the state of the
sources, not of reality. The materials from which ordinary people made
their lives were wretchedly impermanent. Ovid gives a glimpse of a simple
gathering on the Tiber bank for a feast in honor of Anna Perenna. Of those
who attend, some pitch tents, others patch together huts of reeds or blankets.
They spend the night there. When the sun rises they begin the holy day's
dancing-and drinking. The glimpse is that and no more. Afterward,
nothing is left but the trampled grass. What are attested much, much more
 2. Attracting Crowds 25

often are "advanced" manifestations of the religious impulse through

specialized skills and settings, that is, through theaters and telesteria, not
tents; through trained choruses of aristocratic boys and girls under paid
trainers; paid dancers, too, and instrumentalists, not villagers and cottagers
doing their best. The 'structure of religious leadership corresponds in its
human shape to that divine hierarchy of frequency-of-mention which we
examined earlier. Within it, the many were quite subordinate to the few.
They acquiesced in the elaborate tribute of the arts !:>ought by the few for
Zeus, for Minerva, Artemis, and Mars. They acquiesced by streaming in to
the theaters and festival celebrations; not to the extent of quite forgetting
Anna Perenna and her like. There was room for many gods.
Throughout most of the eastern provinces, cities advertised their princi-
pal, most characteristic religious beliefs by sending out delegations of
heralds or by issuing small change, that is, bronze currency from their own
mints and of their own design. They wanted to attract wide attention, in-
spired by a mixture of sins: pride and avarice. So Ma, better known as
Bellona in the west, showed herself op the coins of her home in Pontus,
Men on coins of cities in his home Pisidia, the river oracle XPHCMOC on
coins of Limyra, Asclepius on those of Pergamon, the Mother of the Gods,
all over, and festival presidents on coins of many cities, of Apamea, for
example- by their title and prominence indicating what occasion they and
all th~se mintings chiefly served. Visitors not only fed their hosts' self·
esteem; they were good business.s2 Nor should it be thought that the phe-
nomenon of the cult festival, advertised to and drawing from the popula-
tion of a wide region, was not to be found where the imperial government
forbade municipal mints. On the contrary, the evidence comes from every·
where, even if not on coins, and so abundantly as almost to match, within
the economy, what we have seen cults supplying in higher culture: the very
heart. Their economic importance in itself has no claim to discussion here;
but the role of festivals in the exchange of ideas as well as goods is very
much our business. People assembling for commerce could not avoid ex-
posure to the challenge of others' religious beliefs.
Lyon, for example, had "a festival that was populous with throngs as·
sembled to it from all the provinces," in Marcus Aurelius's reign; under
Augustus, in half-deserted Fregellae near Rome, the people of half a dozen
surrounding towns "gather in it still to conduct fairs and certain religious
rites"; and to these welcome but rare hints from literary sources can be
added more from the archaeological, showing us facilities for trade, booths
or shops, built into the porticoes of sanctuaries or attested indirectly by
quantities of dropped coins in the soil, even by inscriptions recording this
26 Perceptible

or that trader as having his business at this or that temple. 33 To be sure, we

do not know if these signs indicate annual gatherings or, more likely, gath-
erings at closer intervals. They offer their testimony, in any case-the
"cafes and gardens," "the sales-tax-free monthly holy days on the fifteenth
and thirtieth" of each month-in the east as well as the west. In Syria, the
sanctuary wall at Damascus embraced a sacral area twice the size of a foot-
ball field, constituting at the same time the principal bazaar of that great
caravan city. 34 In Greece, the traffic that once nourished Tithorea now, in
Pausanias's time, flowed instead to the half-yearly festival-cum-market at
the Isis shrine a few miles away. Olympia had the same magnetic force on
commerce, earning its meetings the name mercatus 0 lympicus. ss That tells
something!
The fullest evidence of all, however, comes from the cities of Asia Minor.
In this region, the system of free gifts of money and services provided by the
well-to-do to their communities, and a healthy economy during most of
our period, combined to produce relative luxury at the holy days of smaller
centers and remarkable brilliance at the major ones. The liturgists involved
had saved up, sometimes for years, to be able to lavisq quite dispropor-
tionate sums on the duties of their term of office, magistracy or priesthood.
The greater the concourse attracted from round about! Hosts found it a
little hard to maintain their smile of welcome. 36 Their guests were, after all,
a very mixed crowd, in part unrestrained and unrespectable, in part pitiably
gullible, and most giving a good deal of consideration to pursuits entirely
secular. Hence, need for the authorities to put extra amounts of money into
circulation and give extra care to the policing of the marketplace. In time,
the office of festival president could be seen absorbed into that of agora
superintendent. Where that happened, it revealed much about the focus of
their concern. 37 How mixed the interests, how various the types of person
drawn to a festival, can be measured in the scene sketched for us by a con-
temporary: "some people attending out of curiosity, some for shows and
contests ... , and many bring goods of all sorts for sale, the market folk, that
is, some of whom display their crafts and manufactures while others make a
show of some special learning-many, of works of tragedy or poetry, many,
of prose works." ss
Developing out of terrain, seasons, plain routes of travel, and human
needs of all sorts, some of the smaller religious fairs known in the second
and third centuries filled those needs so naturally that, with changes only
on the surface, they were still being held in the nineteenth. 59 Some of the
grand ones drew worshipers "from remote regions for religion's sake," as
Artemis did to Perge or Ephesus, "venerated not only in her own home-city
 2. Attracting Crowds 27

... but by Greeks and barbarians." 40 Whatever their size or area of attrac-
tion, they constituted one of the chief means of introducing someone to a
larger world than that in which he was likely to pass his workaday life.
Thereby they help to answer the chief question latent in the last several
pages: How did the isolated and the uncommitted, people away in their
villages or people who had no interest in a given belief, come naturally to a
knowledge of it? In the absence of missionaries and mass media, it is not
easy to picture the two kinds of individual learning of many gods. Such was
certainly not the chief object of (for example) the high and mighty colonels
and imperial bailiffs who paid a call on the hot baths of Scaptopara, so
handy (as close as two miles) to a rustic cult center and its holy days. Reli-
gious learning can have been of no concern at all to those sons of heroes
who patronized a company of whores on circuit around Thermopylae as it
went from one holy day to another. And the desire to find out about strange
gods was certainly not what accounted for the peddler who set up his skin
tent near the sanctuary walls or for his customers fresh come from the plow,
in Tithorea. Nevertheless, their senses were assaulted by messages directing
their attention to religion: shouts and singing in public places, generally in
the open air, not in any church, and to an accompaniment as loud as an-
cient instruments could sound; applause for highly ornate prose paeans
before large audiences; enactment of scenes from the gods' stories performed
in theaters and amphitheaters by expert actors, singers, and dancers, while
the idols looked on from seats of honor; the god-possessed swirl of wor-
shipers coming down the street to the noise of rattles and drums; and all of
this without the onlooker's offering a pinch of incense or so much as glanc-
ing at a temple. So obtrusive upon the attention of the least interested was
the world of the divine.
Holy places only by chance lay where cities grew. Many lay rather at
some distance away: the Asclepieion in the suburbs of Pergamon, Didyma
twenty-three miles from Miletus, Daphne outside Antioch, Eleusis twelve
miles from Athens. There was thus much traveling out to see them. But
some deities themselves had to make an annual visit to convenient water.
Among them, Isis. Early in the spring her image was taken from its house
down to the shore, to set the ships asail, and again in the autumn, to find
Osiris. Other images went to water for a bath; 41 many were ritually carried
to a secular or cult theater, 42 many simply paid a brief call on their friends
and relations. They were borne on carts drawn by a pair of horses, by a deer,
by elephants (at least in representation); 45 they were carried on a litter with
human bearers (the means of transportation most often shown in reliefs,
coins, and frescoes). Or the parade included no icon, but instead people
28 Perceptible

assigned to carry a reed, a tree, a wooden phallus, a box, a jug, the god's
ornaments or clothing, a miniature temple or image.44 They moved along
to the sound of hymns and pipes. 45 Many processions were nocturnal, lit by
torches. Animals driven to be sacrificed had signs on them proclaiming
their donors' names. 46 In Greece and western Asia Minor, specifically at-
tested, regularly recurrent religious parades number in the hundreds, in
some towns taking place several times a year on behalf of different cults; but
beyond this area, too, and especially as a feature of Oriental cults, they are
known, though not so familiarly. 47 The reason for drawing out the evi-
dence on them is the same that required the describing of other public cult
activities: the need to make as vivid as possible the degree to which testi-
monies and teaching about religious beliefs demanded notice, even from
the most uncaring observer.
But perhaps the point has been labored, and to the neglect of a more
obvious truth: most inhabitants of the world here under inspection were by
no means uncaring observers. Had they been that, there would be no ex-
plaining their efforts to attend great or little festivals a day's journey from
their homes, or a week's journey-in short, there would be no explaining
pilgrims. These were a phenomenon distinct by themselves. Like many a
visitor of festival times, they spent money when they arrived at their desti-
nation; but only to make offerings. That is what Lucian means by saying of
the Holy City, "to the great goddess' temple in Hierapolis comes wealth
from Phoenicia, Babylonia, Cappadocia, and Cilicia." 48 Wealth, that is,
coins, discovered in the excavation of a shrine to the Mother of the Gods,
south of Pergamon, similarly shows the origin of visitors to lie in a number
of cities fifty, a hundred, or more miles away. 49 They had to be lodged in a
special hostel such as has been found in one form or another, and with a few
or with dozens of rooms, in many shrines, 5° though most frequently where
visitors' business would keep them for a matter of days, even weeks. That
could be necessary for the attending of a long festival (Olympia or Pana-
mara), religious study (Jerusalem), consultation of an oracle (Trophonius),
treatment by Aesculapius (Pergamon), or a water cure by deities of hot or
cold springs in many Celtic shrines beyond the reach of this study. Divine
doctors received many calls. Asclepius alone maintained hundreds of offices
open for patients throughout the Greek-speaking provinces,s1 of which
those at Epidaurus and Pergamon are only the most famous. The size of
their adjacent theaters suggests the numbers drawn to them. But memorial
inscriptions can also be used to learn the distances traveled by people
searching out the obscure springs at Vicarello on Lake Bracciano, all the
 2. Attracting Crowds 29

way from Cadiz; others, to Samothrace from far north; and so forth. Pil-
grims were ordinarily men, rarely women. 52
In Egypt, a land to all of which the Nile served handily as main street, it
was customary as nowhere else to pay one's respects to many gods, traveling
if need be many miles. Witness to the custom are some two hundred texts on
the walls of the Isis sanctuary and the buildings it contained, on the island
of Philae. They were inscribed in the period of the Empire, most of them on
the pattern, "I, Apollonius son of Theon, came to the Lady Isis [to offer] the
prayer for ... " (relatives-plus date). The prayer may, however, be versified,
like this: "When we arrived at the island of Isis, splendid and holy, at
Egypt's edge, before Ethiopia, we saw in the river Nile swift boats that bore
noteworthy shrines [miniatures, as offerings] from Ethiopia to our land
rich in crops and worth the beholding, which all mortal men on earth
venerate." u When, particularly in the second century but right up to the
end of the third, too, we find visitors to the remotest shrines of Egypt
coming from Greece and Italy, it is more easily understood why Greek and
Latin literature in the same period should be able to show so many readers
and writers well informed about Egyptian religion. Of course, the same
instruction could be gained secondhand. Every pilgrim who visited some
exotic temple then came home. His neighbors would hear all about it the
next evening, and for perhaps more evenings· thereafter than they would
have predicted, until they were fully acquainted with whatever he could tell
them. Pilgrims not only learned but taught, too.
To repeat: they were a distinct phenomenon. Their journeys were the
physical expression of· a faith greater than prevailed generally. But they
were not unique in their interest. At sanctuaries they were joined by, and
hard to distinguish from, another stream made up of plain tourists. And the
two met a third category: local experts. Lucian at the Holy City mentions a
hereditary corps of hosts to pilgrims; Aristides at Pergamon like Pausanias
in Lydia and at countless sites of Greece encounters and quotes from guides,
exegetes; Athenagoras, Lucian, and Plutarch and his friends quote the
priests both in Egypt and nearer home. 54 And there were people casually on
hand at temples, to be described just as learned or wise men; but of a special
religiosity accounting both for their presence and their lore.55They would
be most drawn to festivals, "a time at which," says Strabo, "it is best for the
participants to see and hear about these matters," namely, details of the
local cult.
Bookish Plutarch sat down with strangers in the precinct at Delphi, near
the temple, to discuss with them the significance of the letter E over the
JO Perceptible

entrance. They were soon joined by a chief priest and other people from the
shrine (not further explained, but ministrants of some sort). In the course of
their conversation they refer to "the tourist-guide view." The scene is
thoroughly believable-confirmed by Pausanias's report of a very similar
argument begun by chance at the Asclepieion in Achaea.s6 It could, per-
haps, just as well have taken place in a secular setting, in which other
conversations about religion arose: a beach at Ostia made famous by Minu-
cius Felix. But that was the scene of a debate no less profound between one
of Plutarch's friends and another philosopher.s7 Religion was talked about
in a public portico, where two strangers met, one of them in the character-
istic cloak of a philosopher. They entered on a debate before a cluster of
bystanders. And in the same sort of building but another time and temple,
Lucian imagines two philosophers of contrary schools colliding before "a
multitude of men, some inside the portico itself, many out in the courtyard,
some shouting and straining forward on their seats" -concerning the
question, Do the gods exist? 58 The stridency of interest is what counts for
our purposes-an interest which is far more easily documented among
Christians in the history of their internal debates but which animated non-
Christians as well. There even survive a few hints of the two engaging in a
dialogue with each other.59
Contemporary with some of these moments, a rhetor of the type we have
seen elsewhere on display in cult theaters offers us a lecture exactly on the
subject of chief concern to us: what were the means by which ideas of the
divine came to be shaped. He lists ideas inborn, those derived from the
poets, and those derived from decreed cult regulations. So far, familiar
sources. He adds a fourth, works of art. He proceeds to an extended discus-
sion of them. They belong of right to his subject. We have seen Plutarch
using them as a means of approaching and understanding the divine. 60
Pausanias in his tours of Greece, like Pliny earlier around Italy or Lucian
in Gaul and the Danube provinces, comments on the paintings he saw in
sanctuaries. 5 1 He expects to learn what beliefs are held at a given shrine by
studying them closely and hearing them interpreted by resident experts.
They are evidently to be looked for in any well-maintained sanctuary in
Greece or, indeed, in Asia Minor, in the Danube provinces, or in Africa.
That much appears in inscriptions boasting of their installation.62
Figured frescoes still survive on temple walls from an area even wider.63
Bas-reliefs were ordinarily painted, as too were architectural details on
.
Greek or Hellenizing temples (the practice is familiar) and on theaters also
(as may be learned from the example at Stobi); most wall surfaces not
marble would be stuccoed and painted; and, for the famous old cult statues
 2. Attracting Crowds

of earlier Greece or Rome, there was a match in the Sara pis of Pergamon,
thirty-five feet high even seated, or in the many images everywhere that were
gilded. That rhetor of the previous paragraph, Dio Chrysostom, whose
ideas on art we can examine quite fully, delivers his address while standing
in the neighborhood of the great Zeus image by Pheidias at Olympia, a fact
no doubt explaining why he should attach special importance to sculpture
compared with painting; but he is surely right in attributing to any true
masterpiece a kind of tyranny over every worshiper's subsequent percep-
tions (Or. 12.53 )-"by this glorious vision, conquering and uniting Greece
first and then others-revealing something so wonderful, so splendid, that
thereafter no one who saw it could form a different view."
Depending on the good will and wealth available to freshen the paint,
sacred enclosures and their buildings, frescoes, friezes, and statues must be
imagined, first, as drawing the eye with a very rich display of color through-
out, and second, as rewarding it with scenes and symbols full of meaning
for even a stranger. To this striking polychromy, inscriptions contributed
their share. They too were ordinarily painted, the letters alone or the back-
ground as well, with the obvious intent of providing better advertisement. 64
The importance attached to some of them is manifest in the efforts of the
authorities to maintain them in good repair, to show them off properly or
otherwise reaffirm them. 65 They constituted the best means by which priests
and pious alike could impress their beliefs on the public entering the
shrines.
While gods and goddesses resided in every temple and could be seen in
secular surroundings, too, at street comers or ensconced in niches in the
town hall, nevertheless it was the stone population inside sanctuary cir-
cuits but outside temples themselves that contributed most to the total of
30,000 statues of deities estimated to exist in the Empire as a whole; and this
population received additions all the time, to the point of needing to be
thinned out. 66 Prolific piety commissioned many of these conceptions, eveq
if many others were the offspring merely of convention: "to great Artemi~
Thermia, this bronze image in obedience to her command and revelation";
"So-and-So the barber, seeing it in a dream, erected an Asclepius to the
Nymphs, for his health, and gave thanks"; "the Ionian Polemo, sophist,
following his dream, set up a bronze statue of Demosthenes the rhetor in the
Asclepieion of Pergamon in Mysia, inscribing on it this epigram .... " 67
Only a selection from epigraphic memorials in the provinces east and west
suggests that a variety of cult reliefs, fonts for lustration, and whole chapels
as well as dedicatory statues were offered by worshipers obedient to the
gods' own expressed will. And if worshipers could not give life to the
12 Perceptible

images they set up, they could and did, by the score of texts surviving and
the 30,000 lost to us, communicate some of the vitality of their faith to
anyone who read on a statue base their affidavits of direct revelation.
No one dared disobey divine commands-or almost no one. Isis of
Tithorea wished only certain individuals at her festivals, and let them
know by appearing to them in dreams. Worth heeding, in the view of local
folk. "They say that someone uninvited entered the shrine and died soon
after. ... I heard the same thing from a Phoenician man," adds Pausanias,
speaking of Isis in Coptos. 6s An often-discussed type of inscription from
Lydia and Phrygia, belonging to our period, relates how some person went
against the will of this or that local deity, Apollo Lairbenos, Artemis
Anaitis, the Great Mother Artemis, or Mis (Men, and in that name settled
with his worshipers in Attica, too); whereupon the sinner is struck by some
punishment to self or kin or property, learns of the trouble "ina dream" or
"through command," confesses openly, and concludes his requital at the
deity's orders by posting an account of the whole affair on a stone tablet, to
advertise the deity's "powers" and "manifestations" (epiphaneiai).6 9 Too
many gods are involved for the practices reflected in these texts to belong to
any particular few; rather, to a region; and a fairly wide one.
Moreover, punishment for offenses against right and wrong, for example,
theft, is as sharp as for any against the rules of worship. The two kinds of
punishment together fall on someone who swears falsely before the altar.
Such perjury receives its due at Palmyra, where Iarhibol reigned; and con-
tempt for a deity brought on its penalties in Mysia near Pergamon and at
Epidaurus, too, as we can see in the very inscriptions that Pausanias him-
self saw and reported in about A.D. 170.7° They recorded from the fourth
century B.c. well over a hundred remarkable medical case histories, for
example, one in which a man came to the Asclepius shrine suffering from
crippled hands, and, "seeing the tablets in the sanctuary" (i.e., of exactly
the same kind as these texts themselves), "he disbelieved the cures and
somewhat sneered at the inscriptions. But going to sleep, he saw a vision
... ," etc. Or another man who, "transporting fish to Arcadia, in prayer
saying he would give a tenth to Asclepius from the sale of the fish, did not
fulfill the prayer. The fish suddenly bit him ... ,"etc. Or a third visitor to
Epidaurus, "suffering from an unhealed wound. While he was in Troezen
about to be cauterized by the physicians, in his s]eep the god stood by him
and told him not to be cauterized, but to sleep in the temple at Epidaurus
... ," etc.7 1
But whether the painted letters on marble stelae told stories five or five
hundred years old, anyone like Pausanias whose curiosity led him to enter
 2. Attracting Crowds JJ

the Asclepieion at Epidaurus would see them set out where they could not
be overlooked. Like the hymns written up on walls with indications of the
tunes best suited to accompany them, like the examples of aretalogy already
mentioned, in which wondrous deeds were strung together in rapid series,
or like the rules of ritual declared to be given by a god through dream,
oracle, or vision, the whole of a cult was displayed to the observer in the
most accessible fashion possible. Exegetes were at hand to address a stranger
and invite his questions. Benches invited him to sit down and talk or think
about what he saw before him: 72 the texts, the friezes on the temple, the
dedicatory statues with their inscribed bases, and the paintings most often
in stoas, where the rain and sun could not spoil them.
In Asclepius's casebook at Epidaurus, an alert editor can find the style of
simple folk, 73 the pinning down of highly improbable tales to named per-
sons and places, and other points of presentation likely to win credence.
Some of these same qualities Pliny remarked in an out-of-the-way sanctuary
deep in the Apennines. There "stands Clitumnus himself, girt in decorated
official costume. That he is a present power and prophetic, the oracular
responses show .... You will read many things written up by many persons
on every column and wall, praising that holy spring and the god. You will
admire most-but laugh at some." The passage serves as reminder that the
same devices of explanation and advertisement found on an elaborate scale
in a great center like Epidaurus were to be found also in the most remote
nooks of Italy.7 4 They were not the monopoly of Asclepius at Cos, Epi-
daurus, or Pergamon-not a peculiarly Greek possession, that is; 75 for
inscribed testimonials to divine cures distinguished shrines also in Italy,
North Africa, and Egypt.76
The modern student is unhappily aware of the capricious capacity of evi-
dence to survive at one point and perish at another. So much depends on the
nature of the materials to which records were originally committed: in the
Greek-speaking world, a great variety of topics on stone, and quite fully
expressed, but in the Latin west and north, a habit of inscription much,
much more shallowly rooted and perfunctory in its style. An ephemeral
literature on papyrus once existed to attest to miracles of healing wrought
by Asclepius in his Egyptian manifestation and, no doubt more often, by
Isis and Sarapis;77 there were prayers on papyrus once pinned to statues;
and almost equally perishable wooden panels once bore painted texts. 78
None of these records still exists today, therefore none of the beliefs and
events to which they bore witness can be substantiated. Luck, not truth. To
disappear without a trace is of course most likely to be the lot of those
simple folk and their faith sensed in the Epidaurian affidavits or smiled at
J4 Perceptible

by Pliny the Younger; hence, a specific distortion that is social, added to the
regional.
The illiterate worshiper could learn from unwritten wonders perhaps as
much as the lettered could learn from the written. He could infer the ob-
vious meaning from a pair of ears carved on an altar as thank offering to a
goddess in Bithynia. She had heard the suppliant. The symbols really
needed no explanatory text. They could be seen in shrines throughout the
area of our study, as reliefs on stelae or on a wall. 79 Their ubiquity is strik-
ing. In the major Asclepieia they seem to have been offered as things or
places (carved, and with a hole in them?) for a visitor to address, like tele-
phones linked to Olympus. And they expressed silently the character often
attributed to the gods in dedications, the character of listening. 80 Without
that, prayers would have no reason.
In healing shrines, ears carved in relief occasionally indicate a part of a
suppliant made well, i.e., hearing restored. Two· or three-dimensional
representations of feet occur, 81 which more often tell of the presence on the
spot of whatever person comm~ssioned the carving; but they, too, may
signify a healing of that part of the body. There are quantjties of other parts
and organs represented in stone or terracotta or, in the western provinces, in
wood, as is known only through mention in a literary source.a2 So much is
lost! There remain only tantalizing lists from earlier centuries, cataloguing
objects in some sanctuary: "on the third row, a leg in relief, uninscribed,
dedicated in the priesthood of Lysias," something else "hung from the
ceiling," or "on the left as you enter," "on the wall," "two tripods, two
small masks, two breasts, two phalluses, a snake, a hand." as It must be
supposed that the contents of temples had not changed by the point in time
studied in these pages; for stray objects of the sort one would expect to find
in the glass cabinets of an old-fashioned medical school turn up, most
improperly, in the excavation of second- and third-century ruins.

3. DISPLAYS AND ACCOMMODATIONS AT TEMPLES

It was customary for men to leave, nailed to the side of some building in a
holy place, a lock of their hair cut off to mark the end of their adolescence.
Accumulation of gifts could produce a peculiarly furry wall. I There were
elephant tusks nailed up at two temples in Africa, an elephant skull at a
third, in Capua (with a clever poem inscribed below it). 2There was a pair of
sculpted camels on view in a shrine a few miles from that skull, a pair of
sculpted lions elsewhere, and live lions, bears, and bulls wandering loose
but tame in the enclosure of Atargatis at Baalbek;' tame sacred snakes very
 J. Displays and Accommodations at Temples

fittingly at shrines of Asclepius, both the famous one in Epidaurus and at

Sicyon also, which visitors fed gingerly;• and at Baalbek again, sacred fish
in a pond- but Aelian rightly tells his readers that "tame fish which answer
to a call and gladly accept food are to be found and kept in many places,"
for instance, at Labraunda in Caria, where he says they had gold collars
around them and gold pendants like earrings on their gills.5 A number of
other ancient authors comment on the sanctity of fish in the cults of Syria.
They were one of the things that caused religion to be talked about. As a
wonder or a curiosity, they drew people to strange places of worship.
The whole city of Hierapolis served as sanctuary for doves held sacred to
Atargatis. The effect can be judged from the scene at Ascalon, in the same
region. When Philo arrived there in his travels, he "observed an enormous
population of doves in the city-squares and in every house. When I asked
the explanation, I was told it was forbidden to catch them ... , and the
creatures, having no cause for fear, had become so domesticated that they
not only regularly share one's roof but one's table, too." Similar conse-
quences must be imagined at Aphrodisias, where the governor intervened
to protect Aphrodite's holy doves; in Phrygia and Lydia, the gods them-
selves punished poachers; Achilles, worshiped in South Russia, guarded or
was served by "hordes of birds" on a holy island; and the Romans in our
period, it must not be forgotten, still put out to bid each year the feeding of
the geese on the Capitoline hill. 6
To complete the picture of religious centers constituting also cultural
centers, with zoological parks, aviaries, museums, concerts, art galleries,
and public lectures, or the equivalent of all these things provided nowhere
else in most cities, we must add botanical gardens: one consecrated to the
goddess of the woodlands, seat (says Pliny the Elder) of "a famous tree."
And he goes on to tell of the Italian bamboo, "which we commonly see in
temples." 7 At Ephesus, the shrine of Artemis had woods around that con-
tained rare trees, and perhaps still the deer parks of earlier times, nea,r the
copse in which ran the sacred spring.s Cybele shrines in every province
owned woods from which annually a pine was cut on the day "the tree
enters": arbor entrat (March 22). 9 It was in a copse that the Arval Brethren,
priests to the Dea Dia in Rome, held their frequent ritual meetings. All over
Italy it was immemorial to feel divine presences in such a place. No different
the Greeks, or other peoples of the east. 10 Indeed, whether from a feeling of
awe before the darkness of thick-clustered trees or to enjoy the refreshment
of their shade, the inhabitants throughout the world here studied attached a
special value to groves, set them apart and regulated their use. Mentions of
these doings and feelings are as many as they are casual, in the literary or
J6 Perceptible

epigraphic record-a subject to themselves, in which the careless student

could be lost without a trace. Safety for us lies in our having a clear and
limited purpose: it is to appreciate the ways by which religion was estab-
lished close to the center of daily life and therefore forced itself on people's
attention. That is easily imagined happening in, for example, the sanc-
tuary of Aphrodite at Cnidus. The space before and around the temple was
not paved but agreeably planted with orchards. "Under the most deeply
shadowy trees were cheerful picnic places for those who wanted to provide a
banquet there; and some of the more well-bred used these, sparingly, but
the whole city crowd held festival there, in truly Aphrodisiac fashion." 11
In this description we encounter a distinctive and ubiquitous feature of
cult centers, a place where a number of people could sit down together to
eat and drink. It is best understood through our first asking what is known
of ordinary social life and the equivalent of parties in the ancient world. To
this inquiry, answer is very hard to give. In larger houses, a dining room
could be expected to accommodate the usual nine guests. Only a really
grand mansion had couches for more. How then could lavish entertain-
ment take place? And for the great majority of people, whose quarters had
no dining room at all, where could they play host to their friends? The solu-
tion lay in sacred tables, where religious and social life were joined in a
common occasion.
The priests of Atargatis have already been seen at a fishy meal by them-
selves, from which worshipers were excluded. Similarly the Arval Brethren:
as might be guessed, being of the nobility and sometimes including the
emperor, they dined well. 12 These companies, however, and their like in
whatever region for whatever god are too private to serve common socia-
bility. Dedicatory banquets are too rare-by definition, once only. 13 More
to the purpose are those regular meetings indicated by the existence in
sanctuaries of buildings called kitchens. Let us look first at Italy. There they
are supplied, for example, by and to worshipers of Venus. Would guests
attend wearing the popular earrings of dangling little Cupids? 14 The gods
met one's glance everywhere unexpectedly. In Aquileia, gateway to the
Danube lands, there are a number of sanctuaries equipped with kitchens
for an equal number of different gods. On the river's bank at Carnuntum
was built, and later repaired by an ex-legionary in the early third century, a
shrine to the spirits of the woods and crossroads with a monumental en-
trance, stoa, and' banquet area. Throughout the whole region there pre-
vailed a cult of the so-called Rider Gods in which banquets were evidently
common, because representations of them in painted stone reliefs are also
common, at least in the third century .15
 J. Displays and Accommodations at Temples J7

St. Paul provides the most familiar evidence for our subject in Greece. He
speaks to the Christian community in Corinth about its members, or about
people who are at least not devotees of some given pagan deity, joining the
real devotees in that deity's temple grounds to share in the eating of
sacrificial meat. His rather offhand reference to the scene as something
quite everyday fits with the frequent epigraphic mention of dining rooms
opening off the stoas that ran around sacral areas. 16 As we would expect,
those Greek buildings and the custom they accommodated found a second
home in Asia Minor. 17 Also predictably, the best-known example was one
built for worshipers at the foremost cult center of all, for Artemis of
Ephesus. The benefactor who paid for it was, significantly, also noted for
his charities to the poor in the city. That beggars and the homeless found
refuge in temples will become clear shortly.
From the many terms used for the facilities that received the diner, it is
obvious that they varied in size and shape, some being big enough to hold
large crowds of guests, others, to hold only nine; and they might be found
only one to a shrine or, more often, judging from excavated remains, in
numbers ranging from three to more than twenty-five.
In Syria, even a little rural shrine might have five separate rooms for
eating; even a remote center might boast a banquet hall with a portico
around it. 18 Two well-excavated sites have supplied evidence on detail. At
Dura on the Euphrates, Syrians in small groups equipped their places of
worship (the word "temples" perhaps conjures up the wrong picture), each
group with its own dining room, eight or ten or twelve persons constituting
a cell or cult society (thiasos, as the Greeks called it) and their quarters
closely following a set model for size and shape-at the Adonis shrine,
anyway, and with a single grand exception which, according to its founder,
was to enjoy a peristyle and wine cellar. The building up of a congregation
of worshipers cell by cell can be seen also at the second site, Palmyra. One
whole temple there, though not very large, turned into a banquet hall seat-
ing about forty-five. Again, in the long, long effort of constructing porti-
coes, entrance colonnade, propylaeum, and so forth around the temple of
Bel, right through from the earlier first to the early third century, the work-
men coalesced into small religious fraternities serving the holy triad that
they had ensconced in the temple already finished; and they built and
feasted in separate chambers. 19 Throughout the city, where four tribes
made up the dominant part of the population and established and main-
tained various cult centers, there were unknown scores of privately funded
sacred dining rooms in which, on just what holy days and how frequently,
no one knows, there gathered the guests of the members in response to little
JS Perceptible

tickets of invitation, for elaborate cocktail parties-so they have been called
-at which there were dishes of hot meat and vegetables, perhaps (or after-
ward only?) pipe and cithara players to add entertainment; and next, the
guests having departed, the hosts settled down by themselves to serious
eating and still more serious drinking, served by butchers, bakers, chefs,
sommeliers, cupbearers, dancers, and musicians. Good manners obliged
the lolling votaries to toast all their companions but at the same time to
observe the sacred regulations against "whoever vomits up his wine."
People from this corner of the world, wherever they went, took with them
their loyalty to their religious customs. So when a temple was erected in
Dacia "to the ancestral deities Malagbel and Bebelahamon and Benefal and
Manaval," the donor provided it with a kitchen; the same in Rome and
elsewhere.2o And worshipers of the one god in particular, the divine pro-
prietor of the town Doliche, hence called Jupiter Dolichenus, added (once
at least "at the command of the god") a dining chamber to his temple.2 1
Our circuit of survey comes last to Egypt. From this land spread the cult
of Isis and Sarapis, along with its cult meals, even to Germany. Tertullian
reminds his western audience of sumptuous feasts in Sa~apis' honor. 22 At
home, however, Isiacism is far better documented through papyri. Eleven
of these resemble the hundreds of surviving Palmyrene invitations. In
almost unvarying words, as if by strict courtesy or convention, the writers
ask the recipients to meet them "at the couch [or dining room] of the Lord
Sarapis." Not always in his temple: "Dionysius asks you to dine on the 21st
at the couch of Helios, Great Sarapis, at the ninth hour, in the house of his
father''; or in the temple of some other deity. The best suggestion to explain
this is that, on any of the more obvious holy days of the year, the numbers of
people wanting to celebrate with their friends exceeded the capacity of a
single shrine, and some parties had therefore to find space elsewhere. 25
They could the more easily do so because there would be many gods in
town, and each or most evidently had commodious facilities for private
banquets, either built permanently in stone or less elaborately in the form
of arbors over stone couches. Users would lie on leaves or straw, hence the
name "mattress" given to a banqueting place as a whole; and they would be
shielded from the sun, hence the word "tent" used for the same purpose.
Aelius Aristides, the rhetor, reminds his listeners that "in sacrificing to
this god [Sarapis] alone, men keenly share a vivid feeling of oneness, while
summoning him to their hearth and setting him at their head as guest and
diner.... [He is] the fulfilling participant in all cult associations, who
ranks as leader of toasts (symposiarch) among them whenever they as-
semble" (Or. 8.54, l, p. 93f. Dindorf.). The passage tells something of the
 J. Displays and Accommodations at Temples 19

human feelings at such occasions, and invites speculation about other,

similar gatherings when defined, select groups added an express religious
element to a meal together. The phenomenon was widespread. "Devotees
of Hercules," "the devotees of Jupiter Axoranus," "the association of
Aesculapius and Hygia" (all these in the west) assembled at periodic din-
ners in temples- but then, of course, entirely secular fraternities used,
perhaps rented, parts of a convenient temple for their meetings or even
inscribed their rules on its walls; and the same customs could be found in
the Greek provinces.2 4 Observing the latter from a distance, Tertullian re-
marks that "for the Apaturiae, the Dionysiac or the Attic mysteries, a draft
of cooks is proclaimed." 25 From the east to the west, too, there spread the
worshipers of Mithras, bringing with them, like worshipers of Jupiter
Dolichenus or of Isis and Sarapis, their own religious practices, which
included common meals. Wherever they built their cavelike chapels, ar-
chaeology uncovers gnawed bones of all sorts of edible animals.26 And for
votaries of every sect or none at all, there were graveside meals in memory of
the deceased. Those too are common. 27
What emotions animated these scenes? The question, because of its in-
trinsic interest as well as its bearing on the Christian Eucharist, has been
often examined. It barely touches our subject proper-fortunately, since
material for an answer hardly exists. Two points may nevertheless be made
without need of any long discussion.
First, the idea that a god might join worshipers who were eating together
was widely diffused, most overtly in celebrations called lectisternia, when
the icon was brought out of its house and was laid on a couch beside the
celebrants; but also in other settings, larger and smaller; and obviously the
emotions prevalent in some one setting could not be expected in all. The
range would run from moments of great intensity (of hope, love, abase-
ment, peace, and more) to moments, hours, or days of a festival in which
religious meaning was all but forgotten. It has been the tendency of recent
studies, in my view very persuasively, to deny much scope to the former, the
more intense feelings.2s ·
Second, many signs prove the grossest indulgence of the appetites at
banquets of the sort being considered. At least it indicates their reputation,
that a novel could describe an initiation at which the banqueters eat, vomit
up, and eat again human flesh, while, in drinking, only the presiding priests
exercise any moderation. A wild, revolting fantasy, no doubt; but the
vomiting of wine we have seen at Syrian meals.29 When Christian graveside
meals enter the literary record a century later, scandalous drunkenness at
them earns rebuke,so and it is a fair guess that their patterns of conduct had
40 Perceptible

been taken from the pagan. When we turn to citywide parties, evidence for
heavy drinking grows. We will consider them shortly. Does all this add up
to something less than the spirit of worship, to shallowness of feeling, a
secular devaluation of once-fervid beliefs? Or can that only be said if some
extraneous standard of behavior, defining what is and is not "truly" reli-
gious, were imported into the world of the Apologists? That was the latters'
intent. It need not be ours, at our remove.
Citywide cult celebrations are described, not lovingly, from a Christian
viewpoint. They are said to be staged in the name of false worships, "to
which most men abandon themselves at festival time and holy days; and
they arrange for drinking and parties, and give themselves up wholly to
pipes and flutes and different kinds of music and in every respect abandon
themselves to drunkenness and indulgence." 51 Against this interpretation
of how people usually acted, an apologist, Plutarch (Moralia 1102A),
might offer another: "It's not the abundance of wine or the roasting of meat
that makes the joy of festivals, but the good hope and the belief that the god
is present in his kindness and graciously accepts what is offered." The pic-
turing of the scene thus aroused in contemporaries a range of emotions; but
the visible events were clear enough. A good meal was had by all, or many
meals: for thirteen days, free meat with hymns and parades to, from, and in
Apollo's shrine at Didyma; or feasting, dancing, and singing for the twelve
days of Aglibol and Malakbel in Palmyra. 52 We will turn later to more
detail about such long-drawn-out affairs, as it can be recovered from the
inscriptions of Caria; but here it is enough to offer no more than the two
quoted descriptions from the Greek-speaking world. Their like are found
very rarely in the west. 55
In opening a glimpse upon these festivities, my aim is to place religion at
the heart of social life as surely as it must be placed at the heart of cultural
activities of every sort. For most people, to have a good time with their
friends involved some contact with a god who served as guest of honor, as
master of ceremonies, or as host in the porticoes or flowering, shaded
grounds of his own dwelling. For most people, meat was a thing never
eaten and wine to surfeit never drunk save as some religious setting per-
mitted. There existed-it is no great exaggeration to say it of all but the
fairly rich-no formal social life in the world of the Apologists that was
entirely secular. Small wonder, then, that Jews and Christians, holding
themselves aloof from anything the gods touched, suffered under the repu-
tation of misanthropy!
In rejoinder, they pointed with elaborate repugnance to "the pollution
around the idols, the disgusting smell and smoke of sacrifices, the defiling
 J. Displays and Accommodations at Temples 41

gore about the altars and the taint of blood from the offerings." 34 Did they
overstate the case? It was a pagan who described "the priest himself [who]
stands there all bloody and like an ogre carves and pulls out entrails and
extracts the heart and pours the blood about the altar." 35 It is clear that the
great bulk of meat (not fish or fowl) eaten in the ancient world had been
butchered in temple precincts, most of which, ill-supplied with water,
could not be swashed down easily, accumulated ugly piles of offal in cor-
ners, and supported not only flies in clouds but stray mongrels as well.
There would have to be wood stacked handy to the altars, and provision to
store the hides stripped off the carcasses, since at least in some temples they
were reserved for the priest. He was entitled, after all, to make his living
from the premises.!& It appears common for him to have received a portion
of each victim which went to "the god's table" and which he could then do
with as he wanted: give for a feast or sell to retailers. That latter choice
would explain Pliny's remark that, as Christians in Bithynia were perse-
cuted, "the temples begin to be crowded and the meat of victims to be for
sale everywhere"; but a number of cult regulations specify that the hide, the
head, or a set fraction of the whole animal belongs to the god (hostile ob-
servers making fun of the mean portion he receives), and such parts are to be
sold.37 As to what worshipers could keep, they consumed it on the spot.
Rules might even forbid their taking anything away, since symbolically all
had been consecrated. Here served those kitchens, cooks, couches, arbors,
and so forth that have been discussed. "Sacrifices were devised by men, I do
think, as a pretext for meat meals." 38
Though most of this detail about the apportioning of sacrifices comes
from services in Egypt, Asia, and Greece, enough touches other regions to
suggest that practices were broadly similar everywhere within the lands
here being examined. The western provinces, however, yield a different
kind of evidence of their own in the form of the remains of feasts, bones,
skulls, and broken cups. The use of horses as sacrificial victims in Celtic
areas seems securely attested but not explained. 39 In the region to the south
and inland from Carthage, Punic custom directed worshipers to tophets
with altars but no temples (perhaps a small chapel or two}, where sacrifice
was made an~ the remains usually consumed in a feast by the party that
brought the animal. Unused parts were buried under memorial stelae,
often decorated with painted reliefs that help to explain the whole routine.
In one such sanctuary dedicated to Tanit, "pele-mele with the charred
wood was a prodigious quantity of half-calcined bones, among which
could be easily distinguished the jaws of oxen and sheep and a large quan-
tity of leg-bones," plus thousands of lamps, hundreds of unguent bottles,
42 Perceptible

and so forth, built up over time to a depth of six feet. 40 The absence of stone
couc;:hes at shrines of that god who was by far the most widely revered in
Carthage's old dominions, namely Saturn, and the Celtic practice of sacri-
ficing horses, together warn against assuming uniformity too readily across
all parts of the empire, where explicit evidence is lacking; while, on the
other hand, in all parts were many gods, one or another displaying the pre-
dominant cult practices-the offering of a sheep that worshipers then
shared with family and friends, and the like customs- that we have been
surveying. What was not native might thus still be familiar.
At the Pergamon shrine of Asclepius, Aristides dreamt the god appeared
to him with instructions to "make an offering to Asclepius ... and distribute
the holy portions to fellow-pilgrims." The latter must be added to the num-
ber of possible or likely beneficiaries of sacrifice. They would include per-
sons making long stays for a cure, of months or years. At the shrine favored
by Aristides, vast though the whole complex was, its recovered plan does
not show clearly just where they lodged; but other Asclepieia offered good
accommodations. Some of the buildings have been described above. 41 Against
modern stereotyping, Asclepius drew pilgrims also for initiation into the
mysteries of his cult at Epidaurus, and Artemis to Ephesus, for healing.
Many or most deities attracted seekers of many sorts, it should be remem-
bered. 42 Inscriptions of Asia Minor speak specifically of a category of per-
sons whose existence we could generally assume anyway, the more or less
permanent hangers-on and dependents of the whole fat, smoking cluster of
b,uildings and activities that Aristides dwelt among. Who these were, besides
paupers and fugitives enjoying asylum, cannot be said. 45 A second category
to be taken for granted St. Paul encountered at Ephesus in the person of
Demetrius, "a silversmith who made silver shrines of Diana and provided a
great deal of employment for the craftsmen"; and the pagan holy man
Apollonius met the like in Athens, a wholesaler of statuettes of gods for sale
to worshipers. 44 Excavation of shrines in Africa and the European prov-
inces often turns up quantities of objects all nearly alike, which had some-
how become conventional offerings to this or that deity and in whose
manufacture there was obviously involved a sizable population of artisans.
These one would expect to find in or near temple precincts.

4. ROUTINE STAFF AND ADMINISTRATION

With temple servants proper, with priests above all, we return more closely
to our intended subject; for, to the extent that they were not purely munici-
pal magistrates of secular character, they took care of dressing the face of
 4. Routine Staff and Administration 4J

religion for its votaries. But it must be remembered how far from purely
secular were most elected officials in cities of Greek or Roman derivation.
This year to the gods, the next to the city-such was the pattern of service
rendered by the local aristocracy. Even within a given office, the two kinds
of duties could not always be kept apart: "the strategos assisted at the tradi·
tiona! procession for the great goddess Isis ... ," "for the visit of the most
distinguished prefect ... , all [statues] in the temple crowned with wreaths
... , anointing of all the statues in the temple with oil ... , workmen accom·
panying the image of the god to go and greet the prefect, wreaths for the
image, a rhetor addressing the most distinguished prefect ... "-so read the
entries in the daybook of a magistrate and of a priest regarding quite
similar public occasions and, for both of them, quite similar duties in third·
century Egypt. It would be the same in any province. 1 As the gods came out
to meet officials, so officials visiting a city for some secular purpose paid
courtesy calls on the gods; 2 and magistrates made free use of accommoda·
tions in the porticoes of sacred enclosures, for their business, in just the
same way that votaries as such met freely in civic buildings. 5 So closely were
sacred and secular services confused in some settings.
In others, not. From the gallery of faith in the Roman world no picture is
better knowri, because none has been more often painted, than that of the
fanatic who, half-naked, whirls about in a mad dance, flails at himself till
the blood flows, gashes his arms and legs, even castrates himself in the
transports of his faith. From northern Asia Minor originally came the cult
of Ma, providing Romans with their first sight of such wild doings; then
from Phrygia, the cult of Cybele and her eunuch priests who howled like
dogs; then from Syria, the cult of Atargatis, another goddess served by
eunuchs; and Isis, whose devotees shaved their heads and twitched about in
dance to demonstrate their grief at the death of Osiris, later to rejoice as
extravagantly at his recovery.• Their long hair was thought to be a sign of
divine inspiration, the longer the better,s and they and other priests, not
only those of Isis and Cybele, wore distinctive costumes. 6 Outside the cults
of the four eastern female deities just named, ascetics are indeed attested-
extremely rarely; and in Cilicia you could see inspired votaries walk bare·
foot on hot coals, in trust or tribute to their local goddess, as still today in
Macedonia the votaries of St. Constantine walk on coals "at the command
of their General," as they say. 7 But the sum of all this evidence for enthusi·
asm in the literal sense is striking by its narrow limits, all thing~ (all gods)
considered.
The essential figure in sacred precincts is no doubt neither the aristocrat
nor the fanatic but the priest, ministrant, or caretaker of whatever title.
44 Perceptible

Often a group of them acted together, answered questions and kept order.
Enthusiasm was not the mark of such persons at all. Mentioned by travelers,
they remain always nameless, humdrum, and helpful. In the sanctuaries of
most gods it was rather the worshipers in whom eccentric behavior appears.
So we can gather from scattered small hints, and from a single description
that survives of people in a temple enclosure. It was written by Seneca:

Go to the Capitol and you will be ashamed of the folly there disclosed, and of
the duties which a deluded madness has assigned itself. One servant informs
Jupiter of the names of his worshippers, another announces the hours; one is
his bather, another his anointer, that is, gestures with empty hands to imitate
the act of anointing. There are women who are hairdressers for Juno and
Minerva; while standing far away from the temple as well as from the image,
they move their fingers as if they were dressing the hair, and there are others
who hold a mirror. There are men who summon the gods to give bond for
them, and some who offer them lawyers' briefs and explain their case. An ex-
pert leading actor in the mimes, now a decrepit old man, used to act a mime
each day .... 8

Certain elements here we can identify: interpretive cult dancing and the
calling of the god to witness and judge suits and oaths. Such customs we
have seen before, though not at Rome. One element is obscure: the naming
of worshipers. Others will need to be discussed.
For with this passage we approach the center of our subject, the actual
idol in its house. Our first-and by anticipation, it may be said, likewise
our last-impression is curiously disappointing. Once past that circle of
devotees that Seneca describes, we are left to the company of sextons. They
have their everyday duties. They can sometimes be made out at daybreak
opening the temple doors and singing a hymn to the deity- this for Sarapis,
Asclepius, or Dionysus. 9 The altars had to be lit, too, each one to a prayer or
acclamation. 10 There followed one or two later junctures for hymns or
sacrifices during the day, and then the doors were closed again.tl But it may
be that very few places of worship were regulated according to so nice a
timetable. Certainly some temples never opened their doors at all, rather
were secluded from their worshipers behind an anterior screen or fence.
Others were never closed. Apparently it did not matter. And some admitted
worshipers to their interior for services that required permanent seating
arrangements or for sacrifices of burnt offerings.t2 Lack of any pattern of
practice, almost total lack of mention of gods seen in their own dwellings,
suggests that idols played no very active part in the ongoing life of their
cult. They represented the pictorialization prevailing in a community at a
given moment, caught by an artist, thereafter to serve as a kind of aide-
 4. Routine Staff and Administration 45

memoire to the faithful. It was what the faithful did in the precincts around
them that counted for much more than what could be seen, or never seen,
deep in the inner sanctum.
To both precinct and temple within it, night brought the illumination of
lamps dedicated as offerings or of statues holding torches. 1' The effect had
something magical about it, a sure sign-part of the evidence deserving
discussion later on- that persons in charge of the premises consciously
arranged things to arouse certain feelings in observers. The same inference
may be drawn more specifically from the revised liturgy of the Arval Breth-
ren, whereby, after the Brethren's cult dance, the statues of the goddesses in
the temple were rubbed with perfumed oil, candles were lit all about, and
the temple doors were thrown open.• 4 The anointing of idols, if not merely
mimed, made them gleam in lamplight. It was also a symbolic attention,
like the dressing_of them in rich clothes. As if they had been great dolls, they
were prinked and bejewelled and shown off in processions by specially
titled servants: "robe-bearers," "adorners." 15 The same literal piety was
seen earlier washing them and stretching them out on couches in the
company of their devotees for a feast.
Much of the long-elaborated, colorful sequence of activities that went on
about the gods in their daily or yearly calendar has left its trace in sources
for Greece and Asia Minor (actually, parts only of these broad areas), but
not elsewhere. How is this unevenness of evidence to be handled? The most
obvious recourse is simply to indicate what can be found where. But beyond
that, it would be good to know if paucity of evidence corresponded to
paucity of show in the cults of the province of Africa Proconsularis or
Gallia Lugdunensis. It appears certain that Mithraism, lsiacism, and Cybele
worship were well known (though not of course known to literally every-
one, either in the west or in the east) throughout the Empire. With them
they brought a full liturgy, in every sense of the word. It figures in a long list
of pagan and Christian Latin authors. Authors, however, by no means tell
the whole story. If it were not for the two words written next to figures on
the wall of a Roman chapel, in a text all but destroyed, we would never have
learned that Mithra was addressed with hymns.•s They and other features of
cult are relatively familiar in the east, not in the west. The chief cause for
the difference lies in habits of epigraphy, by which the Greeks set down a
large variety of subjects on stone, rather discursively, while the Romans did
not. We have confronted the consequences of those customs above. The
people they conquered took on their masters' habits. Thus it could be
argued that, in Latin lands, the example of liturgy of eastern type and
origin was placed before people, therefore no doubt copied, but not reported
46 Perceptible

in the type of source on which we depend in Greek lands, namely inscrip-

tions. On the other hand, had the Romans possessed a native elaboration of
cult to match that of their eastern subjects, it seems odd that they should
have imported so much from the east into practices of emperor worship.
Such borrowing is plain in texts from Gythion and Arsinoe, for example,
and the role of the Dionysiac artists is specially revealing. But all that evi-
dence has still to be put together into a single picture, one that does not
belong to the gallery that is my concern here.
Perhaps what is of concern, the salience of worship that could not be
ignored in daily life, can best be illustrated by the customs of Stratonicea,
near the southwest comer of (modem) Turkey, with its two nearby sanctu-
aries, Hecate's at Lagina and Zeus's at Panamara. The cult of the latter is
especially well documented. Indeed, it may be known in fuller detail than
any other in ancient times save Judaism. It is, however, quite neglected,l7
perhaps through appearing nowhere in any classical text, only in inscrip-
tions, and through being neither Roman nor Greek. Students who attach
much living significance to those latter two words in the world of the
Apologists, except as they refer to language itself, have (I would think) a
good deal of explaining to do; but that is a separate matter. Enough to say
here that the coloration taken on by one of those regional deities with
which this chapter began throws useful light on the degree of uniformity
that may be looked for in our area of study. It was, as I said at the outset, a
proper melting pot.
Panamara was the home of a male divinity whom visitors and conquer-
ors settled on as Zeus, whatever the natives once had called him. The sanc-
tuary occupied the flat top of a hill. The circuit wall enclosed a space about
100 by 85 yards. In and around the space and on the temple itself were more
than four hundred inscriptions. They were published in quantity begin-
ning about a century ago. There was also a Hera temple in the sanctuary,
receiving new paint and plaster one year, as we are told; a dining hall; "the
god's tables" supplied by the gods' share of the sacrifices offered; a building
called the Komyrion; and a monumental portico dressed in stucco. Among
the votive objects to be seen, perhaps the most precious were four feet made
out of gold to commemorate the gods' active manifestations or epiphanies.
Around the sanctuary were ox and sheep pens for sacrificial victims and
inns nearby for those who were going to eat them. Inside were seating ar-
rangements large and formal enough to be referred to as_a theater and room
for crowds to sit or lie at meals under tents. Wherever the eye rested, there
must have been the inscribed and painted letters of honorific decrees, testi-
 4. Routine Staff and Administration 47

mony to miracles, or regulations specifying the music and victims appro-

priate to this or that act of supplication or thanksgiving.
"Since, by the celebration and veneration at the festival and mysteries we
enjoy great benefits from the gods' manifestations ... ,"begins a decree of
the Stratonicean senate and citizenry. 1s The pride and piety that speak in
the text spoke also in written invitations that went out to senate and people
of many cities-some elsewhere in Caria, but also in distant Miletus and
Rhodes-and in posters and by heralds to the population of Stratonicea
itself, from the priest of Zeus, asking everyone to join his festivals. They
were the annual ten-day Panamareia, the one-day biennial Heraia, and the
two-day Komyria every fifth year. At the beginning of the Panamareia, the
idol was set on horseback and taken along the road, a part of which is still
paved, for a sojourn in Stratonicea, during which his priest used the city
theater to present dancing, singing and acting by professionals hired for the
occasion, and in the public baths distributed free oil to the bathers, men and
women in their separate buildings, plus cash to the needy in the market-
place, wine, and take-home meals. At the end of the sojourn, those who
accompanied the god to his home received gifts of sweet pastries and wine
en route.
For the other two festivals (the Komyria having been instituted only in
the first century to solemnize what had been a private rite of adolescence),
the amenities were still more lavish: aged wines at banquets with no head
table, for the ordinary arranging of guests of course involved worse or less
food to hoi polloi; no head table, but drinking parties, love and good cheer,
symposia, agape, and euphrosyne, often advertised as part of the "mys·
teries" which begin to appear in Marcus Aurelius' reign in connection with
the Komyria. 19 In a period and sequence not easily fixed but nearly com-
plete by the later second century, men are first (and necessarily ever after)
admitted to the Heraia, and women to the Komyria; and persons not of
Stratonicea-Roman citizens, country folk, even slaves-are invit{:!d, and
then feasted; and then new items of increasing luxury and frequency are
added, in a great tower of celebration built by competitive generosity
among the leading citizens over many decades. In the third century the so-
called gymnasiarchy of the high priest, meaning the period during which
he supplied anointing oil to the baths and gymnasia in the city, totaled
twenty~two days; in the fourth century, thirty-four days.
But that included the Key Procession of Hecate in Lagina, whose shrine,
a couple of hours' walk from Stratonicea in another direction than Pana-
mara, also fell under the presidency of the same annu<J,lly elected priest. He
48 Perceptible

would be expected to improve the facilities there, for example, by commis-

sioning and paying for "three porticoes and monumental gateway and the
stoa along the front of the temple toward the food market, and a decorated
sideboard for the Table of the Goddess"; 2o and such a generous person
could expect to be thanked by vote of the senate, citizenry, Elders, and in-
habitants of the shrine, who, with the hired entertainers, lived there year·
round.
From among the children of those inhabitant families, the priest chose a
choir of thirty to sing "the customary hymn to Hecate just as it has come
down from the past." 21 This was to be done every day in the municipal
senate building, the choristers wreathed in flowers, holding flowers, and
accompanied by someone playing the harp. A charming sight, before the
lawmakers bent to their labors! Outside, in the portico, since a date near the
tum of the second century, stood statues of Zeus, Hecate, Artemis, Asclepius,
and Hygieia, statues which wrought miracles. In honor of these, musical
performances began to be staged on the spot. 22 Were we to choose one point
in space and time that brought to a focus the beliefs and practices described
in our first chapter, surely it would be this columned portico on some
morning around the year 200.
 n
Debatable

1. NEEDS AND ANSWERS

Was every one of these many gods just an invention? And did the whole
structure of belief that we have surveyed thus rest at last on pure delusion? A
challenge to be taken seriously. As superstition, so-called by some, much of
that whole survives today; much (a little changed) receives attention daily
in the newspaper, for readers who see stars. Historians, however, moved by
some natural persuasion or by the dictates of their craft, pretend a terrible
impiety. They must pretend, not that the gods lived and ruled, but that they
did not exist-yet served.
The altogether human needs they served give us our starting point in
examining the life that lay, not in paganism observed, but in paganism felt
and thought out. Those needs are reflected in the activities that went on in
any center of worship-activities most of which, however, might have
fitted just as well in quite different settings. Entertainment or barter or
sociable intercourse need not have been sought of absolute necessity in
temple precincts and nowhere else. What really did belong? The answer
should lie in whatever remains when everything expendable or extraneous
is somehow stripped away. There is no more obvious place to begin the
search than in Asclepieia. Those particular shrines were very numerous
and, if all other places of resort for healing were counted in as well, drew to
themselves more suppliants than any competing category of belief. What
could be found there? Dreams and health. The chief business of religion, it
might then be said, was to make the sick well.
A simple test can be made through inscriptions. They tell us what was
expected of gods, sometimes by attaching epithets to them, for example, in
address or hymn to "the Healer," sometimes by specific thanks for a cure.
By the former practice, Asclepius alone is "the Savior," without further defi-
nition; Isis and Sarapis, too, emerge as specially great healers, workers of
endless medical miracles; but many or most gods could heal. 1

49
50 Debatable

A semifictional biography tells us of sickness caused by a maleficent

spirit, who can only be destroyed when a holy man, come to help, calls on
the power of some appropriate deity. The anecdote is not particularly rep-
resentative of beliefs and practices prevailing in its dramatic setting, western
Asia Minor in the later first century; 2 but it provides a convenient transition
from the curing of the body by Asclepius, Isis, and other gods, to that of the
mind. Mental sickness, too, could be and most commonly was explained as
possession. Cure lay through casting out the devil. And the same holy man,
Apollonius, is seen to be capable of this feat also.' Powers such as his were
commonly on sale in marketplaces throughout the area of our study-so
the rather sparse and casual mentions seem to indicate. Though exorcism
provided agreeable shivers to the reading public and fed more serious con-
victions among evidently unlettered folk, the art itself had no great fame or
audience.
Christians, however, had their own views. They stood out as frequent
and powerful exorcists, " ... as you can learn even now from things done for
all to see: for many persons possessed by demons, everywhere in the world
and in our own city, have been exorcized by many of our Christian men";
"for some people incontestably and truly drive out demons, so that those
very persons often become believers, those cleansed of evil spirits"; "let a
man be produced right here before your court who, it is clear, is possessed
by a demon; and that spirit, commanded to speak by any Christian at all,
will as much confess himself a demon in truth as, by lying, he will else-
where profess himself a god"; and "traces of the Holy Spirit are still pre-
served among Christians, whereby they conjure away demons and effect
many cures." 4 It would be hard to frame more specific assertions than these,
from both halves of the Empire and from the mid-second century on to the
230s, joining then with tales of saints in action in the second half of the
third century, and so on to the declaration of continuing powers by writers,
eastern and western, in the reign of Constantine. Such testimony suggests
how helpless people felt before mental illness and how inevitably they
turned, therefore, as they did also for the cure of bodily illness, to a source of
superhuman aid. It duly met the needs of Christians. But while churches,
high in their hierarchy, had designated exorcists, pagans lacked any cor-
responding deities, lacked temples known as places of recourse for the
possessed, on the model of Asclepieia, and had instead to trust to luck or to
some not very respectable help bought in the shadows.
Therein appears a point of weakness in paganism, perhaps inexplkable;
also a point of division in the spectrum of pagan beliefs: roughly corre-
sponding to our word "religion" was the sum of those things to be sought
 1. Needs and Answers· 51

openly in sacred precincts. Other things-a winning horse, sexual pleasure,

revenge, foreknowledge to be used against one's adversary, and a variety of
trivial conveniences of the kind the Sorcerer's Apprentice tried to master-
all belonged to the realm of magic. 5 For whatever reason, exorcism was
included. The point of division was fixed by moral considerations, not
exclusively but more than by any others. When Apologists charged that
among their persecutors could be found some tolerated pagan who prayed
for the death of a fellow being, meaning a gladiator, the reproach proved
the rule. Whatever Christians might believe, few pagans thought that
gladiatorial combat was immoral, therefore few would have hesitated to
send up a gladiator's prayer. 6
What pagans did pray for-their needs engendering and giving outline
to the divinity they thought they could discern above them-was health,
first. That is the best and most easily proven domain of supplication. Others
can be illustrated out of the very small minority of votive inscriptions that
specify why they were set up. There was beauty, close to health: thanks "to
Minerva for the restoring of hair"; and prayer for new life, "to Leto, that my
wife bear a child"; also "to Zeus Helios the Great Sarapis, savior and giver
of wealth"; "to Silvanus, from a vision, for freedom from slavery"; for relief
from tax payments; many vows for protection from natural disasters like
lightning and earthquakes, or rescue from drowning or (most often) from
unspecified "great dangers," for example, on a journey; for protection from
one's enemies or the safe return of a squadron from patrol; for safety-and
booty; or "for the safe-being of the colony and its senate and people ...
because he [Jupiter Best and Greatest] by his numen tore out and rescued
the names of the decurions that had been fixed to monuments by the un-
speakable crime of that most wicked city slave." Which returns us to magic
distinguished from prayer. Some wretched streetsweeper had exercised his
right to strikeJ7
It is disappointing that habits of inscription should not have producfd
in our period of study a richer body of material; but the sampling just
offered is useful in measuring the reach and variety of prayer; and there
remain several categories for mention which, if they indicate nothing un-
expected, at least define the center of our subject, human needs. Those were
the needs, beyond health, which touched an individual's household, do-
mestic animals, and crops. The first member of that trinity occurs in vari-
ants of the formula, "for himself and his" (i.e., his or her people or kin or
the like). 8 The prayer is so often expressed that, in Latin, it reduces to an
abbreviation, a mere four letters. The fact signifies as much as a thousand
texts in full form. As to the second member of the trinity of needs, inscrip-
52 Debatable

tions speak most often of the essential plow oxen, but of other beasts that
may be vital, too. 9 Almost none of the evidence, however, comes from Latin-
speaking areas; and, in the Greek-speaking, what survives is very little and
disproportionately found in Phrygia. As to the third member, there is no
relation between the importance of winning one's food, on the one hand,
and, on the other, the likelihood of prayers on that score being recorded
epigraphically. A god or goddess of fertility does not preside over every
regional pantheon-surprisingly, since agriculture certainly presided over
real life. The nearest approach is Dionysus, guardian of the vine that classi-
cal myth proclaimed him and addressed in that capacity just often enoJ.lgh
to meet our expectations. Perhaps folk closer to the soil, and further from
writing, held to faiths totally unrecorded. 10
Other matches exist explicitly or by implication in prayers to Mercury-
Hermes from travelers, traders, and market supervisors; to Silvanus from
cowherds and hunters; to Poseidon, from those spared in his shaking of the
earth; and to Zeus from those spared by his bolts, the gods being seen in the
actions assigned to them by all the handbooks. There are special connec-
tions we do not understand between young women and Hygia in her
temple near Sicyon, as between children of both sexes and Mars Iovantu-
carus near Treves.n Of spheres of divine functions, indeed, a great deal
more can be sensed than demonstrated. More still was probably left uncer-
tain in the minds of worshipers. Warning has already been given against
supposing that Apollo or Isis could not heal so quickly as Asclepius. And
Artemis in Cyprus or Perge, "national" deities that reigned alone, must
have answered prayers for aid of every sort, not only within the bounds fixed
by Greek convention. Jupiter and Zeus could do everything, witness whole
lexica of epithets tacked on to their names.
When we survey the full range of human requirements, then, from the
most central to the peripheral, looki'ng for a logic that might explain the
origin and vitality of religious beliefs, we discover the expected lines of
connection quite broken and twisted or, more often, to be drawn only by
conjecture. We are to blame the loss of evidence, not the inadequacy of the
whole system of cults in themselves. It is not to be supposed that comfort
was not somehow afforded to all the various causes of misery in the human
condition.
It is easy to sense, and worshipers certainly make no effort to hide, in their
relation toward the objects of their worship, a spirit of self-interest, of do ut
des and contract. That spirit has in the past often drawn reproach. It still
does-witness most recently the critic who contrasts "the Isiaccult[which]
possessed a moral system purer, more elevated than that of classical Roman
 1. Needs and Answers

religion. Who did not feel his soul shaken with feelings of admiration as he
watched those votaries of Isis plunge thrice into the frozen Tiber and,
shivering with cold, creep about the temple?" Reference is offered then to
the source of the picture, Juvenal.1 2 Juvenal did indeed have strong senti-
ments about it: he found it all ethically and esthetically repulsive. And to
the notion of the Christian Lactantius, that God could not be expected to
love worshipers who did not first bring their love to Him, a pagan might
have answered that what mattered was rather the service that the deity could
provide, since a god (as Aristotle had long taught) could feel no love in
response to that offered. The modern or alien critic of prevailing pagan
views, judging the reasons for worship, starts from assumptions that are
only his. Their own, pagans found quite comfortable.
Among felt wants, the modern observer expects to find none sharper than
the need for life, promised for ever. But, like a deity to insure good harvests,
assurances of immortality prove unexpectedly hard to find in the evidence.
Even the longing for it is not much attested. What can be gleaned from the
several kinds of source has been often discussed. The chief features need
only brief review.
People belonging to one or another of a small number of cults, and it:t
small groups, sought further lessons in their beliefs, lessons learned through
rites designed to catch the imagination and arouse awe. Impressiveness of
presentation could be heightened by rules forbidding the lessons to bf
talked about with outsiders. Obedient secrecy of course obscured the his-
torical record forever. One group, nevertheless, in the worship of Dionysus,
can be faintly discerned through inscriptions, developing more formal
ceremonies of instruction, at least in Italy, in the later second and early
third century. During the ceremonies, participants may have received prom-
ises of afterlife. But evidence for all this is unfortunately very little and very
indirect. u Similarly with Isiacism: the evidence lies in the concluding
chapters of Apuleius's novel, in which his hero Lucius undergoes a lengthy
and most expensive course of instruction at the hands of Isis's priests. At the
end, he is fully satisfied by her promise, "You shall live in blessedness,"
vives beatus; and when life is over, he may continue to worship her. He is
the envy of everyone for being renatus, reborn "in a sort of way" -defined
as having earned the goddess as his patron and at the cost of no more than
temporary bankruptcy. There is, however, no word of his being renatus in
aeternum, which is what counts.l4
As to Mithraism, it grew like Syrian cults, cell by cell. At meetings in
underground chapels, members dined together, sang hymns, and from time
to time participated in rites that brought promotion through a series of
54 Debatable

titles and roles. One line of a graffito on the walls of the Santa Prisca
Mithraeum in Rome, almost illegible, seems to read rebus renatum dulci-
bus atque creatum, "reborn and created for delights"; others tell how "you
have saved us(?) by the shedding of eternal blood." "Saved" for what? "Re-
born to eternity through secret baptism," answers an inscription-a for-
gery! 15 Nor do Celsus and (a century later) Porphyry serve to clarify the
matter much. The one speaks of beliefs in the passage of Mithraists' souls
through the seven spheres of the planets, the other, of passage after death
into some different living being. 1& Which is not to everyone's taste in im-
mortality, even if it could be reconciled with the few other fragments of
belief that survive.
When Sabazius cult by the mid-third century had come into contact with
Christian doctrines in Rome, one of its priests depicted his departed wife in
a fresco, banqueting among the blessed. There in that scene was immor-
tality clearly portrayed;!? but only a part of the portrait, probably a very
small part, could be considered pagan, to throw light on unaltered eastern
ideas. A mile away, across the Tiber, a certain Gaionas, servant of a Syrian
shrine there, died, so his epitaph declares, "paid in full to death-Gaionas'
little soul." 18 So flippant a farewell ill suits the moment of rebirth into
eternal blessings. The inscription cannot be used to support any broad
interpretation of afterlife in the minds of the inhabitants at the center of our
time and area of study.
Other scraps of evidence survive without adequate context. There are
indications of inner circles or more advanced degrees of revelation in a few
Greek cults, but what was won by them cannot be discovered. Part of their
value lay in their being reserved for special seekers, and their deep secrecy .1s
When Theon of Smyrna, some time in Hadrian's reign, speaks of initiation
rites as bestowing "a blessed state of divine grace and companionship with
the gods," it is not clear what rites he has in mind; when Plutarch, close to
the same decade, speaks of finally joining "the company of the sacred and
pure," "the fully liberated and released" as opposed to "the uninitiated,
uncleansed mob of the living," he does not explain what he means by "The
Great Rites," the ones producing these results, whether they are the Eleu-
sinian or some other; and Tertullian discusses rites of advanced instruction
through several different phases or moments, as do the preceding two
authors, but in a jumbled paragraph which fails to show what belongs to
Mithraists (named), what to various other unspecified mysteria. "He both
celebrates the offering of bread and introduces a likeness of resurrection,"
runs one key sentence. The last word may as well mean "of the new man,"
reborn in his life like Lucius, as "of the man that dies." 20
 1. Needs and Answers

From all three authors' pictures of initiation, what seems to emerge in

clear outline is a considerable clientele-let us say, many thousands at a
given moment, but not tens of thousands, since our glimpses are only of the
educated classes-in the Greek-speaking world, among whom prevailed
some faith in an agreeable afterlife earned by attention to ritual purity; and
in Tertullian's world, perhaps similar views more localized in region and
class, reflected also in the epitaph of the departed Julia Benenata: "For I,"
she declares, from her tomb at Mactar in Numidia, "who always lived in a
pious body, inhabit, thanks to divine law, the sweet Elysian Fields of
Proserpina, that is the sun whom I have seen on high, with the constella·
tions." Echoes of Vergil? And surely of Christianity, now, in the late third
or fourth century? Of Isiacism, too, or Neoplatonism? But the ideas are
widely shared commonplaces; whether literary or genuinely religious, it is
very hard to say; and (returning us to an eastern setting, possibly Egyptian)
there remains the one generalization that survives from our period, inserted
casually in a religious tract: "as hoi polloi believe, when the soul leaves the
body, it turns into an anima1."2 1
In all the "Oriental" cults in general, whether of Atargatis, Mithra, Isis,
or Cybele, the element of resurrection has received emphatic attention in
studies old and new-attention emphatic but not always firmly controlled.
It should really not be taken for granted, as it often is assumed, that people
who believe a god might rise from death also believed in such a blessing for
themselves as well. The conjecture needs support-and finds none.22 More·
over, in one of these Oriental cults, though the death of Attis is primordial,
there is no mention of his resurrection in sources before the end of the
second century at the earliest. 23 A second feature of the same cult, ritual
cleansing in the blood of a bull, has been taken as a sign of belief in the
possibility of life after death. What it was actually thought to grant, how-
ever, was an extension of one's earthly existence, in a state of ritual purity,
and then only for a limited term of years.24
On the views of death most commonly held in the world of the Apolo·
gists, there exist one or two statements of quite broad import by contem·
poraries. They should be brought forward as a sort of summary to the
positive evidence assembled in the preceding pages. Justin (to begin with
that figure of the mid-second century) declares that pagans do indeed en·
vision some afterlife, witness their belief in magic (he means, practiced at
tombs upon the spirits of the deceased) and possession by demons. But, in
saying this, he chooses peculiarly roundabout ways of substantiating his
point, at least, if the Elysian Fields had been much talked about. He goes on
to appeal to statements, unspecified, from the oracles of Amphilochus,
56 Debatable

Dodona, Pytho, "and many others." That too is not probative, because of
its vagueness.2s Lactantius can do a little better. He cites not only anony·
mous philosophers and seers but a particular reply from Apollo at Didyma.
"Does the soul survive death?" Yes, says the god, it is borne into the aether,
there to exist forever. The passage has been quoted above.26 Jt is as interest·
ing for what it does not say as for what it does: it instances no easily recog·
nizable, universal, or at least very familiar deity to name, whose followers
all trusted in his power to save them from extinction. Nothing like that
existed.
In striking the true balance of the evidence we should return, too, to
Plutarch's circle. How pious it was can be sensed from many conversations
he records or imagines. One of the most forceful is inspired by the subject of
Epicureanism, naturally enough, since that was the chief stronghold of
atheists. During the exchange, one of the company wishes to underline the
delights and rewards of piety. "No visit," he says (IIOlE), "gladdens us
more than a visit to a shrine, no season, more than a festival, nothing done
or said, more than what we see and do in regard to the gods, whether we are
present at secret rites, or at dancing, sacrifices or initiations." But if those
last had conferred immortality, surely they would not have received so
casual a mention, added by afterthought to such agreeable diversions as one
might find at a church picnic. Initiations, teletai, can have meant nothing
more soul-shaking than those big, open spectacles encountered in the first
chapter, above, as they unfolded in theaters. Spectators received no special
promises.
The pagan Celsus takes Christian ideas of resurrection as being, at best,
metempsychosis misunderstood; at worst, ridiculous; and the pagan Caeci·
lius, in Minucius Felix's dialogue, agrees.27 The pagan Seneca rejects with
scorn the whole Greek afterlife, Ixion, Cerberus, and the rest, in favor of the
disembodied spirit reserved to a second, Stoic existence within the very
heart of light. 28 He represented the cast of mind most often to be found
within that class that gave its thought to abstract questions; he represents
its strengths, also its weakness. For, with no very significant or prominent
exceptions, people of all directions of belie£, to the extent they could con·
ceive of life continuing beyond the grave, did so in terms of the divine
spark, as we might call it, the disembodied soul. That was very much a
philosopher's view, one of those convictions one may hold without always
remembering one holds it. No one cared greatly that he might gain eternal
life if it were not really he that gained it, rather, some animula, some particle
ephemerally spun off from the Great Soul, or the like. What was felt to be
essential was one's true self, a personality; and, in default of that, Stoicism
 1. Needs and Answers

or similar ideas of immortality had little to do with people's most earnest

longings.
Inscriptions here as on other points hold out the best hope for a broad
sampling. "Savior" in them, or "salvation," had to do with health or other
matters of this earth, 29 not of the soul for life eternal. Or in epitaphs, people
so often joke about annihilation that the jokes at last congeal into com-
monplaces or abbreviations: "I was not, I am not, I care not," boiled down
to six letters. 50 Or, last, in rules of cult associations, even those intimate
gatherings of the faithful around a table that so instantly recall the Eucha-
rist- but wrongly: the most telling note is struck by the injunction, "do not
vomit up your wine." If there is no absolute certainty that banquets were
not thought of in times past (as they have been so often by modern inter-
preters) as conferring immortality, there is no scrap of testimony to such a
thing; and the silence itself is quite extraordinary. No, at festal meals, as at
teletai of dancing and music, the wine, women, and song satisfied wants
more simple and secular. 51 At Panamara, at the two-day Zeus festival, the
crowds invited from far and wide to what were officially termed "mysteries,"
and who there received the wine from the priest at solemn meals, enjoyed a
fellowship with the gods among them-that may be assumed, that would
be typical of such cult meals- but no life promised forever after.
The aim of the foregoing discussion is the quite elementary one of estab-
lishing the qegree of correspondence between common emotional wants
and the satisfaction supplied by religion. That aim it seems natural to
pursue first on the level of the individual, but the individual multiplied by
millions. Thereby we should gain some sense of the vitality o£ paganism-
key word in this second chapter.
Before resuming examination of individuals' feelings, however, it might
be best to consider the satisfaction offered by religion in larger contexts. It
was certainly recognized throughout antiquity, at least by people able to
look at their world with any detachment, that religion served to strengthen
the existing social order. Explicit statements to this effect, and very interest-
ing ones (because the juggling of high principle and selfish advantage
always rewards the spectator), can be found across a string of well-known
writers, who protest the crucial role of religion among the masses. Public
largesses such as were described in the preceding chapter-the enormously
expensive distributions of wine, liqueurs, meat, pastries, more wine, and
cash- should be made, says Plutarch, "on an occasion affording a graceful,
noble pretext, through honor to god that draws everyone to piety; for in hoi
polloi at the same time a strong belief in, and conviction about, how grand
the divinity is, and how august, is engendered when they see those whom
58 Debatable

they honor and consider great, themselves so liberally and zealously com-
peting in regard to the divine." 52 Which by no means convicts Plutarch of
hypocrisy. He had his faith and believed in its truth; hoi polloi had theirs,
and he believed in its usefulness.
Many gods in one particular region, Phrygia, could be counted on to
avenge themselves upon the wicked. That certainty was encountered earlier
(p. 32), along with similar ones thinly scattered in other quarters as well.
Most characteristically, it was the man forsworn on whom some punish-
ment descended. ss Most often, in both halves of the Empire, it was the water
of a sacred spring or river that scalded or melted away the flesh of the per-
jurer. But (magic aside, which does not quite belong to this book) the
wretch who disturbs a tomb is threatened, too, in epitaphs sometimes
found in Phrygia. One invokes vengeance from "the gods of the Roman
People," no doubt conceived to be supremely effective.'4 And frequent
abasement in Phrygian sanctuaries before a god whose wrath is feared finds
parallels occasionally elsewhere, both as an individual and as a general or
loose-floating belief. Isis, wherever she was worshiped, was thought to seek
out criminals and strike them blind or paralytic, Sol-Helios saw their every
misdeed, and the pagan Celsus professed faith on behalf of himself and
others that the wicked would be made to suffer. So one is taught, he says, in
undergoing initiation into the mysteries. He means, of course, sufferings
on this earth as distinct from the pains of hell fire. 55
A second distinction must be kept in mind, too: between misdoings
defined in terms of laws divine or human. In terms of the former, offense lies
against ritual purity or the sacredness of a god's own dwelling. Perjurers
will be caught only if they sin in the very face of the god, the virgin-not-a-
virgin, only if she immerses herself in sacred waters. When people expect
divine vengeance, it is usually in circumstances of this sort, not through
supposing that the human and divine sense of right and wrong will be
identically offended and that retribution will thereupon reach out to catch
the criminal wherever he may be. Of such a grander faith there are very few
signs indeed, at least in cults untouched by Judaism. Phrygia provides an
exceptional inscription; Lydia and Lycaonia a few more, nearby; and there
are further scraps and hints in Hermetism and Mithraism. 36 What emerges,
however, are glimpses of an expectation of punishment-that is, guilt-
attached to no one cult in any significant manner, rather little talked about
or counted on, and generally felt in regard only to basic tabus: parricide, for
instance, and incest. With these conclusions we may return to look a second
time at Plutarch's views. They represent the end of a string of statements on
the social usefulness of religious faith, going back to the fifth century B.c.
 I. Needs and Answers 59

Perhaps not chance but change of perception and diminishing detachment

accounts for their being almost the last. And Plutarch does not, after all, say
exactly what his predecessors had said (that religion keeps the masses
down, or honest); rather, that the masses' faith should be stimulated so as to
direct their thoughts upward in reverence for nobility: to be found among
the gods, found likewise among the local-nobility.
Naturally, he could also see the practical value in having as many people
as possible discern a sort of harmony and mutual respect between the gods
and the city magnates. We have earlier heard an orator inventing and pub-
lishing in a theater certain wonderful visitations from above, in favor of
this or that sitting magistrate. 57 Cicero told the Roman Assembly that
Pompey deserved a high command, being born to it by divine plan, divino
quodam consilio natus esse videatur. He made the proposition incontest·
able with his favorite clausulass_not that he himself believed a word of it.
We might accuse him, and his brother orators in the east, of empty compli·
ments to which not even the most credulous audience could have attached
much meaning, were it not for the evidence of private supplications on the
people's behalf-that is, by a citizen for his village, city, or neighborhood.
It is hard to see what gave rise to these other than a perfectly sincere convic-
tion, a certainty that the gods did care about the whole unit which offered
them its cult. 59 More proof of this sincerity appears in plague times, when
city senates tried to enlist the direct help, or at least the advice, of some god
in their salvation. •0 Again, we ourselves may be skeptical. In desperation,
any measure might be tried, however little credited. Yet some of our inscrip-
tions, to say nothing of Aristides' report on his private efforts, declare after
the event that salvation had been really won by prayer: plagues or earth·
quakes had been ended, unspecified benefits bestowed, the Goths scattered. 41
That partnership, mutually respectful, between the powers above and the
leadership below, was seen actually to work.
But how, precisely, did it do this? By what means could any contact be
established with the beings of another world?
From testimonies to a living faith (which we are not done with yet) we
must turn briefly to its quite mechanical functioning. Aside from rare and
rarefied speculation on sympathetic magic, inhabitants of the Apologists'
world thought first to touch the gods through images, because that was
where the gods lived; or at least, to images they could be brought by
entreaty, there to listen and to act. Whether or not they fitted exactly, whether
they looked like their portraits in stone or wood, they were to be found
inside. Christian observers, who had no reason not to be accurate, report
this as the generally prevailing idea. 42 Moreover, three Greek pagans at·
60 Debatable

tribute miracles to statues of men who lived in their own, the second, cen-
tury. One statue fell over and killed a man and then, being punished by
drowning, got itself fetched up again from the sea. A second provided cures
to suppliants. A third and fourth respectively provided cures and oracles.
And a statue of Antinous in the city renamed for him wrought unspecified
miracles, as even a Christian could concede. So ready were people to attrib-
ute powers to images even of mortals, so ready to see the divine even in their
fellows! 43 A final glimpse takes us to Palestine in the fourth century, there
to find women consulting the image of Aphrodite on their choice of hus-
band, or the sick cured by the weeds that grew at the base of a certain bronze
statue. Local inhabitants, questioned by a touring bishop, regretted they
could not identify it. They suggested, however, that it looked very much
like Jesus. 44
There was a second means by which contact could be made with the gods,
beyond their images. That lay through sleep. In antiquity, dreams as such
constitute a very large subject, studied for centuries by learned men-their
handbooks published, passed on to the next generation, and so fattened
into miracles of sound scholarship under the names of Artemidorus, and
others. But that is all by the way. So too are the interpreters that served Isis
or Asclepius and their customers in dormitories, another large subject in
which specially figures the name of Aelius Aristides. More to the purpose
are the sheer numbers of testimonies to the appearance of divine figures
before the minds of sleepers. 45 They introduce an extraordinary variety of
gods and goddesses spread broadcast over all provinces, acknowledged
typically in the form, "To Such-and-Such, from having seen," or "having
been advised" or "commanded," or "by order" or "in a dream." The formu-
las occur also in literary sources that describe nocturnal revelations. 46
When a man died, he might somehow earn promotion to demigod and, as
"hero," could appear to worshipers in their dreams. So could the Genius of
a given legion. The two examples convey how quickly and carelessly the
practice of consulting deities through sleep might spread, even to the most
newly invented inhabitants of the heavens. 47 Corporate bodies might act on
dreams as well, presumably not vouchsafed on one night to every member
of the town council. 48 1t was their priest's experience, perhaps, and then
interpreted to them? And sailors off the north coast of the Black Sea could
count on the hero Achilles to tell them in their sleep just where the best
anchorages lay. That was common knowledge; visions of the Dioscuri seen
by sailors, also common; but less well known, thatPatroclus, too, might be
seen in local waters. All this Arrian discovered in his tour of the area during
Trajan's reign.49
 1. Needs and Answers 61

Famous only in their own region, these heroes' visits from above serve to
correct the impression that might be gained from Asclepius's visits, famous
in every region: it was not only in the latter's hundreds of shrines that
people counted on contact with the divine through dreams. Origen, exter-
nal observer and therefore to be heard with special attention, declares that
"it happens to lots of people that a dream reveals to them that they should
do something" -instructed by an angel or some other means, he adds. He
does not intend pagans alone. Celsus, too, is quoted, "that someone in a
certain condition, dreaming or out of his own wish, may have an experi-
ence, an illusory vision that carries some message to him, as has happened
to tens of thousands." 50 "Out of his own wish" -evidently, then, just as
one could have them at Asclepieia (else why venture the journey?), so one
could turn them on at will in any temple. "I have come to see a dream that
will give a sign to me on the matter about which I am praying," reads an
inscription from a shrine in the Greek-speaking world. And from the
Latin, the incident in a novel in which a woman is sick: "and for that reason
I inquired for a treatment through dreaming; and I was commanded .... "5 1
While there are scattered indications that experts might be called in to the
interpretation, that practice was a local matter and not prevalent.52 Nor-
mally, one helped oneself. Iussus, "I was commanded," is so frequently a
boast as to be abbreviated in Latin inscriptions, v(otum) i(ussu) d(ei)
f(ecit). 55 In Greek inscriptions, equivalents are, if anything, more easily
found. And the only deity whose votaries seem to enjoy a detectably greater
access to his will (such is the impression gained from inscriptions) is Jupiter
Dolichenus, from the east, but probably better known in the west. The peak
of his popularity belongs just in the middle of our period of study.
In the world of the Apologists, contact with the gods lay most often
through signs and scenes longed for and granted to the suppliant in sleep,
thereafter in waking to be interpreted. The process served pagans as prayer
served jews and Christians. It would of course be quite parochial to sup-
pose that, in every land and every age, the plea for help can only follow
certain forms or not exist at all.
But a third means of contact, neither through idols nor dreams, joined
worshipers to the objects of their worship, namely, oracles. This is the
means most familiar. It is characterized by the interposing of a third party
(not invariably but almost always needed) between the two that must com-
municate. A priest, priestess, seer, or spokesman of some sort gives the
deity's answer to spoken or written questions. Plutarch wrote a well-known
essay on the decline of oracles, meaning in central Greece. Very sensibly, he
attributes the change to diminishing wealth and population, though in
62 Debatable

nearly the same decade the Pseudo-Plutarch declares that "prophecy enjoys
honor among all men." 54 Lucian, noting the decline at the center, sets
against it the high repute of other oracles to the east, where there are various
indications of a readiness to credit them, in other sources, into the third
century: Amphilochus in Cilicia, Apollo at Miletus, above all Apollo at
Claros.55 The veracity of oracles, many of whose responses were locked in to
past history and so to speak immortalized by Herodotus and later writers,
constituted a natural point of debate not only among various philosophic
schools-Cynics and Epicureans especially expressed their disbelief-but
between pagans and Christians. 56 That debate suggests continuing vigor in
the institution and in the credit on which it rested, regardless of the popu-
larity of any particular center. More obviously, however, it brings to the
fore an element of dissent within the body of polytheism.

2. THE VITALITY OF PAGANISM

The health of that body we must now begin to examine. It is convenient to
look first at the group just mentioned, Epicureans, who represented the
furthest extreme of disbelief. But aside from their name's being attached in
obloquy to anyone who doubted a local oracle-"Atheist or Christian or
Epicurean"- they hardly appear in the sources for our period. 1 It is rather
their company that calls for comment: they are close to Christians, and that,
in mid-second century Pontus, was not good. They are close to atheists, and
that was a great deal worse. In public, as many statements make clear, to
deny the reality of the gods was absolutely unacceptable. You would be
ostracized for that, even stoned in the streets. 2 There is no need to recall the
scenes of persecution in Smyrna, Rome, and Lyon in which the mob rails at
Christians as godless. 8 An Apologist acknowledges his opponent's feelings:
you "believe that to worship no god at all is impious and unholy." A con-
temporary, but a pagan, Aelian, condemns the atheist who, with alien
"wisdom," disputes whether the gods exist or not, 4 and Lucian, a half-
century later, declares that "the great majority of Greeks and all the non-
Greeks" (which includes Romans) are believers. 5
To have the same author, in contradiction of himself, tell us there were
no worshipers left to Zeus any more is certainly puzzling; puzzling, that our
learned friend the Epicurean of Oenoanda, Diogenes, should have dared
instruct his fellow citizens in the freethinking of his sect, spellesi out across
the most public walls in town (above, pp. 10-11); and puzzling, that the
same ardent Romans who burnt the "atheist" Justin should, in Juvenal's
day, have laughed at anyone professing faith in an altar or temple. So the
 2. The Vitality of Paganism 6J

poet tells us, anyway.s Pompeian walls display dozens of blasphemous

graffiti, insults to Venus (patron deity of the town), or, in a tavern, an
obscene painting at Isis's expense. We may take their like for granted
elsewhere, if there were other sites so well preserved.? In Africa, down the
coast from Carthage, a rich man commissioned a semipermanent, semi-
public bit of humor: a mosaic wall with figures and captions making fun of
the gods at their feasting. 8
This is the kind of mixed bag of facts that renders general statements
about religious life so hard to frame or so easily criticized if they are framed
too narrowly. What, for instance, can one make of the assertion that oracles,
"it is true, enjoyed a recovery in popularity in the second century"- for
which a single inscription is cited, recording help sought by a city in
Sardinia from Apollo in Claros? 9 Such characterizing of the feelings and
thoughts of fifty million people on any day out of thirty-six thousand has
something ludicrous about it, as if one were to measure the pulse of the
western world on the basis of a single headline in the St. Albans Sentinel.
Worse than that, perhaps: since religious feelings are not something to talk
about in public, in some of their aspects, they must prove all the harder to
assess from the outside. The more need for care.
Let us consider a few further contradictions suggested by the important
role of inner emotions in religion. Lactantius tells pagans to their face (and
there is a sort of reliability in such testimony, out in the open and instantly
subject to challenge by contemporary readers) that theirs is "a superstition
about those gods of theirs ... in which, for all I can see, there is no more
than worship by the finger-tips ... , nothing required but the blood of one's
flocks, and smoke, and foolish libations." In contrast, Porphyry declares
"the one who loves god cannot love pleasure or body; but the latter sort of
man will love money and so be unjust, and the unjust man is unholy, both
toward god and his ancestors, and a criminal in his conduct toward others.
So he may sacrifice hecatombs and adorn the temples with a myri,ad oifer-
ings, but he remains impious and godless and, in true calling, a sacrilegious
person." 10
Or we may turn back from the later Empire to the earlier, there to read,
"Vesta, be gracious! To you we now open our lips for worship-if indeed
we may join your ceremonies. I was completely absorbed in prayer, I was
aware of the divine powers; and the earth, joyful, shone back with a dark red
glow." And another's assertion that "all men feel a powerful longing to
honor deity and pay cull from close up, drawing near and seizing hold with
persuasion, offering sacrifices and crowning with wreaths. Just as tiny
children, torn away from father or mother, feel a terrible longing and de-
64 Debatable

sire, and often reach out their hands in their very dreams to the absent ones,
so to the gods, men who rightly love them for their beneficence and kinship
are eager to be and to talk with them by any means." II
Moving passages, reminders to their audience, western or eastern, of
emotions shared in the innermost heart! But consider for comparison the
scene of devotion that Seneca painted (above, p. 44), in which simplicity of
feeling and intensity itself are derided as madness. And in other pages too
long to quote, leading up to that scene, he dismisses the dancing or howling
or bloodied priests of Cybele or Isis or Bellona, all of them as revolting as
they are demented-the very manifestations of "true" piety that modern
students most often point tol 12 Even Porphyry's plea for god and ancestors,
almost equating the terms of the two, and sounding a note so exactly in
tune with all those praises of one's inheritance that were heard in the
first chapter (above, p. 3)-even Porphyry's plea Seneca denies head on:
"What of the fact that we even join the gods in marriage, and dishonorable
marriage at that, the marriage of brother and sister? We give Bellona in
marriage to Mars, Venus to Vulcan, and Salada to Neptune. But some we
leave unwed, as if no match could be arranged, especially since some are
widows, such as Populonia, Fulgora and the goddess Rumina. I am not at
all surprised that there has been no suitor for these. As for all this obscure
throng of gods, assembled through long years of ancient superstition, we
shall invoke them, but with the reservation in mind that their worship
belongs rather to custom than truth." The repudiation of mythology here
is echoed by other writers as well.1s
Enough of such contradictions-which, useful though they may be to
indicate the bounds of the possible, chiefly demand resolution. It cannot be
found, however often sought, in the change from one century to another-
from "the religious crisis in the Roman world at the end of the Republic
and beginning of the Empire" to "the second century with its powerful
upsurge in religious life," then the mid-second century to the early fourth,
all one "age of anxiety," in its later phases coinciding with "the decline of
paganism in the second half of the third century." 14 Conventional and
representative of a sort of scholarly consensus regarding successive develop-
ments within our subject, that succession of periods must nevertheless be
looked at critically. All of its parts are made of, or meant to describe, religi-
osity, not only the whole nexus of thoughts and feelings about the divine,
but their intensity or vitality as well. For the final period of decline, sub-
stantiation is sought in the much-diminished epigraphic record, not in
what it says but in its very silence. The silence itself is a fact plain and un-
contested, though its interpretation is something else again. We will return
 2. The Vitality of Paganism 65

to that matter, later. But for the other periods, the sources used are almost
wholly literary, since "inscriptions seldom tell us much about the under-
lying personal experience." 15 Here too is a fact uncontested, witness the
passages called on in the last few pages to convey a sense of what emotions
characterized moments of active worship. They are drawn from a meta·
physician, Porphyry; a poet, Ovid; a rhetor, Dio Chrysostom; and still
others that display or suggest deep attachments or experiences might be
found in a novelist and rhetor, Ap!-lleius, in a natural philosopher, Aelian,
or in other poets. 16 Inscriptions provide no match.
There are, however, two objections to be raised against reliance on such
authorities. First, they represeQt only a small group. The fact has been
stressed before. Summing up a good deal of discussion, it is stressed again,
below (p. 88). Second, even that small group they do not represent accu·
rately.
To examine the latter objection: as is clear in the opposing pairs of quo·
tations ~hat were just presented, the religious feelings most common in the
Empire, and those considered suitable or "true," covered a very wide spec·
trum indeed. Explanation lies in plain human variety, not in place or
period. Moreover, the range indicated in the comparisons could be easily
extended by the inclusion of prinking and frivolity in sacred processions,
erotic or abnormal titillation in cult groups (or in novelistic fantasies about
them), exaltation roused by concentrated and emotive word pictures de·
scribing the great powers and loving kindness of a god-and so forth. Ex-
amples of all such passages in the literature of our two centuries of study
have been offered above. 1' Except from a sectarian point of view, then, it
would be wrong to pick out one or a few particular dispositions as charac·
teristic of paganism.
Indeed, it is my
I
guess that the variety of feelings satisfied in the practice of
religion differs inconsequentially from age to age, however much accepted
patterns of behavior to receive those feelings change in shape; for, as Nock
wrote, "there are fashions in religion as in everything else." 18 If my guess
should be righ4 religiosity in any society as a whole would likewise differ
inconsequentially across time, and the several phases or periods which it is
conventional to distinguish in the religious history of the Roman Empire
would lose all meaning. Whether one age is "more religious" than another,
that is, whether it actually stimulates its inhabitants to significantly dif·
ferent levels of emotion in the service of their god or gods, is a question
almost too important to leave alone; but surely even those who choose to
offer an answer to it for ffieir own living generation, if they were really
pressed, would admit they had no evidence anvwhere near adeauate in
66 Debatable

scope or weight to support any opinion. Hope of discovering the truth

about a remote and ill-documented period it would be foolish to entertain.
Let us rejoin the circle of writers from whom we turned aside a moment
ago. None in the present connection is better known or more certain to be
referred to than Marcus Aurelius, the Stoic emperor. His Stoicism, how-
ever, stands out in isolation. There are no signs of a similar devotion in the
preceding half-century. Suppose that we knew nothing of his meditations,
or that another spiritual diary of the same bent but from a half-century later
were discovered, would historians have any difficulty in accommodating
either possibility? Would either be clearly at odds with its context? Surely
not. There is a quite inadequate, and therefore indiscriminately welcom·
ing, texture of evidence across the whole period of this study. Accordingly,
the emperor's devotion has human interest but not historical. Even-or
perhaps one should say, especially-a careful examination of the man and
his time must conclude by admitting that we do not know to what extent
he typified his own aristocratic circle. 19 The same admission should guide
the student of Plutarch's circle, earlier, or Porphyry's, later.
Marcus Aurelius and the other names most likely to occur in a book
about the cults of the Empire are in any case self-selected for the interest
they felt in their own spiritual life and things divine. They wrote on those
subjects. Not everybody did. Some people are simply more given to belief
than others, like Teucrus of Cyzicus, who journeyed to Pergamon in search
of relief from epilepsy. There Asclepius "appeared to him, and they struck
up a conversation, and he developed a quartan fever, and through it recov-
ered from epilepsy." They "struck up a conversation"- extraordinary! Or
D. A. Lazurkina at a congress of the Communist party not many years ago,
who reported, "Yesterday I consulted Lenin again . . . and he said to
me.... "2o
There is epigraphic confirmation for the point, perhaps. Among vows
paid to the more prominent deities, where some very rough statistical tests
are possible, and particularly among those vows that depart at all from the
bare formulas of dedication, dedicants who identify themselves as servants
of the god-priests or the like-are present disproportionately, in com·
parison to ordinary dedicants. Some of them are quite lowly. 21 The cult of
Asclepius at Epidaurus presents a striking case: the sanctuary is filled with
scores of altars set up by the priests themselves.22 Another illustration lies in
the cult of Jupiter Dolichenus, whose priests made themselves exceptionally
prominent. 25 It was naturally to their advantage to celebrate the deity who
fed and sheltered them; no less natural, we should grant them, to feel a
genuine faith. That faith prevailed over advantage is suggested by those
 2. The Vitality of Paganism 67

priests of one deity who offered tribute to another. 24 Taken as a class, they
were surely representatives of human nature at its most devout.
It would not be too cynical, but simply wrong, to attribute to the profes-
sional servant of the gods real doubt about their existence or efficacy; for
most of the most highly admired and educated people in the world about
him were quite at one with his convictions-at the top, the emperor Marcus
Aurelius, who doubted wonder-workers but credited healing dreams, above
all, those sent by Asclepius. Everyone agreed: Marcus's teacher Fronto, for
one.2s But there is no need to call up from earlier pages the proofs of un-
feigned belief to be found in virtually every writer of the second and third
centuries whose works survive. They all believed; which cannot be expected
to mean that they all believed everything. The variety of religious convic-
tions or systems (to which we will return) matches the variety of emotions
regarding the divine; but the former can be examined with some hope of
understanding them. The latter, as has been said, lie beyond reach of
profitable discussion.
What is untypical about the religious views of the highly educated we
may identify by following them out beyond the surviving literary corpus.
Will that distort those views? Apparently not. Some literary works are
known only through copies on stone. Our sources epigraphic and literary
join without a seam. 26 But acceptance of the gods is as widespread in
inscriptions as is implied, for instance, by Lucian, when he counts listeners
to atheism as so few.27 We would have been misled had we given too much
weight to his own scepticism, or Seneca's. Here is one illustration of the
danger of relying only on a small selection of writers. For a second illustra-
tion: the gods ·addressed in inscriptions are described only in very broad
strokes and simple affirmations, in a shouted phrase, for example- "The
One," "Greatest," "Savior" -or in a string of epithets without integration:
"swift-running," "dwelling on high." 2s Examples of the style have been
offered in the preceding chapter. The resulting picture of the gods is oddly
pointillist, as if in acceptance of each one's birth and growth out of the
visions of all his votaries, individually, by quite uncontrolled accretion-
pointillist and motionless: it is rare that the gods are seen in action in any
episode of their story, though a life or biography many of them are indeed
known to have possessed. 29 The literary treatment of religion in this respect
also matches the epigraphic. After Ovid, back at the tum of the era, and all
the way up to Claudian and Nonnus at the edge of another great break in
history, writers show little interest in anecdotes from "mythology," as we
would call it. So it would be misleading to draw our own impressions too
much from certain exceptional passages in Plutarch, let us say.
68 Debatable

Literary evidence, at least in Greek, joins closely with one other type,
papyrological. There is no sharp break detectable in moving from the style
and interests that broadly prevail in one, to the other. Good reason, there-
fore, to trust to Egypt for some impression of the place of religion in litera-
ture-though that province was famed for being particularly addicted to
sacred lore and studies. Out of some thousands of papyri, whole or frag·
ments, that are of literary character, only a small number focus directly on
religion. Numenius, Ammonius, Plotinus, and Porphyry make no appear-
ance; other philosophers very rarely and for treatises on government or
ethics or logic. ao There is a Hellenistic scrap of Orphism, there are two or
three scraps of aretalogy, quite a few hymns and paeans (so far, the same
mix found in inscriptions, minus the votive), plus bits and pieces of fiction
describing scenes of worship. Throughout this little corpus of texts there is
a distinctly regional quality, to which Isis and Osiris especially contribute.
Had anyone demanded a larger share of the written word for piety, there
was an answer ready to meet him: all the big temples had their own libraries,
containing no doubt the same sort of prose and verse in praise of native
deities that still survives, but in much, much greater abundance; beyond
that, representing Hellenic culture in a broad sense, there was Homer. A
majority of all identifiable literary papyri are the two epics, or commentary;
and Hesiod is high in popularity, too. Those, it might have been said, were
the old and the new testaments of truest Hellenism.
In trying to understand the faith prevailing in the lettered classes, use is
commonly made of Gnostic, Hermetic, Chaldaean, Neoplatonic, Orphic,
or Neopythagorean texts. With favorite selections drawn from other genres,
they constitute the material from which the modern account of religious
feeling in our period of study is typically composed. They are, however,
given an importance quite out of scale with what contemporaries conceived
their faith to be. "The underlying personal experience" (above, p. 65) is
indeed far better found, at its very deepest, in Gnostic texts and the like. To
call this deepest level "religion" without further qualification misrepre-
sents the subject.
So much for the first objection that must be raised against choosing one's
sources to fit one's modern preconceptions-against reliance, that is, on a
small group of writers. The choice must rather widen to include experiences
by no means intense, but no less a fact; concerns by no means profound, but
no less sincerely expressed; and indifference or positive rejection almost as
strongly represented as high-wrought religiosity. This said, nothing is sub-
tracted from the subject. It is only taken away from a little, rather peculiar
collection of books, and restored to a very much larger circle of real people,
 2. The Vitality of Paganism 69

some of whom (like any sampling from the Canterbury Tales, Simplicissi-
mus, or War and Peace) felt religion at the heart always; most, at the finger-
tips, as Lactantius puts it; and a few, never at all.
The second objection lies against the presenting of religion in our period
only through the testimony of writers and readers. How few they were in
the entire population has already been emphasized. That they were dif-
ferent remains to be asserted.
Plutarch may stand for a type thoroughly accepted among the well edu-
cated and well-to-do. He had a broad range of friends and interests Roman
and Greek, even (but not untypically) reaching beyond, to write our only
extant essay on Isiacism. He counted as more devout than most. In his col-
lection of antiquarian oddments called Quaestiones Romanae, he takes up
about a hundred customs having to do with Roman religion. For example:
(1) Why do the Romans consider Kronos the father of truth? (Because they
suppose Kronos is Chronos, Time, and time discovers truth?) (2) Why do
they pay honor to the dead in December? (Because in that month all grow-
ing things have done with their lives?) (3)Why may the priest of Jupiter
never anoint himself in the open? (Because it is not decent for a son to strip
before a father, and Jupiter is the father of all?) (4) Why are slave women
barred from the shrine of Matuta? (Because her Greek equivalent, I no, was
jealous of a slave woman on her husband's account?) (5) Why may the King
of Sacrifices never address the people? (Because the Romans expelled their
kings, and wish to limit his semblance to cult acts?) And (6) Why is Rome's
guardian deity never named? (Because that protects him from being con-
jured away?) 31
The answers he offers to his questions he arrives at through six charac-
teristic modes of thought. They lie in punning, in symbolic or esthetic
appropriateness, in analogy with mortal customs, in echo of myth or a
deity's life story, in historical accident, and in suitability for achieving
some result, through binding or pleasing a deity. Only that last could be
called functional, that is, only in that category of explanation does Plutarch
consider the gods and their will as really existing. All the others (the next
to last being the most often invoked) assume that cult is wholly man-made.
His approach, not only in the first category, is bookish in the extreme, and
in a sense also extremely rational.
By contrast, an inscription from third-century Dacia in semiliterate Latin
describes how Aurelius Martinus Basus and Aurelius Castor Polydeuces,
"standing about, saw the spirit of an eagle come down from a hill upon
three snakes. One big viper tangled up the eagle. The aforementioned freed
the eagle from danger. They set this up as their vow deserved, willingly," to
70 Debatable

the sky god Jupiter of Doliche. 32 Somehow they felt the divine had entered
th.eir lives, and in a fashion far from bookish and rational, they did what
needed to be done.
The differences in religion separating the upper class from the lower
(and no middle really existed) can be more easily sensed than brought out
explicitly through the evidence. That they existed, at least one of the two
well knew, and reminded the other in clear and often scornful terms. ss They
may all be gathered under one alone: "superstition." In tum, however,
what we and the ancients, too, indicate by that term is a crudeness of convic-
tion in regard to both the gods' sphere of action and their nature. The two
elements in "superstition" now to be examined should be kept distinct so as
to grasp the meaning of the word clearly and to understand the chief class
differences in religious views.
The wonderful happening in which the two Aurelii participated brought
the god of Doliche down from the sky to the very roadside. It was a sign that
they could touch-as Timoleon's soldiers could touch the load of parsley
on the mule's back, when they passed it on their way. They took it for a
portent. Plutarch tells the story (the explanation does not concern us),
calling it an instance of superstition. 34 He uses the word to express his dis-
agreement with those who would take a perfectly ordinary event, as he saw
it, and ascribe divine dimensions to it. The more a person saw the gods at
work in the material world-moving things around, for instance-and the
less a person explained in terms of'natural causes, the clearer was the
presence of superstition.
Over the course of the centuries chosen for examination here, supersti-
tion within this meaning certainly increased. The fact is best sensed (to say
"measured" would imply a degree of accuracy beyond our reach) in the
greater prominence of magic; for magic, after all, is most shortly defined as
the art that brings about the intervention of superhuman powers in the
material world-"moving things around, for instance." Proof of the prac-
tice of the art grows more abundant, most obviously in recipes and hand-
books written on papyrus. It is to be found in every province, for example,
in leaden curse tablets. And people who should have known better come to
credit invocations with an efficacy that, in some previous century, would
never have been believed. It means nothing that a late orator attributed an
ineffectual speech to hexing by jealous detractors; an early orator, consul in
Cicero's day, offered the same excuse; but when the very emperor resorted to
wizards to aid him in his wars, times had changed. 55 Perhaps no new ideas
are to be discovered, but old ones are found in circles previously immune to
 2. The Vitality of Paganism 71

them, now ready to acknowledge the direct, visible intervention of super-

human powers in the world at human bidding.
With the change, no doubt partly cause and partly effect, philosophy in
the old sense fell out of favor. This too is a phenomenon well known. It is
often remarked among the upper classes. It shows itself in times of peace
and prosperity in the later second century, or at least the symptoms that
characterize the developed disease are by that period already discoverable.
Its origins therefore cannot lie in the decline of wealth, schools, libraries,
endowed chairs of learning, and the like, although all these latter indeed
suffered in the third century. Alternatively, to explain the waning of phi-
losophy as a mere fashion, in Nock's term (above, p. 65), provides no
serious answer unless the fashion itself can be derived. But contemporaries
had their own explanation: "philosophers agree about nothing-one of
them even says that silver is black. You can hear more uproar from a house-
hold of philosophers than from a household of madmen.'' 36 And variations
on the theme: "I frequented the schools of the philosophers [at Rome] and
found nothing but preparation for the overthrow of doctrines; competing
and arguing; and tricks of syllogizing and the imagining of premisses"-
from all of which the wise man would turn away to more profitable studies. 37
At the end of our period, the emperor himself before a learned gathering
could describe "Socrates, elated by his debating powers, making the weaker
reasoning appear the stronger, and pll!lying about with mutually contradic-
tory arguments ... , and Pythagoras as well, while pretending to practise
self-control in a special degree, and silence, too, was caught in impos-
ture ... , and finally Plato, up to a point, was wise, but in other matters he is
found to have erred from the truth.'' se They had all been wrong, every one.
Disciplined abstract thought, tested by challenges made sharp in the
course of fierce fights over many, many centuries, the challenges themselves
not to be handled or even understood by the casual observer, had passed
from favor. As in the history of warfare periods emerge in which attack
prevails over defense, so in the history of thought a time had come in which
the nays had it. Every proposition could be overthrown by ten different
arguments. All the arguments were in every student's hands, or handbooks.
And the casual observer whose interest counted for most had spent his
youth, not in a university, but in some catechetical school or, more likely,
in a barracks. Thus the balance of techniques of debate, and then political
and economic troubles, and the hastening advancement of people new to
wealth and cultural leadership, combined to bring on the moment when
Constantine could simply wave philosophy aside without being laughed at.
72 Debatable

That was the test: ridicule. Fully to sense the meaning of Constantine's
preposterous pontification, he must be imagined speaking at Plutarch's
table. There, his views would have produced delighted grins; likewise, no
doubt, in the company of Lucian or Apuleius. Lucian knew of opinionated
ignoramuses in very high places indeed, followers of the pious fraud
Alexander. They were the equal in gullibility of the population of Abonu-
teichus where Alexander set up shop. Lucian expects his readers to laugh at
them, as Apuleius could hope (a little anxiously, in a small town like Oea)
to raise a laugh at yokel accusations of magic. He practiced no magic, he
insisted in his defense, but scientific experiment in the tradition of Aristotle.
Who but a clod could misinterpret that? With his trial, we have passed the
mid-second century. We still feel a difference- the difference between "reli-
gion" and "superstition" -separating the literate few of Athens, Rome, or
Carthage from the people of remote centers like Oea in Tripolitania or
Abonuteichus in Pontus. Another hundred years pass, and gullibility is no
longer a target for ridicule. In the most educated circles that the Empire has
to show, enchantments, trances, and wonder-working raise no laugh; rather,
fear and awe. It is rationalism, as w~ would call it, that now must defend
itself; and it is easily put to rout by Constantine. Most of his listeners-not
all, for such large changes come about very gradually- no doubt shared his
views. 39
To return, then, to our point of departure, in the contrast between the
lettered elite and the broad masses of the population, and between the reli-
gion of the one and the superstition of the other: it is clear that the differ-
ences were sharper at the beginning of our period than at the end. It follows
that a sharper objection must be raised against the use of some member of
Plutarch's circle than of Constantine's, to represent the whole world of the
Apologists. Elite and masses, in their views about the supernatural at work,
had at the end drawn much closer to each other. What divided them still
was only a matter of complexity or depth of explanation.
A final sign of their approximation: by piety, that is, by a life lived in
accord with divine will and laws, one could draw down upon one's whole
community all sorts of blessings-wonderful happenings, or good crops
and bread at low prices. "The priestess Alexandra of Demeter Thesmo-
phoros asks" -so begins an inscription of the second century from the
oracle of Apollo at Didyma. And her question follows: "Because, since first
I assumed the priesthood, the gods as never before have been visible through
their attentions-and this, sometimes through the virgins and matrons,
sometimes through men and boys-what means such a thing, and toward
what destiny?" The answer, though largely lost, can perhaps be recon-
 J. How the Divine World Was Envisioned 71

structed: the cause of the phenomena she asks about was the devotion of the
priestess herself, here, as reported also of others in other inscriptions, at
Stratonicea (for a specific miracle vouchsafed) and elsewhere. 40 We can
recall, too, a traveler of the period talking with a Greek shepherdess who
claimed "the power of prophecy given her by the Mother of the Gods; and
all the shepherds in the region and the farmers used her for the fertility and
security of their crops and ftocks.''4 1 Or we can compare another tourist at
Atargatis's shrine, seeing the pillar saints atop their pillars, "whom hoi
polloi believe to be up there in the company of the gods, requesting benefits
for the whole of Syria." 42
Intercessors for divine favor thus appear widespread in the eastern prov-
inces- but more easily to be found in the hinterlands and periphery (Olbia,
Cyrene, Caria, and the islands) thanatthecenterofthings; more venerated,
too, by the ignorant and simple. It is not till the third century that the very
emperors are acclaimed in open ceremonies as winning bountiful harvests
for their farmers and calm seas for their sailors, by their piety. 43 The change
is not abrupt, a matter of emphasis rather than of innovation; but it as-
sumes in the gods accessibility to direct and specific appeal, it assumes their
willingness to make their favor felt by visible tinkering in the natural
world. Earlier, by contrast, when a panegyrist credited Trajan with averting
famine, it was not the emperor's prayers that had brought supernatural aid,
but his shrewd administration that had mobilized quite human forces. 44

3. HOW THE DIVINE WORLD WAS ENVISIONED

Almost to see, or actually and with one's own eyes to see, the natural order
suspended through the exertion of one's will upon the divine, or a wizard's
or holy man's will, constituted only one element in superstition. The other
element was accused of misrepresenting the nature rather than the reach,
accessibility or sphere of the gods. It attributed to them an unpredictable
malice forever to be feared and deflected through seemingly irrational acts.
Plutarch wrote an essay on the subject (in Greek, deisidaemonia, "demon
terror"), which he calls folly because it corresponds to nothing in reality.
The gods in fact are kind; there is no cause for dread. Well said, the super-
stitious man would have rejoined; but he himself had a better understand-
ing of reality; and in that, his own mad actions and panic were actually the
height of good sense. 1 Only Clement of Alexandria penetrates into the
implications of deisidaemonia much further, in supposing that wicked
people will invent wicked, savage gods, just as benevolent people invent
the reverse. The result, a superstitious dread of gods who should indeed be
74 Debatable

feared, is logical-"reasonable," as he says. Normally, "superstition" is

used by ancient writers simply to indicate religious beliefs they do not
share, without their bothering to justify the pejorative term. Justification
must be guessed, and usually can be guessed, by drawing in to the judgment
quite familiar ethical or esthetic standards: temple prostitution or castra-
tion to serve the god better was "superstition," and so was rolling about in
the mud or shrieking and dancing in public.
Needless to say, condemnation of actual, existing cult practices arises in
the library or school, among observers of a certain education and fastidi-
ousness. They are the same on whose experiences it is common to draw for a
picture of religiosity in its various degrees and levels. They appear to have
their own forms of engagement, however, and their own forms of detach-
ment, just as the masses demonstrate their own loud blasphemy, contra-
dicted by unconditional zeal.
Of the one, quite tiny, circle, Plutarch is a good member to advance our
discussion further. He wrote at the outset of the time. span here chosen for
examination, and from the geographical center, near Delphi. His connec-
tions with the Roman as well as the more obvious Greek population have
been noted once before. When we glance through his essays in search of his
view on the true nature of divinity, to be contrasted with superstitious
views, we find him repeatedly insisting on goodness as its essential charac-
teristic: goodness in the sense of active beneficence as well as inactive virtue.
Divinity can never be the cause of ill. We must trust what is "suitable,
reasonable, and likely" on this score, not what the poets have written. The
gods cannot have engaged in violent quarrels-such opinions are "neither
sound nor true" -against each other or against or for men. They cannot
have suffered wounds from men, as Homer relates, or dismemberment or
other mishaps, as Dionysiacs or Egyptians believe, according to myths both
"barbarous and unlawful." They cannot be imagined lowering or dimin-
ishing themselves by too direct involvement in human affairs. To suppose
otherwise would be "naive and childish," disrespectful to the greatness of
divine goodness, arete. 2 And Plutarch scatters in his writings a good many
normative words, showing that he had very clear ideas about the nature of
the divine, even if he does not happen to spell out some of the (to us) most
interesting ones. No doubt he supposed they needed no explanation.
His contemporary, Dio Chrysostom, and (born about when these two
died) Lucian and his generation, can be used to amplify Plutarch's testi-
mony. The divine, says Dio Chrysostom, is goodness complete and the
source of all that is good in our lives: good esthetically, that is, glorious to
look at, and glorious ethically, for our contemplation. As such, gods can
 J. How the Divine World Was Envisioned 75

delight in gifts only from mortals resembling themselves, the just and
good, though their delight implies no satisfaction of a need. Divinity of
course stands in want of nothing, not even of statues or burnt offerings.'
Indeed not, Lucian agrees. It is impious to imagine the gods being bought
or swayed by men; and yetthe contrary view, thatthey care nought for men,
is equally wrong. The key to the dilemma is not to portray them too much
like men, as the poets do; by whom they are even lowered to physical ugli-
ness or menial employ, like Hephaestus, all sooty and "always in the fire, by
his trade"; crucified, like Prometheus; physically restrained in shackles or
wounded in combat; caught up in emotions, in love; or melted down, in
idols. In Egypt, they assumed the shape of monkeys, birds, and crocodiles! •
All such pictures of the immortals do them violence, by setting them in
postures or acts which would be condemned as weak or wicked if they were
human.
Lucian and others do not see the difficulty in applying two standards, one
anthropomorphizing, the other, more abstract, whereby the gods must be
everything that is best in man-handsome, benevolent, high-minded, and
so forth-but also what is best in creatures of some other, higher species
altogether. That difficulty arose from the conftation of long-established
criticisms by Plato, Xenophanes, and others of the past, whose views could
not be made to fit together logically.
Making a cast beyond these obvious authors, for beliefs scattered more
sparsely but more widely among others of the second and third centuries,
we catch some fragments of the Pseudo-Apuleius, the Pseudo-Aristotle,
Iamblichus, Porphyry, and others; also of Apologists. In part, at least, the
latter spoke to pagans, therefore intended to anticipate and withstand their
criticism. It needs no demonstration, of course, that Christians were wholly
of the world around them-of what other can they be supposed?-and
drew in its everyday assumptions with their very breath; that those among
them who engaged in debate with pagans had had the same schooling,
beneath or in step with the catechetical; and that in such debate they could
best hope to win by anchoring their assertions to points of common con-
viction. Hence their free characterizing of beliefs as ''absurd,'' ''irrational,''
"folly," "incredible," "outrageous," and the like, without need of refuta-
tion step by step. All parties shared the same preconceptions. Accordingly
we have for illustration two men who never saw each other (conveniently,
they straddle the midpoint of our period), Celsus and Origen, the former at
one point summing up some arguments of Christians in order to answer
them, while Origen the Christian quotes. that passage itself in order to
supply the surrebutter. 5
76 Debatable

It is "not reasonable" to consider idols as gods, when they have been

manufactured by men, and, worse, by men of low social status and morals;
and the point was long ago made by a pagan Heraclitus (of the first cen-
tury), so says the pagan Celsus-it was no invention of higher-minded
Christians. Celsus is indeed right in bringing out how much derisory or
outraged criticism of current cult practices, theology, and mythology could
be found in pagan writers. Here it is aimed at implications that gods are the
mere creatures of men.
And to continue the survey of divinity as it is portrayed in these less
obvious sources: gods or divinity can do no ill, being goodness perfect and
complete. 6 That, like all the points now to be summarized, has also been
found or implied in Plutarch, Dio Chrysostom, and Lucian. Further, the
gods are infinitely remote from the material world, themselves incorporeal
and insubstantial. 7 To imagine that they have any need of the world would
imply some incompleteness in them; rather, it is of the essence of divinity to
have no desire, no wish, no lack or feeling at all.B
It follows that the gods cannot change, assume other shapes, grow up or
grow old. Assuredly they cannot die and be reborn, like Osiris. They cannot
be cut up, wounded, put in chains, tossed out of Olympus, crippled; nor
have they appetites. They do not eat or drink, defecate, or fornicate. Of
course not. Rejection of such pictures is registered or implied in the writ-
ings of Plutarch and Lucian but also of Heraclitus and Celsus. 9 And no one
may rightly accuse the gods of adultery, sodomy, theft, perjury, cowardice,
murder, or wicked or disgraceful acts of that sort-again, features of belief
shocking to pagans and highly convenient to Christians. 10 The gods should
never be thought of or portrayed as dependent, servile, or menial. The
opposite is the truth. 11 Still less should they ever be described as monsters of
any sort, misshapen, abnormal, or even as animals: Egyptian crocodiles
and so forth. 12
From conceptualizations, the higher criticism turned to visible routines
of worship to make its point. Idols that were in the first place sawn, glued,
nailed, and filed could hardly be divine. The materials of their manufacture
were base, and they endured the birds that shit on them and mice that nested
in them. 18 It was equally misguided, if the gods were conceived aright, to
suppose that they could "taste good strong thick stupefying incense-smoke,"
or respond-still more wicked folly-to human sacrifice. 14 Of all the danc-
ing, singing, miming, or recitation of prose hymns; of all the anointing,
bathing, wreathing, robing, and parading about of images; of all the toast-
ing, holocausts, and cheerful tables; of ivory, gold, sublime skills in painting
and carving-really nothing remained that held the faintest interest for
 J. How the Divine World Was Envisioned 77

Olympus, if that high realm and all its denizens in fact existed. Probably
not.
Certainly not in the sense or shapes that Homer meant, certainly not in
the Dionysiac's or Isiac's demented terms-not if the pagan purists were to
be believed. The gods really lived; but at a great remove. Cult could not
reach them. It might be inoffensive, never persuasive. Mythology, not only
as the poets had written it but as the Phrygians embraced Cybele in it, or the
Syrians, Atargatis, was folly or insult to the true beings above. The sacred
had lost its story when its enlightened critics finished with it.
But who cared? The inappropriateness of common forms of worship,
seen through the eyes of Seneca or Porphyry, appears not to have deterred a
single soul from the inheritance of his tribe. If anyone listened to Epi-
cureans or Stoics, no signs attest to his conversion. Which is not to deny
that conversions may have been made-let us say, must have been made-
but not in numbers at all detectable. The same limits perhaps circumscribed
the Apologists. There is no knowing what effect if any they achieved, when
they attacked religious customs of their time-idolatry, for instance, or
burnt offerings. Had they managed to find many readers for the kinds of
arguments and conclusions that have been summarized in the preceding
pages, surely Tertullian, at the outset of his Soul's Testimony, need not
have deplored the fact that "no one turns to our literature who is not
already Christian.''
And surely, too, we would expect some sign of the higher criticism to
show in the corpus of surviving inscriptions and papyri. It is there- barely:
the one oracle issued by Apollo from Claros, speaking of the hereafter and
recorded on stone; from Egypt, the one fragment of Pseudo-Plutarch's
otherwise lost Placita philosophorum. Otherwise, of Plutarch, Philode-
mus, the Pseudo-Aristotle, Lucian, Sextus Empiricus, Celsus, Porphyry,
and all the Apologists from whose collective testimony the foregoing views
about the nature of god have been reassembled, there exists not a word, not
a scrap of papyrus-against some hundreds for Homer alone. 15 One prov-
ince, and the survival of evidence by chance (unless chance is rather an
element to be desired, in this sort of testing), should not be allowed to stand
for all the Empire. But what is there better? And besides, why should any-
one expect a broader audience for what we might call analytical theology?
No, the literature of religion par excellence was hymns, in every century
(above, pp. 15£.); and hymns present a world of ideas quite cut off from the
higher criticism. Anyone may compare the two and confirm the fact.
In any case, if one were made too uncomfortable by inherited worship
and mythology, escape into beliefs less gross and more intellectually satis-
78 Debatable

fying could be had through reinterpretation. Menander the rhetor in the

third century, describing how to write prose hymns suitable to whatsoever
deity, calls attention to one useful device: the discovery of ainigmata,
riddles. He means a sort of decoding, whereby the speaker detects, or pre-
tends to detect but actually invents, hidden intents in traditional material.
By this trick everything can be spruced up: the older hymns and rites, now
outmoded; cult objects at initiations that offend prudery or common sense;
local images and lore; stories of deities that tell too much or too little. 16 In
the same century, Porphyry takes up a stretch of an Orphic hymn for illus-
tration: Zeus, as commonly portrayed, "is seated, the firm seat of his power
being shown through riddle (dtvtn6~evo~). for his upper parts are un-
clothed because he is visible through his thoughts and in the celestial por-
tions of the universe, but the parts to the fore are covered because he is
invisible among the things hidden below." And Menander again interprets
Apollo as the Sun, Hera as the air, because of the resemblance of her name
to the word in Greek. Rather elementary examples. 17 The stories of Kronos
eating his young bewildered and revolted anyone who stopped to think of
them; but they could suggest cryptically that mind turns in upon itself. So
says Sallust the philosopher. Pausanias worried about them, too, but "grew
to hold a more thoughtful view of them. In olden times those Greeks who
were considered wise spoke in riddles, not straigln out. Accordingly, it is
my supposition that these legends about Kronos are a piece of Greek wis-
dom." 18
Reinterpretation had a long history before the period of our study. The
art found favor among Jews and Christians as well as pagans. Seneca
thought it nonsense, Dio Chrysostom scorned its exculpations of Homer;
but their respective contemporaries Cornutus and Plutarch made frequent,
reverential use of it in defense of existing religion-even Egyptian. 19 The
tearing apart of Osiris could be understood in a new and most unobvious
light; so also the same fate suffered by Dionysus; and passageti in Homer
and Hesiod could be enriched with marginal commentary pointing out, for
example, the former's confirmation of Empedocles.2°
Most practitioners of reinterpretation felt or appealed to a loving awe, an
automatic veneration for everything in the distant past, feelings that were
emphasized earlier. It was too much to throw out one's heritage, Iliad,
Theogony, and all, only because of the parts at odds with a university edu-
cation. And was the more modern enlightenment itself so sure to be right,
after ali-or was it not rather to be supposed that, in the old poems, old
legends and old rituals, men of yore had hidden Cyclopean verities? Up to
 J. How the Divine World Was Envisioned 79

the end, however, until Porphyry, interpreters either said explicitly that
what was drawn out by unriddling should be kept from ordinary folk; or,
Porphyry included, they proposed ideas of a sort that few people could have
understood. Everyone venerated antiquity, no doubt; but by no means
everybody thought its lore required to be rewritten.
Where enlightenment crossed antiquity, the two could be realigned to fit
well enough with each other. That, beyond allegorical interpretation, pro-
vided another way of tolerating both one's religious heritage and the ideas
most in fashion among intellectuals. By realignment, the old systems of
belief were made parallel but subordinate to the new, as intermediaries
between man and the "real" god(s).
Plato had seen the possibilities in this, his followers developed his
thought, and from Middle Platonism it passed into common currency
within that tiny circle who read philosophy at all. They believed in "de-
mon" intermediaries, 21 to which could be attributed everything gross,
wicked, bizarre, or irrational. To higher planes could be attributed passion-
less perfection, sedate entirety, and remoteness above everything material.
According to various thinkers, there were more or fewer planes: of
demons alone, beneath the divine, three kinds, so said Plutarch, Apuleius,
and Plotinus; two kinds, higher and lower, according to Celsus; several
gradations not all called demons by Iamblichus, including "angels" and
"rulers"; or spheres above spheres at unimaginable distances from earth,
each with its special denizens. Seven was a favored number for such regions;
but there might be ten; or, if one thought about it- better, in Dr. Johnson's
phrase, "if one abandoned one's mind to it" -there might be 365. Encoun-
tering Gnostic systems, on the Iuna tic fringe of such speculation, Irenaeus,
Clement, and Hippolytus gave hundreds of pages to their destruction,
repeatedly in their attacks characterizing their adversaries as quite mad-
Irenaeus at one point hanging on to his own sanity, the reader feels, only by
giving way to a choice parody of the wilder flights of Valentiniaqism ..2 2
"Mushroom growths," he calls the Gnostic systems elsewhere (1.29.1 ).
They were as sudden in their appearing as they were light, and quite as
perishable.
But demons were something else again: very deeply rooted in paganism,
western as well as eastern, even if much more easily documented in the
latter. First, the word itself. It occupied a place apart from "gods," a little
lower, in common usage. But there were passages in what might be called
sacred writings, notably in Homer, where the two terms appeared synony-
mously. 2s Christian writers drew a sharp distinction, reserving the proper
80 Debatable

At Dura on the Euphrates in 1920, British troops happened on this and other wall-
paintings among the free-standing ruins. In due course, word reached the American
Egyptologist James Breasted. He visited the site, took color photographs, and pre-
sented them with a lecture to the French Academie des Inscriptions (see its Comptes
rendues o£ 1922, p . 240). From this resulted the excavations at Dura in the 1920s and
1930s, by the Academie, Cumont, Rostovtzeff, and Yale University. Courtesy Yale
University Art Gallery, Dura-Europos Collection.

title for their own God and the lower rank for everyone else's. Their scorn
would have had no sting to it had they not been speaking in well-understood
gradations.
In paganism itself, then, there were agreed-upon ranks of supernatural
power. But the fact emerges less clearly from the vocabulary in ordinary use
than from people's mental picture of the divine order. It is often revealed.
By far the most usual outline is a regal one: there is a king god, Zeus, from
Homer on; there is a queen, or rather, two at least: Isis and Juno regularly
receive that title; and Men, Osiris, Attis, and Helios are occasionally called
king, too.2 4 These are all eastern, perhaps not by coincidence. From Hel-
lenistic times, epithets common for deities define them as absolute masters,
like Hellenistic rulers. In Asia Minor, Men is regularly "despot," tyrannos.
In Syrian cities, like Hellenistic rulers, gods wear military uniform-a mix
 J. How the Divine World Was Envisioned 81

As appears in the restored drawing by L. North, the faded painting in the Bel temple
shows a Syrian triad of deities to the left, dressed in Roman military costumes and
with gold disks (the later Christian aureoles) behind their heads. Below are the
patron Fortunes of Palmyra and Dura. To the right, sacrifice is offered by the
Palmyrene troop commander Terentius, holding some holy scroll, and by the priest
Themes Mocimi, who in A.D. 239 shows up in a duty roster assigned to the troop's
chapel (ad signa, c£. Corp. pap. lat. 33l=PDura 89, and below, chap. 2.4 n.70).
Courtesy Yale University Art Gallery, Dura-Europos Collection.

of Persian, Greek, and then, in our period, prevailingly Roman styles. The
sky god of Doliche, in northernmost Syria, wears the baldric, sword, cui-
rasse, and tunic of the emperor. Mithra on Tarsian coins has the same
costume. 25 It is the early first century in Palmyra when the practice becomes
noticeable in reliefs; in eastern coins generally, not till the second half of
the second century. In disturbed times and places, the unarmed man was
weak; surely also unarmed gods; so they were fitted out for war. That may be
the explanation at first. Later, they borrowed from the prestige of the em-
peror. Portrayal as Roman soldiers touches the images of the Rider Gods in
the Danubian provinces and of Saturn's guards, the Dioscuri, in Africa.2 6
Inscriptions from Germany refer to Sol as "the unconquerable emperor,"
Jupiter as "princeps of the gods," and in Italy even Christ appears in im-
82 Debatable

perial military regalia under the title Christus Imperator, before the mid-
third century.27
God is given his Guards by Apologists, in the second, third and fourth
centuries, just as Helios is so defended by his worshipers in a magical
papyrus post-300, or as the philosophers' supreme being is, in the second
and third centuries. The imagery is sometimes drawn out into armies, com-
manders, prefects, governors, and messengers around the throne.2s Such
were the agents which had to be assumed in order to bring divinity into
touch with the material world or, more crudely conceived, into control of
countless subjects and responsibilities.
Because the divine could will no evil, theorists supposed that demons
served as instruments of vengeance and punishment; likewise as instru-
ments for the dirty or menial business of magic, and for the mechanical
business of providing signs and oracles. It need not be Apollo in person
who spoke at Delphi. 29 Demons were assigned specific spheres of activity-
oversight of animals, delivery of men's prayers to higher authorities, con-
trol over mortal events or natural processes 5°- and specific cities or tribes of
men. 51 The pagan and Christian pictures of divine administration were
identical, however different in origin; both could have accepted what the
emperor Julian later said in his oration against the "Galilaeans" (143A-B),
that "over each nation is a national god, with an angel acting as his agent,
and a demon, and a 'hero,' and a peculiar type of servant-powers and sub-
ordinates." The titles he chose would have aroused some disagreement; the
authority of the Psalms and Plato'sLawsmighthave been matched against
each other-always supposing that the hierarchy had been worked out at
all in the minds of disputants on either side, even by those (both Christian
and pagan in both eastern and western provinces) to whom the thought of
minor beings in action was quite familiar and in whom their names had
always produced a shiver. Not everyone had taken the time to arrange his
ideas in any logical shape. But there would have been no dispute that such
powers did exist.
"Spirits," or however they might be termed, could be found everywhere.
Evidence is more abundant than might be expected, considering that it
must be found in the realm of the shadowy, the shameful, the illegal, and
the illiterate. Everyone believed. But since belief procured no major benefits
-those lay in the gift of the gods-it attracted no priesthood or other
control. Instead, there was a great deal of free borrowing, extemporizing
and invention by ignorant votaries. The description of the spirit world-
the bottom or base of the pyramid-as "the syncretistic, rotting refuse heap
 J. How the Divine World Was Envisioned SJ

of the dead and dying religions of the whole ancient world," while no
doubt a trifle dramatic, conveys the right impression.s2
Magic without doctrine; devils ·without priests; prayers unintelligible;
worship homeless; and ignominious realms of rule over a single house, a
single field, cow, racehorse, gladiator, rival in love or adversary to one's
career or party-all, together, constituted the broad underpart of the world
above this one, the part with which mortals felt themselves to be most
directly in contact. Further above reigned beings of grander dimensions,
still confined, however, in their authority to one village, town, or city; next,
a smaller number of what Christian writers call "Great-Demons," all one
word, like Apollo or Mars. But did the logic of this structure culminate, as it
ought to have done, in one supreme being at the highest, the most remote
point, in the universe?
The answer is yes, in the minds of some; almost, for more; no, for the vast
majority. It was a question requiring more thought than most people were
inclined to give to the relations, not between mortals and immortals, but
among the latter by themselves. Moreover, to describe Jupiter as prince of
gods or as "saving your ancestral gods from every destruction," in public
documents,ss was to imply a subordination not very flattering to the inferior.
To accept it diminished them and denied ancestral worship. How univer·
sally such disrespect was condemned has been emphasized in earlier pages
(above, pp. 2-3). Perhaps that was the reason, then, why testimonies just
now cited, which assume some gods to be subordinate to others, cannot be
easily found before the very end of our period. By that time, the Roman
Empire here beneath had had some centuries in which to suggest itself as a
model for the Empire above. That the human should supply a way of being
conceived to the divine has been shown to be quite natural.
In the second century, Apuleius writes with real anger about a character
in his novel, a baker's wife, "enemy to faith, foe to modesty, who spurned
and kicked aside the divine powers of sure religion, and substituted for a
true religion the audacious imagining and invented rites of a god whom
she proclaimed the only one"; while, in the earlier third century at Rome,
when a worshiper set up an inscription to "One Zeus, Sarapis, Helios,
maker of the universe, invincible," someone else came along and substi·
tuted the name "Mithra" for "Sarapis" 54-two indications that too high a
claim for one god gave offense to the worshipers of others. That sort of
intolerant behavior in paganism was extremely rare.
Indeed, the illustration from Apuleius may be unique; for the second
belongs to a phenomenon which only appears to be monotheism. In reality,
84 Debatable

it is the melding of several gods into one chief: "Zeus Helios the Great
AU-God Sarapis," this on an altar from second-century Carthage. 55 Zeus is
worshiped as Papa and Attis, all at the same time, in Bithynia; he is "Zeus
Greatest Helios Olympian, the Savior," in an inscription from Pergamon;
"Zeus Sarapis" often on gems and amulets, "Zeus Dionysus" in Phrygia or
Rome. 56
More commonly still, supremacy is concentrated in the sun, natural and
visible master of at least the eastern skies. So in Mithraism the sun, in
hyphenation with Mithra, is the supreme deity. Inscriptions from the west-
em provinces make that clear. They also call the sun "Helios" more often
than "Sol," and the dedications are disproportionately in Greek, not Latin,
indicating the god's special favor among people of Greek-speaking origin.
In western Asia Minor, the sun enjoys favor as a witness to men's oaths,
with his all-seeing eye; but that presents him in no regal role. 57 Better, in
Egypt, magical papyri invoke the sun in such terms as "Lord god who
grasps the whole, gives life to all, and rules the universe." ss
In the Levant, the sun is most at home. In Emesa, and among Arabs
generally as Aziz, he reigns uncontested; and as a god great but not supreme,
like Apollo, he is worshiped in Palmyra, Baalbek, and other towns and
cities.s9 A mosaic ftoor from the area (Tiberias) and from the later third or
early fourth century portrays him in a characteristic pose, standing in his
chariot (like Apollo), wearing the commander's cloak (the paludamentum,
like the Roman emperor), holding in his left hand the orb or globe of rule
(again, like the emperor), and raising his right hand in the typical gesture
of benediction and mastery. 40 What was said earlier about how the heavenly
powers were most often and naturally arranged in the mind's eye, tore-
semble the shape of earthly powers, is well illustrated by this picture. It very
much resembles also a well-known depiction of Christ in a mosaic beneath
St. Peter's basilica in Rome.
To examine a little further the form in which the divine world was com-
monly envisioned, Roman imperial coins can be used very helpfully, since
they reached and presumably did not conftict with the ideas of an enormous
audience. At the beginning of our period they advertise Sun or Sun-of-the-
East, Sol Oriens, in quite conventional fashion, to suggest that the region
was friendly during Trajan's and Hadrian's wars. The same conventions
govern portrayal of Gordian receiving the orb of sovereignty from Sol,
toward the mid-third century. 41 Sol as invictus, whose invincibility is as
specific as the beneficent power of his raised right hand, appears on Antoni-
nus Pius's coins first (and about that time, too, in inscriptions); then on
Commodus's coins; and on Septimius Severus's (and at that point the
 J. How the Divine World Was Envisioned 85

epithet is first applied to an emperor himself on his coins); then on Gallie-

nus's and with great but not exclusive emphasis on Aurelian's; briefly, on
Probus's; again, in A.D. 305-310, minted by Galerius and (in many issues)
by Maximin Daia, with Sol bearing the special title, "Guardian of the Em-
perors and Vice-Emperors," conservator Augustorum et Caesarum, on
issues of both Maximin Daia and Constantine; and finally, after the Tetrarchs
had all been destroyed (all but Licinius), Constantine and Licinius each
resumed a specially advertised relationship with Sol, Constantine carrying
it forward to a point some years beyond Licinius's death. 42 From all of
which it has been thought that solar monotheism spread out from the east
over the center, west and entirety of the Roman Empire, until in the reign of
Constantine that ruler, convert to the Sun himself, changed faith a second
time and so became a Christian.
What we have tripped on here is a point at which religious and political
history intersect. There is no way to keep the two distinct, despite the ob-
vious fact that coins and dates and individual personalities of emperors ill
suit the style of our discussion so far. At best, those implications of the junc-
ture which have to do with emperor worship can be left aside. Imperial cult
hardly belongs to this study; and no one supposes that the worship of the
ruler himself, even where he plainly wanted to be seen as the embodiment
of a god, constituted monotheism. It is supposed, however, that the sub-
ordination of all deities to one, and a kind of personal equation with that
one, was sought and advertised by successive emperors off and on over the
course of the whole third century, until at last an orator could declare,
"Surely, Constantine, you have some secret bond with that divine mind
which, delegating our care to the minor deities, thinks fit to show itself to
you alone." 4' The idea that, even over the gods (not "demons," note), a
greater being presided, has been encountered in earlier pages, where the
pyramid of powers was described and where people's perceptions of the
celestial and the terrestrial hierarchy were compared. It may seem only the
last stage of a logical development that finally subordinates the many of
traditional paganism to the one, Sol.
To make this seem more natural, too, there are two earlier nexus of events
to be added to that which ends in Constantine- two very much less impor-
tant and well known, but still quite often described. In A.D. 218, Elagabalus
had come to the throne. He was hereditary priest of the solar deity at Emesa,
in whose favor he now took every measure he could think of to lower other
deities publicly. The principal image of Caelestis-Tan it from Carthage, for
example, was shipped over to pay homage in Rome. A new temple was
built .there for the Sun, new festivals and ceremonies instituted, and every-
86 Debatable

thing advertised on coins. "He used to say that all gods were the servants of
his own god, while terming some its chamberlains, others, its slaves, others,
its servants of various sorts." 44 Words from a doubtful source, the Augustan
His tory- but at least the picture they offer of a celestial monarchy is recog-
nizable as perfectly conventional. Thew hole structure of worship vanished
the moment Elagabalus died, in 222, leaving no sign it had ever been con-
ceived.
Its obliteration was no doubt due to the unpopularity of its author, since
in the 270s Aurelian renewed the experiment, and under his hand its history
can be traced long after his death, in the form of a Sun temple he built in the
capital. It honored a Syrian deity, from Emesa or Palmyra (the story is con-
fused); and games and priesthoods were established; and the cult announced
on the currency through the most emphatic legend possible, "Sol Master of
the Roman Empire." 45
These reigns and their particular religious focus are often thought to
support the view that "if the solar cult had not succumbed to Christianity,
... it could well have become the permanent religion of the Mediterranean
area." 46 The prediction, though of course meant only to indicate the
general force and dimensions in the cult, seems very hard to substantiate. It
ignores or it overvalues evidence, especially evidence that concerns the effi-
cacy of imperial patronage. We will return to that. For the moment, how-
ever, our business is only with the prevalence and depth of monotheism, of
which the solar is certainly the most prominent species. And further sup-
port for its claims can be sought beyond ~he actions of Elagabalus, Aurelian
and Constantine, in wide if blurry statements that are scattered through the
literature of the time. ·
One of Plutarch's friends refers casually to Apollo, "whether he is the
sun, or the Lord of the Sun and Father, and of all else beyond our seeing"-
that statement at the opening of our chosen time span; and this second from
the other end, in which Firmicus Maternus glorifies "the Sun, Best and
Greatest" (Jupiter's usual epithets), "who holds the center of the heavens,
the mind of the world, the moderator, chief of all and prince." Between
these two speakers, Porphyry's voice is heard, too, in Neoplatonic essays
which Firmicus had read.47 All three writers drew their ideas, though in no
very clear or discriminating way, from Platonic and Stoic teachings. Fur-
ther, Dio Chrysostom reminds his audience that "some people say Apollo,
Helios and Dionysus are all one, and so you," the citizens of Rhodes, "be-
lieve; and many people combine into one strength and power absolutely all
the gods, so that there is no difference in honoring one or the other." Quite
true-another commonplace that turns up in Porphyry later, but in Seneca
 J. How the Divine World Was Envisioned 87

earlier. 48 God is one but his force, vis or MvaJ.U~, is many, expressed
through the various familiar divine personalities. So a Roman says, it
matters not whether you receive help from "Lucius" or "Annaeus" or
"Seneca"; and a Greek, Plutarch, in analyzing the riddles hidden beneath
the Isiac legend of "Typhon's plot and usurpation, that is, the power of
drought," and so on through other tales, checks himself: "These matters,
however, resemble the theology of Stoics, for whom the generative, nour-
ishing spirit is Dionysus,'' etc., etc. 49 1t appears thus to beapartofthe intel-
lectual heritage of the times that god might be one; all "gods," simply his
will at work in various spheres of action; and the interpretive structure, as
accommodating of Zeus at its center as of Sol or of any other traditional
deity, no matter which.
But that was all quite abstract doctrine, taught in universities. It left few
signs in surviving literature- beyond the library and school, only one
trace. An unknown Cornelius Labeo, a writer who is perhaps most easily
dated in the earlier or mid-third century, sought interpretation of an Orphic
verse: "Zeus is One, Hades is One, Heliosis One, Dionysus is One." What
did the poet mean? "The authority of this line rests on an oracle of the
Clarian Apollo, in which another name for the sun, too, is added, who is
given among other names, in the same holy lines, that of lao. For the
Clarian Apollo, upon being asked which of the gods was meant by lao,
spoke as follows: 'Initiates must hold their secrets-yet know! lao is Hades
in the winter, Zeus in spring, Helios in summer, and lao in autumn.' The
force of this oracular saying, and the interpretation of the divinity and the
name, whereby Father Liber [Dionysus] and Sol are meant by lao, Cornelius
Labeo treats in his book titled On the Oracle of the Clarian Apollo.'' 50 So,
like others before and after him, Labeo had asked the gods to speak for
themselves. He had sought truth at the source. And if the result was more
Clarian than clarity, at least it did not conflict with the wisdom passed
down from the philosophers: many gods were really aspects of a single god.
That finding had been brought out of the schools into the open. l
But it has been seen in the open already. Dio Chrysostom in a public ora-
tion at Rhodes ventured a general statement about the beliefs to be found,
or assumed, in his audience. To some, he attributed monotheism. It can
only have been mixed Neoplatonism and Stoicism of the sort we have dis-
covered in half a dozen Greek and Latin authors already. That the orator
was right to expect it among his listeners, however, even in an attenuated or
confused form, can be shown out of a rhetor's handbook compiled by a
certain Alexander in the second century. In describing how to eulogize a
god, he distinguishes the wiser opinions, largely Platonic, from those of
88 Debatable

hoi polloi. According to the former, "the gods were engendered by the first
god-as is also the common opinion .... And to some, god seems to be one
and the same, and unites in himself the power, MvaJ.w;;, of a number of
gods, as they say Helios and Apollo are the same and Selene and Artemis
and Hecate are the same; and veneration is offered by all peoples or by some.
For not all gods are recognized among all, but some by one people, some by
another. If, however, the god happens to be universally recognized, that is
the greatest praise." 51 And last, because later in the century than Dio
Chrysostom or Alexander, this declaration by Maxim us of Tyre, rhetorician-
philosopher: "Amid all these contests internal and external, amid all con-
troversy [on other questions], you will see throughout the world one uni-
form rule and doctrine, that there is one god, king and father of all things,
and many gods, sons of god and his coregents. The Greek says so, likewise
the non-Greek." s2
These writers are all generalizing; all are in touch with a wide public,
too, and two of them are known to have traveled around a good deal. It
would be surprising if what they said could not be accommodated to the
other kinds of evidence that have been gathered, so as to yield conclusions
broadly applicable across the whole scene, and throughout the two cen-
turies, of our survey.
We must first confront the very term "monotheism.'' Like most big words,
and "-isms" worst of all, it is no friend to clear thought. It indicates
acknowledgment of one god only. Very good. But it suggests no definition
of "god." That, as we have seen in our discussion of demons, was a crucial
point of disagreement between the Jews and Christians, on the one hand,
and most other people in the Empire, on the other. The two sides were
united in perceiving a pyramid of powers above them-real powers capable
of suspending the laws of nature. They were united, or at least many of their
more intellectual leaders agreed, on the qualities, origin, and distribution
of power( s) within the pyramid. Plato had established their common ground.
Within the pyramid, however, the "monotheists" of our everyday defini-
tion discovered the hatred or enmity of the greatest power toward all others
beneath, an enmity based on a moral vision. "Polytheists" perceived no
split within the pyramid. They could only distinguish the supreme god
from others by the amount of power he possessed. If they gave him all-
that is, if they adopted monotheism in its radical sense-they must take
away power from every other god, thus denying or obliterating everything
in the pyramid save the top. To have done so would have involved the
destruction of their whole culture. That, it hardly needs to be said, could
not come easily.
 J. How the Divine World Was Envisioned 89

Accordingly, pagan "monotheism," though not for the reasons usually

put forward, barely existed at all. The schools still taught, as they had
taught for many centuries, that all the gods were but the applications of the
one, or that they drew their strength only from One. So rarefied a doctrine
left the gods in fact quite undisturbed and independent entities, so far as
ordinary worship was concerned. It was a way of conceiving things as
harmless as it was restricted in the circle of its supporters. Once outside that
circle, whether Stoic or Neoplatonic, there remained only the question of
what name to put to the top of the pyramid. That there should be a top was,
of course, the most familiar of notions universally, reaching back to Homer,
to Etruscan Jupiter, to the local ba'al of this or that Syrian city. Adherents
of one or another might assert for their favorite a special eminence-not
monotheism, this, but megalodemonia (to borrow a neologism from Cle-
ment and Eusebius). The Great Spirit would be portrayed in perfectly
friendly relations with others not quite so great; would have a special claim
but no exclusive right to gestures or titles suited in strict logic only to a
single supreme being- "unconquerable" or "highest"; and would defend
or accept the veneration of worshipers having each his own, different Great
Spirit. "Live and let live" -even among the jealous Tetrarchs, bowing to
each other's "supreme deity" with perfect affability.s 5
It may well have been easier to do so in those late days than earlier. Rather
than indicating ascendant monotheism, "Sol Invincible" on so many coin
issues of Maximin Daia may only have shown the popularity, almost the
need for mere politeness's sake, of flattering language, used more and more
routinely in addressing anyone of a higher status. The later Empire loved
hyperbole. It loved shouted phrases of clarion superlatives: "the very best!"
"unique!" "savior!", offered to the mayor or governor as enthusiastically
as to the god above.s 4 To call Artemis "the greatest," then, so far from indi-
cating a serious view of her supremacy, meant no more in later Ephesus
than "great" had meant in St. Paul's days; to serve and acclaim a god named
only "Highest" did not rule out the veneration of Sarapis and Apollo; and
writers in both Greek and Latin routinely shifted from the singular to the
plural of "god" in their theological discussions without the least sense of
any consequences implied. Modern readers have a hard time getting used to
the practice.ss ·
All this discussion of ours starts, it should be recalled, from the need to
distinguish between the more organized, explicit theology of "the lettered
elite" (p. 72) and that of the masses. In what structure the gods disposed
themselves was a matter explained in somewhat different ways at different
cultural levels.
90 Debatable

One practice hard to understand but very commonly found at all levels is
polyonymy. Some triple forms have been noted: Zeus Helios Sarapis, for
example. But a local goddess, Perasia in Cilicia, was addressed in inscrip-
tions as "Selene or Artemis," Hecate, Aphrodite or Demeter, all the same to
the dedicant, who thinks to magnify her in this fashion. 56 Similarly, the
Mother of the Gods, Cybele, borrows or lends characteristic articles she is
shown with, so as to be portrayed as Bellona or Astarte or Ma of Phrygia; or
her name is simply run together with that of some other female deity. In
Apuleius's novel, the hero hesitates whether the goddess who saves him is
Artemis or Persephone (Proserpina), the latter called "polyonymous" in
inscriptions (and so, sometimes, is Cybele); but he rightly settles on her
being Isis. After Zeus, she was the most truly polyonymous of all gods in
antiquitys7-witness what survives from the beginning of our period: hun-
dreds of lines of an address to Isis (the opening and closing of the text being
lost), " ... ruler of the fleet, of many guises, Aphrodite, ... savior, ruler of all,
the greatest," Persephone (Kore), Athena, Hestia, "in Lycia, Leto, ... in
Sinope, of many names, ... in Caria, Hecate," and soon, through city after
city round the Empire to Italy; "first in the festivals of the gods ... thou, of
things moist, dry or cold, from which the whole is created,'' and a great deal
more to the same effect, typical of the genre of extended prose hymn. ss The
editors suppose the author was a priest, a likely conjecture. He evidently
enjoyed a congregation patient of long sermons.
That audience must have included, over a lifetime, everybody within a
day's walk of the temple who felt the least interest in worship (though no
doubt a few did not, even if they avoided giving offense as overt atheists).
The priest through hymns thus taught a large class, one rather different
from gatherings that heard professional rhetoricians. The latter, whether
dependent on Alexander's or Menander's handbook or masters by them-
selves, like Dio Chrysostom and Aelius Aristides, brought to their theme
what higher culture could contribute: pillage from "poets and philoso-
phers." That was the pair to whom Apologists forever had recourse, for
support or, more often, for a target (above, pp. 9f.). In discussing mono-
theism as in discussing the afterlife, rhetoricians and Apologists alike
spoke to a tiny minority of listeners who read Plato and commentaries, a
larger minority who liked to hear about such deep subjects, but a majority
who preferred their Homer uninterpreted. The different preferences and
relative proportions of these groups can be sensed distinctly (above, pp. 68£.
and 77). All together, however, they made up a number far smaller than
those addressed by the priest.
The priest taught a very simple lesson: Isis was great! He wished before
 J. How the Divine World Was Envisioned 91

his listeners only to magnify her name. Isis was great! -or Zeus or Sara pis,
Asclepius or Liber. How did he know? Clearly, because so many people
said so, in one city after another all over the world. They worshiped her
even under other names. To report and repeat them was a work of magnifi-
cation, not theologizing.
But it naturally proved more acceptable as the habit of free, peaceful
travel settled on the peoples of the Empire. On a smaller scale and in the
eastern areas, that had been seen in Hellenistic times; now, more noticeably,
in Roman times. Given the pilgrims, traders, tourists and, above all, sol-
diers and civil servants moved about willy-nilly from province to province,
comparison of cults was inevitable; and the conclusions inevitably enriched
the praises of this or that deity as being really (so worshipers believed) one
and the same beneath a dozen local faces.
The people of Rhodes, inquiring of an oracle, were given a little hymn to
sing in which Attis was hailed as Adonis and Dionysus, both; in Africa, it
was the new god Antinous, Hadrian's younger friend, who at his death was
hailed as Dionysus-more correctly, as Liber and Apollo, too. 59 So perva-
sive and vital and itself polymorphous was the practice of discovering one
god to be another. Accordingly, illustrations have been sought in unwritten
evidence as well. All sorts of bas-reliefs, coins, gems, and mosaics have reli-
gious subjects. Each deity had his or her characteristic attributes-items of
costume, an object held in the hand, a stance or gesture. Hence, the equiva-
lent of polyonymy should be discoverable in art. As might be predicted, the
earliest signs show up in the east. Mints of cities like Mylasa and Alexan-
dria, even before our chosen period, jumble Zeus and Poseidon together, or
Zeus, Poseidon, Ammon, and Neil us (the Nile river personified)-later fol-
lowed by more and more inventions and combinations all the time. 6 Cult °
reliefs in a shrine of Jupiter Dolichenus on the Aventine hill in Rome
combine the god with Sarapis, assign him a consort (Juno, naturally) and
depict her with certain traits also of Isis. 61 And from the middle Danubt
region the twin Rider Gods, never named, are shown on plaques crowded
with the Dioscuri, Sol, the stars, Epona (Celtic equestrian goddess), Mith-
raic symbols, the tree of life with the snake coiled around it, and much else.
The high point of their popularity falls near the end of the third century. 62
Generally speaking, the phenomenon of tossing together many gods or
their symbols falls a little earlier, let us say, in the Severan age; but, if there
is no gre~t profusion of examples before that, later times right through the
fourth century supply plenty, judged against the proportions of all art
surviving.
It is tempting to infer from this body of evidence that the religious ideas
92 Debatable

herein expressed were of the same sort, promiscuously eclectic, and so to use
it all as proof of a loose but powerfully working syncretism. Tempting, but
risky. In a period of extreme concentration of power into the hands of a Sun
King or the like, European artists best pleased their patrons by frescoes
crammed with figures of mythology, heroes and heroines; and to affirm the
monarchy of Rome at the height of the Counter-Reformation, Bernini de-
signed St. Peter's chair to show a golden mob of dozens of angels and
cherubim aswirl above the throne. Art has its line, cult another. In the
Rider-Gods reliefs, symbols appear to have been added just to fill space,
bearing no relation to the deities honored; in the Saturn reliefs of Africa, the
complex overcrowding is rightly recognized as belonging to art of the
period, especially the earlier third century, whether sacred or profane; and
contemporary fancy coins, Roman imperial medallions, may be used con-
veniently to illustrate the style, displaying "a strong tendency to elaborate
and to introduce subsidiary figures ... for scenes of liberalitas, adventus or
profectio, battle scenes, scenes of imperial sacrifice before a temple ... or
historico-allegorical scenes with the emperor standing or seated in the
presence of deities or personifications.'' 65 N ock cites these views in support
of the argument that the later Principate was a time of widespread, easy
syncretism. The coins, however, are all of secular subjects. Like the evi-
dence for ideas about the afterlife in sarcophagi reliefs (see section l, n.l3),
art seems to introduce more that is problematical or irrelevant than helpful
to our understanding.
Discussions of the phenomenon are very likely to include a favorite item,
the emperor's chapel. Alexander Severus, we are told, for his private prayers
set up the images of ancestors and predecessors, Christ, Abraham, Orpheus,
and Apollonius of Tyana.6 4 The source for such a tale, the Augustan His-
tory, would rather tell against it were there not so many and so universally
distributed examples of many gods being gathered into a single place of
worship. For illustration, take Mithraea, in which are found dedications to
or images of Silvanus (Ostia), Sarapis (Rome), Venus (Bologna), Vulcan
(Metz), Mercury Cissonius (Strasbourg), and Attis (Strasbourg again)-
these and others by luck of preservation: for Mithraea were underground,
and their collections more naturally kept together. A rarer accident informs
us that Mithraic reliefs were set up in a shrine of Jupiter Dolichenus on the
lower Danube; Asclepius was honored in Men's sanctuary at Antioch in
Pisidia, along with various other gods; provisions were made for the cult of
Artemis, Dionysus, Hecate, Zeus, and Heracles in Apollo's shrine at Claros;
and in the same god's complex of buildings at Bulla Regia in Africa stood
statues of Saturn, Ceres, and Minerva. 65 An interesting inference may be
 J. How the Divine World Was Envisioned 9J

allowed. None of these intrusions into the sacred space of the host deity
can have been made without permission of the priests in charge.
Plutarch's friend Clea, herself priestess at Olympia, was also initiate in
the rites of Osiris. She, then, could hardly have objected to the accommoda-
tion of a second loyalty; no more the priestess of the Sun at Philippi,
initiate into the mysteries of Cybele and of Dionysus. A cult association of
Hercules set up a dedication to its own god in the temple of Jupiter Doli-
chenus in Rome, and "the votaries of Sarapis," another guild, built a
meeting room for Isis and Cybele in Rome's port. 66 Examples abound of
ministrants of one sort or another erecting an altar or a plaque or them-
selves signing some honorific inscription, in worship of a god other than
the one they served. The practice can be observed without distinction of
honorand, whether Roman, traditional Greek, Oriental, or Celtic; without
distinction of area; and only circumscribed in time, perhaps. It may be that
such actions are more often attested in the period after A.D. 150 than before.
But even that is not sure. 67
These apparent betrayals of one's god were of course not only open, else
never known to the present; they were divinely authorized. "By the inter-
pretation of the rites of Sol," a worshiper honors Liber and Libera. Obvi-
ously the priest himself had overseen whatever was done; or a village
honors "Zeus Galactinos according to Apollo's command"; a "priest of Sol
invictus saw to the dedication to holy Silvanus, from a vision"; and so on,
by direct order from Hercules or Men or Apollo. 68 It can only have been
priests who guided these acts, seeing in them no betrayal at all. No one but
priests can have permitted the placing in the temple of Dolichenus, in
Rome, of a relief that shows the god sitting next to his consort and holding
busts of Sarapis and Isis: he had welcomed his friends from Egypt into his
house. Priests directed that the feasts of Iarhibol and Aglibol in Palmyra
should fall on the same day. 69 The accommodation, fraternal welcome,
courteous referral, or punctilious deference shown in one or another part of
the surviving testimony seems to an unbeliever merely the interaction of the
worshipers and priests. But worshipers and priests naturally saw it as the
reflection here below of relations existing in the world above. Tolerance in
paganism operated at both levels, until Christianity introduced its own
ideas. 7° Only then, from Constantine on, were gods to be found at war with
other gods.
Returning to Alexander Severus's chapel where he had assembled various
holy images for worship: the collection seems to demonstrate the opposite
of what is sometimes concluded from it. Not syncretism but discrete beings
were on display. By real, undoubted dissolving of the images of two or
94 Debatable

twenty gods into a single one in people's minds, and by the repeating of the
process twenty or a hundred times, the teeming numbers of primordial
paganism, of paganism among tribes and cities prior to the Greek and
Roman conquests, had indeed been very much reduced. The solvents that
brought about this reduction were still at work in Alexander Severus's day.
But they had always worked in the gentlest fashion, without affecting the
essential character of polytheism. In its upper reaches, as we have seen,
"monarchic" or more abstract pressures of interpretation toward unity of
worship did exist. They should not be credited with more than their actual,
quite limited force and significance. 71 They did not reduce Christ, Abraham,
Apollonius of Tyana, and Orpheus to a single figure in the emperor's
chapel (if that ever existed); rather, Christ and Iahveh were drawn into
polytheism on the latter's terms, simply as new members in an old assembly.
There, awaiting fuller incorporation, for a long time they stood at the
edges, in magic and folk belief.
As to emperors of the period like Aurelian or Diocletian, making great
show of close relations with Sol or Jove, at the most they asserted for their
patrons some relative superiority, not a power sole and unique. Jove had
succeeded Sol as Sol had succeeded to the position once occupied by Her-
cules under Commodus, or Apollo under Augustus. Had Constantine not
intervened, why should the series not have been extended further still? Or
perhaps there would have been slow change. Change is a sign of vitality.

4. CONVERSION

Yes, but if paganism really had retained its nature in good health through-
out our period of study, why at the end were there so many scores and scores
of thousands who had left it for the Church?
The question recalls the aim proposed some pages back, to distinguish
several broad types of theology characteristic of several types of person:
Marcus Aurelius at one extreme- he may be chosen for his very long-
drawn-out, choice education and for the strong element of philosophy in
his faith, a consciously worked-out system of explanation-and at the
other extreme, the nameless peasant giving his penny to the shepherd-cum-
seer in the hills behind Olympia (above, p. 73). By drawing the necessary
distinctions, some offset can be offered to the appeal felt in the former type,
whose intellectuality and moral code speak to our own; whose beliefs,
moreover, can be known in some detail because they have been written
down and preserved for us across the ages. Marcus Aurelius in our minds
stands all too easily for "religion in the Roman Empire." Anything toward
 4. Conversion 95

the opposite extreme is observed without much sympathy or long attention

by Pliny, Pausanias, or modern tourists-through-time. Historians, in con-
trast, apportion their focus on the past according to the significance of
people and events-that is, according to the change these wrought on other
people and events around them. As there is no sign and very little likeli-
hood that many were affected by theM editations, the extreme they and their
author represent have correspondingly small claim on our notice. Human
interest is one thing; historical, another-unless it can be shown that
Marcus Aurelius in fact imposed his views on his subjects. To the structure
of influence in his world we will return.
Meanwhile, we have moved along the spectrum of theology from Neo·
platonism and Stoicism and the higher criticism (pp. 74-77) to the con-
venient theory of intermediaries, demons, and the like; then to the shape in
which all supernatural beings together were thought to arrange themselves;
and so to successively wider categories of evidence, hymns, inscriptions,
and symbolic representations, in which successively wider categories of
person dealt, to express their ideas about the divine order. Whether by plan
or good luck, we arrive at last among the masses, in whom can certainly be
found the kind of historical significance we are looking for. What was
essential in their beliefs?
The answer to that question, likewise the answer to the question why
such throngs were attracted to Christianity, is the same. It may be seen in
the moment of conversion. Converts sought reality, they sought truth, and
the definition of what they sought can be seen in what produced a change in
their allegiance. There are plenty of explicit descriptions of the moment.
The heretic Marcus, active in the Rhone valley in Marcus Aurelius's
reign, won recruits to his doctrines through turning water into wine, or a
little into a lot. Another Marcus, two hundred or more years later, brought
water to drought-parched Gaza, where "some of the pagans, seeing what
great wonders God had performed for us, now believing, opened the gate
and joined our throng, shouting 'Christ only is God, He only has pre·
vailed.'" Or in Caesarea in A.D. 306, a miracle is performed "and all the men
and women acknowledge the one and only God of the Christians." Or in
another Caesarea, this city in Pontus and a half-century earlier, the very
priest of a local cult sees a miracle, and "when this happened, the man
believed in the Word, on the very instant, and leaving all, followed" the
Christian. Or further back by another half-century and in North Africa:
"the witness from your gods, then," says Tertullian of exorcisms, "is what
regularly produces Christians." Or when the governor in Egypt questioned
the Christian Phileas, '"Was Christ God?' Phileas responded, 'Yes.' Cul·
96 Debatable

cianus said, 'What has convinced you that he was God?' Phileas responded,
'He restored sight to the blind and hearing to the deaf, healed lepers and
raised the dead to life, made the dumb speak, and cured the infirmities of
many ... .' " ' All but one of these passages were written by someone living at
the same time as the events he described. A great many other witnesses to the
same effect could be brought forward, separated from the events described
by some generations but not necessarily the less to be trusted for that reason.
And many extremely emphatic statements about exorcism by Christians
should not be forgotten. Some have been reported, earlier. Their point in
common was the simplest: announcement of supernatural powers new in
the world it would be quite irrational to credit, without proof of their effi-
cacy before one's own eyes. That was what produced converts. Nothing else
is attested.
Speaking of the votaries of Pythagoras, Iamblichus says, "they recount
these things" (various wonders) "to inspire belief"; and, he adds, since
none of them was within the scope of humankind, "clearly it was necessary
to accept what was told about him as belonging not to a man but to some
higher being.'' By the same logic Marcus Aurelius was reduced, or raised, to
his own faith: "If anyone should ask where have you seen the gods or how
have you persuaded yourself of their existence, so that you are so devout, I
answer ... , from the continual proofs of their power I am convinced that
they exist, and I revere them." And another governor, this time skeptical
not of Phileas's god but of the oracle of Mopsus, submitted the oracle to a
telling test; when it was passed, "that governor was overwhelmed and made
obeisance and revered Mopsus forever.'' 2 So pagan sources tell the same
kind of stories as the Christian.
The last incident about the oracle is related by Plutarch, who in another
passage extends the Il.ne of logic a bit further. In weighing the merit of
mortals who are accorded temples and worship, as if divine, false claimants
can be picked out because "their good fame flourished only a short time,
and then, convicted of false glory and imposture, with impiety and unlaw-
fulness, 'of a sudden fate, like smoke arising,' (as Empedocles had said) they
flew off.'' 3 True divinity, in other words, will prove itself by its wide or
long-lasting impact on the human scene. Therein lies a further test by
which mere magicians and manipulators of minor, dark spirits can be dis-
tinguished. The Apologists stress this as much as pagans. And still further:
the divine is beneficent, as we have seen; demons alone, not gods, do evil; so
if the effects of superhuman acts are bad, or somehow reward the wicked,
then they have been produced by demons, whereas if they are good and
benefit good men, they are really divine. 4 Finally: the divine has no needs,
 4. Conversion 97

or different needs, compared with humankind. Its working can be judged,

and its prophets known apart from frauds, through their superiority to
material things. sAsceticism will mark them. At the least they will not make
money out of their converts.
To credit the divinity behind a name newly presented to one, a person
had to discover the qualities generally thought to belong to a god. Of
course. Those qualities were: a constitution and substance somehow differ-
ent and above material nature; the ability to do things humans could not
do; and the applying of this ability in ways helpful and desirable to wor-
shipers, indeed, helpful to everyone. Individual instances of conversion
(really reducing to two well known, Lucius in Apuleius's novel and Aelius
Aristides) only confirm in greater psychological depth the outline of cause
and effect that can be traced in scenes involving hundreds or thousands-
scenes which in turn fit exactly with the outline of common views about
divinity discussed in earlier pages. The fit is predictable, the process fol-
lowing from the fit, ending in a new belief, likewise predictable. The in-
habitants of the Apologists' world were, after all, rational beings. That does
not mean only or entirely rational. Conversions might lead on to a feeling
of overwhelming awe or deepest thankfulness, as we have seen; to love felt
toward a deity; even to love felt from a deity-once divinity itself was
proven.
But we are now fairly involved with the gaining or losing of believers in
one or another cult and with the types of belief that moved great numbers.
In pursuing these subjects, perhaps the point to begin with is evangelizing.
Aelius Aristides devotes many pages to describing persons of a certain
type known to his audience and only too well known to himself. Like him-
self, they spoke in public, took the title "philosophers," sought out and
consorted with the rich and respectable, asserted their candor and freedom
of speech, and claimed to offer their wisdom to a wanting world without
fee. Actually, he says, they are abusive, greedy, scheming hypocrites, and
atheists into the bargain.6 They may be put beside the eminently respectable
Diogenes of Oenoanda, to suggest the wide variety of preachers on meta·
physical matters; but they may also be dismissed; for clearly their relation to
paganism is quite tangential, indeed hostile.
More to our purpose is "Julius Eutecnius, native of Laodicea, the ad-
mired ornament of Syria.... When he addressed the Gauls," so says his
epitaph, found at Lyon, "persuasion flowed from his tongue. He circulated
among various races, he knew many peoples and afforded training to the
soul among them. He entrusted himself constantly to waves and seas,
bringing to the Gauls and to the land of the West all the gifts that god
98 Debatable

ordered the fruitful land of the East to bear- for god loved mortal man."
And so the text trails off into fragments. It reminds the reader of Saint
Irenaeus. Even the dates fit. But had Eutecnius been a Christian, surely the
inscription would never have been set up in public; and had his like been
found only in Christianity, in that case too he would not have been described
in terms to be confused with them; so we must suppose he was a pagan of a
type unusual but not unknown. 7
A parallel lies at the opening of a third-century Gnostic text, where a
disciple recounts how he was dispatched by the Spirit to preach about what
he had learned of god's beauty. As he preached, men assembled to hear him
and, when he bade them repent, a part laughed at him, others held back, but
some were persuaded and asked for deeper knowledge.s With Gnosticism,
however, we approach the Judaeo-Christian tradition, in which despatch
of emissaries from a central organization, and other formal aspects of
missionary activity, were perfectly at home. 9
The credos carried by Eutecnius and his like cannot have been within the
reach of very large numbers of listeners. Besides, they lacked the ingredient
seen to be essential to actual conversion: the proclamation of wonders.
That figured prominently, however, in evangelizing of a different sort,
centered in the oracle at Abonuteichus. The fraudulent Alexander who had
settled there sent out his agents everywhere to spread the fame of the re-
sponses. It was a conscious campaign. 10 Eusebius, speaking in broad terms
of pagans, describes how, "among them, prophecies and oracles are con-
tinually talked of, and cures and healings of all sorts of illness, and punish-
ments of the impious, and of this you might see them spreading the report
and inscribing stelae and crying it up in every corner of the earth." 11 The
picture he paints is of course perfectly accurate. We have surveyed some of
those stelae in the previous chapter. Posted where no one could miss them,
their letters recut and repainted periodically, they testified not only to the
wonderful deeds of their patron deities but to a set practice of the deities'
servants: priests could naturally be expected to present the cult which they
supported, and which in turn supported them, as impressively and attrac-
tively as possible. They are even to be found spreading the word of their
festivals by heralds, at least from the two shrines of Panamara and Lagina.
But such activity represented no system of beliefs; it sought to change no
one's life; and it quite took for granted, and assumed that listeners likewise
took for granted, the true divinity of the god advertised. It focused rather on
the attractions to be had at the shrine: healing, foreknowledge, or a feast.
On the road or in the marketplace, one indeed encountered the representa-
tives of Cybele or some other god-who begged, not preached. 12 Of any
 4. Conversion 99

organized or conscious evangelizing in paganism there are very few signs

indeed, though it is often alleged; of any god whose cult required or had
anything ordinarily to say about evangelizing there is no sign at all.
A priest in one Mithraeum in Ostia set up some cult reliefs in a second,
where also he served as priest; a priest of Dionysus erected a cult building
"at his own expense, for his dearest homeland," Hierocaesarea; and others
encouraged worshipers to adorn their shrine. That encouragement may be
generally assumed, rarely proven.u A single deity, Jupiter Dolichenus, en-
joyed an active and prominent priesthood. It figures in the epigraphy of the
cult, in the northern and western provinces, far more often than, for ex-
ample, that of Mithra in Mithraic inscriptions. 14 1ts members throughout
Pannonia joined in a common prayer for the emperor and his son appar-
ently at the time of the imperial visit to the region in A.D. 202. Perhaps the
prayer acknowledged the imperial funding of a temple built at Gorsium in
Pannonia a few months earlier .15 In Dacia, three priests combined to erect a
temple to Jupiter Dolichenus; and one later repaired a temple himsel£. 16
Perhaps all of them drew on temple funds, not their own, since such funds
accumulated in other cults and since (unlike some other cults) that of
Jupiter Dolichenus did not draw its ministrants from the aristocracy. Most
of them came from the east.17
Inscriptions occasionally show Isiac priests abroad being drawn from
Egypt: one in Rome, a second in Aquileia, and other doubtful cases. The
liturgy at the end of our period still required a knowledge of Egyptian.
That could have been learned-or faked. 18 Perhaps Mithraism, too, em-
ployed some strange tongue at points in its services; the mother of Dusares
was hymned in Arabic at Petra, the divine twins at Samothrace, in Locrian;
and to Cybele, even in Italy, it was proper to speak only in Greek. 19 But for
the worship of the Sun in Dacia, the congregation of local Palmyrenes
imported a priest from Greece.2o Overall, expatriate cults seem to have lost
at least some of their native character within a generation or two.
Which raises the question of control and uniformity. Was there in fact
such a thing as Isiacism, without further qualification? Was there a Sarapis
or a Saturn?
Obviously not. Consider first, as a test case, Roman Jupiter. From several
considerations, his was a worship most likely to have been the same wher-
ever it was found: at home, closely under the hand of authority and tradi-
tion; in its forms abroad, safe from the familiarity that forgets or casually
corrupts. Jupiter sat enthroned atop many a hill at the edge or center of
cities in Italy, Gaul, Germany, Greece, Pannonia, or Africa, in shrines
called Capitolia. Therein, each year began with special prayers taken up
100 Debatable

immediately after those offered in army camps: prima in principiis, as

Tertullian says, secunda in Capitoliis. 21 In Arsinoe in Egypt, the town
council in A.D. 215 picked out one of its members "for oversight of every-
thing that belongs to our ancestral deity Jupiter Capitolinus," according to
the instructions of the local bailiff of imperial properties-himself no
doubt still feeling the weight of Caracalla's edict of 212. The emperor had
written, " ... I would give thanks to the immortal gods for watching over
me in time of the recent unforeseen conspiracy-on which account, think-
ing that I should, with pious magnificence, be able to do something com-
mensurate with their greatness if I could attract to the sanctuaries of the
gods.... " 22 But when, a little later, in Stratonicea, we discover one man
serving simultaneously as priest of that god and of three others, we must
surely doubt whether he, any more than the local nominee at Arsinoe,
could have been much of an expert in his duties.2 3
That amounts to no more than negative guessing. Jupiter temples at
least bore the right name, even correctly transliterated into Greek. A match
for Capitolia can be found in the "Vatican Hill" at Lyon and again at
Kastel near Mainz, both named after the Cybele shrine in Rome and sug-
gesting dose imitation of the Roman model. 24 The tiny Almo river near
Rome, in which the statue of Cybele, taken out from the city, was annually
bathed, had its descendant in Milan, just as Isiacs built a Canopus at one
place or another or worshipers of Apollo named local springs "Castalian,"
at Antioch and elsewhere. 25 Duplication of such physical features can only
have derived from worshipers' sense of a sort of unitary character in cult. An
even dearer proof of that sense appears in the careful duplication of cult
statues, sometimes attested in words expressly. 26 Then there are the holy
days of Cybele, observed in an unchanging calendar wherever she was
recognized. 27 Hymnodes of Sarapis in Rome closely followed the Egyptian
calendar, and hymnodes of the emperor in Pergamon closely followed old
Roman rituals. 28 Lastly, when the cult of Ma of Comana in southern Cap-
padocia took root to the north in Pontus, her new home was called Comana,
too, and her new services "used the same procedures in sacrifices, divine
inspiration and veneration for priests." So Strabo reports (12.3.32), al-
though he wrote a century before the period in which we are interested.
These scattered details are meant to suggest the limits of uniformity, not
to exhaust the evidence by any means. The use of identical hymns in shrines
spread widely over the Greek-speaking regions can be demonstrated in the
survival of the actual texts on stone (above, chap. 1, section 2, nn.79f.).
More might be said or very plausibly conjectured about the role of those
songs in fixing the character of cult across time. Or we could infer, almost
 4. Conversion 101

prove, the same changelessness in cults all using the rules of sacrifice in
force at Diana shrines in Rome(above, chap. I, section 1, n.66). The typical
attributes with which deities were portrayed-a bushel basket for a crown,
a two-headed ax in the hand, wings on the feet, and so forth-remain re·
markably consistent and in wide use. Enough, however, on such familiar
topics.
Equally diffuse but not so often discussed is the evidence for the lack of
uniformity in cults. Again, a sampling suffices to show the outline of the
subject. In Cicero's day, as his readers know because he took such pains to
tell them, that wretch Clodius defiled himself and jeopardized the very life
of Rome in no act more expressly than in attending the rites of the Bona
Dea. Men in them violated tabu. But in the provinces, in our period, in·
scriptions record the names of a number of male votaries. Shocking and
inexplicableJ29 But no doubt the faithful of Baalbek would have been
equally shocked at the alien portrayal of their ba'al in his temple (itself also
quite odd) on the Aventine hill in Rome. Isiac ceremonies in the west de·
parted from Alexandrine, though perhaps less than one might have pre·
dicted; Dionysiac ceremonies, being not so much out in the daylight, took
on a hundred forms; and the sanctuaries of Saturn, recognizably Punic in
plan around Carthage at the beginning of the second century, along with
recognizably native icons, had taken on an altogether different appearance
by the end of the century-partly Roman, now, partly "classical" in a loose
sense that included Hellenistic elements as well. 5° The development is not
surprising. Though a distinctly regional figure, underneath his Latin
name, Saturn had no one great temple serving as center, model and arbiter
to his worship. So far as the evidence indicates, no one enjoyed authority
more than anyone else over questions of correct liturgy, iconography or
temple construction.
A good deal is known about Mithraism, for a final illustration. Its cells of
worshipers, like those of Dionysus, arranged themselves under a varying
hierarchy; in the reverse of what happened with the Bona Dea, women
slipped into a cult usually reserved to men; and the central depiction of the
god and his great act, the slaying of the bull, took on a variety of postures,
attributes, and connected episodes. The god himself was equated not only
with Sol and Apollo but with Mercury, too, and he sheltered a great con·
fusion of other gods in his shrines. 51 Other features could be added which
have been found by excavators at one or a few Mithraea but not in others.
"The greater part of the people," said Seneca of common Roman rites,
"know not why they do what they do." 52 He could as well have meant
lsiacism or Cybele cult, in which all sorts of conflicting episodes and inter·
102 Debatable

pretations of the divine story circulated. 33 It was not from neglect that the
religious heritage had become, over the centuries, festooned with airy,
blowing, trailing tales and customs. Rather the opposite: too much atten-
tion. Perhaps no deity had roots into a deeper past than the Maid, Kore,
Persephone-whose worshiper Damian, at the very end of the time here
chosen for study, still approached an oracle in ignorance of the right term
with which to address her in hymns. "Savior," answered Apollo. 34 And
Menander the rhetor, speaking of hymns, recommends the composing of
some in a fashion consciously loose and inventive, for example, in "shap-
ing some new deity, Envy, attributing to her as her veil, Jealousy, and as her
sash, Contention" -this in the third century, too, so the free-changing
quality of paganism had lost none of its vitality. 35 Everything mentioned in
earlier pages concerning polyonymy should be superadded at this point,
everything about the various levels of understanding among believers, and
everything implied in the "national" origin of the larger cults. The sum
was confusion. No counterforce for order existed.
It is, of course, conventional to talk about "state cults." I cite a few illus-
trations of that, above (chapter 2, section 3, n.42). Beyond that, modern
accounts commonly attribute to one or another emperor the advocating or
imposing of one or another cult. If such a phenomenon could really be
discerned, it would present us with just the force most logical for the attain-
ing of uniformity in religious beliefs. None, however, can be found. Like
any city, Rome had its temples. They were supported by taxes. Local
authorities administered them, and administration involved control. All
this is too well known to need elaboration. It might perhaps be thought,
further, that the gods in the capitol enjoyed a position defined and sup-
ported throughout the empire. True, temples were built to house them out-
side Italy, often involving conscious imitation, as we have seen. But imita-
tion included Cybele in its embrace with Jupiter, and without any help
from public funds. Plenty of examples do exist to show state monies spent
in the provinces on religion, but for the benefit of local shrines whose
spokesmen attracted some single imperial gift. 36 The conviction that a real
power lay behind some name-Apollo, let us say-must surely have been
strengthened by deference and tribute offered from the emperor himself. In
that sense, imperial patronage may well have produced conversions to the
worship of Apollo. The patron, however, never intended them and still less
meant to define or unify Apollinian teachings.
Hadrian's establishing of a new cult to honor Antinous brought the
whole of his authority into play. His success was sharply limited: compliant
cities built temples, individuals put up prayers, and honorific games were
 4. Conversion JOJ

staged. The games lasted; the name, however, disappeared from people's
minds. Hadrian's effort, the most emphatic of several he undertook, had
never struck root in the west at alJ.37
His successor, the almost motionless old Antoninus Pius, gave Cybele
more place in his coins than had been usual. A special kind of sacrifice
known as the taurobolium is not attested before his reign on behalf of the
throne but is so afterward; and an undated law offers rewards to inhabitants
of Rome's port city who, for the emperor, offer a sacrifice which is perhaps
the taurobolium, according to instructions from the high priest of the
goddess. These data have been assembled to show Pius instituting a formal
presidency of the cult and a new rite with subsidies, not only at Rome but
elsewhere. Which could all be true-or none of it. The dates necessary to fit
the pieces of evidence together into one whole are either unknown or fixed
only by arguments from silence. Moreover, the taurobolium for the throne
can be found on inscriptions in the western provinces but not in the capital
itself. The emperor's rewards seem to have stimulated no new piety. 38
But, if we pause for a moment to examine the scene without preconcep-
tions, we are bound to ask where the very idea of "official" or "state" cults
comes from. Surely they have been attributed to the Roman world by
reasoning from alien, generally modern and Christian, times. Search for
them therefore proceeds with tell-tale indirection, effort, and paucity of
proof. In paganism uninterpreted, at any level lower than the throne,
zealots were hard to find. In court and noble circles natural to all the An-
tonines, they were decidedly "bad form." Why should the emperors depart
ex officio from the ordinary rules of behavior?
It was perfectly acceptable to favor one deity more than another. One
could speak of "Domitian's god" (most likely Isis), meaning whichever he
was known to honor specially or which especially aided him. That recogni-
tion throws light on a phenomenon very common but nowhere directly
explained, the adding of the epithet "August," that is, "imperial,:' to a
deity. 39 Perhaps the deity so described was envisioned as a face more effec-
tually turned upon men's affairs and therefore worth specifying in one's
prayers; perhaps only compliment and loyalty were offered to terrestrial
authority. No one from the past has been good enough to tell us what he
really thought he was doing when he wrote any religious message upon
stone. But the entirely spontaneous turning up of "Sarapis Augustus,"
"Mercury Augustus," or "Diana Augusta" in dedications by private per-
sons throughout the west, obviously without conceding the emperor any
monopoly on these gods' attention, suggests that what was most in the
dedicant's mind was not the gods' claim to extra veneration but rather their
104 Debatable

relation to the emperor. As the other side to that same fact, the emperor
himself could announce the relation without trying to impose it on others
ex officio. He differed from his subjects only in his responsibility to see that
the gods received enough thanks and respect to win their favor for the
realm. That is more or less what Caracalla said in his edict of A.D. 212
(above, p. 100); and it was in defiance of that that his cousin Elagabalus, the
zealot, wrought his own destruction.
Though the whole family of the Severi have been portrayed as "Orien-
talizers," there is no evidence that any but the mad boy Elagabalus wanted
or tried to change anyone's religion. 4o From his reign we pass to others in
which Sol and Oriens were advertised or favored. The sequence of events
was sketched above (pp. 84-85). Until after 312, however, there is no sign
of imperial pressure for uniformity in cult. The independence, not to say
license and shapelessness, of paganism had suffered no disturbance from
above.
No one contested the government's right, of course, to do whatever it
wished in this area. The potential for interference is indicated by a number
of odds and ends of administrative action. An ancient committee for over-
sight of religion, the quindecimvirs in the capital, formally installed the
Cybele priest of Lyon in A.D. 160, and his like, too, in Arausio in Gaul and
in several Italian cities subsequently. Those were all coloniae. Perhaps
other centers enjoying that legal status were similarly controlled in other
provinces. 41 The priest of Hercules Augustus in Apulum in Dacia was in-
stalled by the governor and other priests in Smyrna by the emperor and
Roman senate. 42 And government intervened, often by local request, in
matters of little concern to us, but vital to temple management: the law
regarding rights of inheritance by temples, adjudication of their boun-
daries, scheduling of their festivals, even building of new facilities ..s Early
in the second century in Bithynia, the younger Pliny planned some urban
redevelopment that would have touched a Cybele sanctuary. He inquired
locally what rights might exist of record in a lex dicta templo, a founding
charter; but, he discovered, local customs of consecration were un-Roman.
That tells us that Republican formalities were still commonly in force in
Italy as the mos that Pliny expected to find abroad. 44 At home, when he
planned a shrine in Tifernum, he applied to the local senate for the assign-
ing of the land to be consecrated and so got what he needed.
To determine the fixity of beliefs gathered under the name "Isis" or
"Dionysus" or "Mercury," the need was seen to determine also if there
existed any force in the empire at work in that direction. So our search
began, it will be recalled. Thus far it has uncovered almost nothing. Having
 4. Conversion 105

passed over the obvious areas, in the educated elite and in the imperial
administration itself, we have arrived at Tifernum in Italy. Did cities even
within their own restricted boundaries seek, or manage, to control local
beliefs?
Though the matter has never been studied, clearly town governments in
the west controlled the use and allocation of space within some limits, per-
haps within the city walls. That accounts for the very common use in reli-
gious inscriptions of the abbreviation l. d. d. d., "in space given by decree of
the decurions." 45 A fair number of inscriptions show town governments as
such, and with town money, erecting a temple, idol, or dedication, most of
all in Africa but occasionally in other regions. 46 Pliny mentions how the
people of Hispellum in central Italy had set aside funds to build an inn and
maintain a hot baths at a shrine outside the town. Money well spent: such
facilities were bound to be attractive-at Scaptopara in Thrace, they were
to prove dangerously so, later. But it was only hoped that tourists, pilgrims,
and the idly curious would contribute a little to local markets. 47
Civic pride, an ardent wish to put one's town on the map, to advertise and
embellish it, was a salient feature of Greco-Roman civilization. Everyone
gloried in the god or goddess who watched over him and his fellow citizens,
gloried in the size of the shrine, and loyally supported with his pocket or
applause the most ruinously generous subsidizing of worship. Such feel-
ings best account for the reward of membership in the Pompeian senate
extended to a little six-year-old for his (that is, of course, his father's) re-
building of the Isis temple after the quake of A.D. 62. 48 In Narbo, the
authorities collected money to pay for a particularly expensive kind of
offering on behalf of the emperor-the taurobolium, quite often made by
cities, not by individuals, in Gaul-and determined the type and calendar
for sacrifices for the numen of the emperor. The act would be a prime illus-
tration of secular authority controlling worship were it not actually a part
of the imperial cult. 49
In the eastern provinces, a match for Narbo's act, but quite unconnected
with emperor worship, can be found in Teos. There, the senate and people
laid down the rules for the honoring of "the god presiding over the city,
Dionysus, on each day, by the priest and the ephebes," and so forth. The
details are spelled out fully, including the matter of who pays (the god,
from temple resources). 50 At Ephesus, it was the Society of Elders (gerousia)
that oversaw most religious concerns; elsewhere, it was usually the same
bodies that are found at Teos. Use of sacred buildings for secular, public
business, and vice versa; the honoring of sacred officials in secular settings,
and vice versa; the prominence of the one or the other at each other's cere-
106 Debatable

monies; the alternation of offices from sacred to secular in a given man's

career; and the confusing of aims and benefits in assemblies for the gods or
for civic business-all reflect the closest relations. They were nowhere
plainer than in Stratonicea, where we have seen how town senators, de-
liberating as such, included among their leaders the past and present in-
cumbents of the chief priesthoods. 51 Implicit in such overlap was the poten-
tial for interference in religion at the narrowest local level quite as much as
any we have seen at the provincial or imperial level; and we may now repeat
our challenge: was this control not used to produce uniformity in cult?
The question may sound a little odd. How could there be anything but
uniformity in the worship of Zeus of Panamara at his very shrine (or of the
like deity in any sanctuary anywhere)? But the matter is not so simple. Even
at the absolute roots of faith, paganism tolerated disagreement. It may be
discovered up and down Syria, where a triad was worshiped, somewhat like
the Capitoline, except that it consisted of a male sky god, a consort, and
another male. At one town or sanctuary, however, one member of the triad
would bear one name, another at another. 52 At Sidon, Byblos, and Hiera-
polis, the tourist who inquired within the very walls of the shrine would be
told two contradictory stories about one or another myth or feature of wor-
ship; the same again in Greek centers. Sometimes dispute lay between
versions of the wise and hoi polloi. Still, the wise might disagree among
themselves.
We have already noticed, too, that rites and beliefs changed over time:
among the Arval Brethren in Rome; among Mithraists in Italy, Pannonia,
and Syria; among the devout of Cybele in Italy and Gaul, or of Saturn in
Carthage and the African backcountry. 53 So long as paganism retained its
full life, it went on growing. Even if a part of it had somehow been well
unified across all the Roman empire at a given moment-which we have
seen to be never the case-we might nevertheless expect the vitality of faith
to have produced within that part continual disuse of old practices and
introduction of new. Our expectation is met. In a sizable number of cults
well enough documented for us to tell true innovations from features that
are simply not earlier known to us, a general refreshing can be seen over the
course of the second and third centuries. It affected the rituals associated
with Demeter at Pergamon, Artemis at Ephesus, Hecate at Lagina, and the
hoax at Abonuteichus. 54 All these developed their own "mysteries" because,
perhaps, that was the thing to do. It was a fashion, at least in the Greek-
speaking cities. Each had to have some religious festival, too, and it would
be improved or altered from time to time, at Cos, for instance, or at the rela-
tively well known shrine of Zeus at Panamara. The feast days there, as we
 4. Conversion 107

have seen, underwent several significant, and perpetual minor, elabora-

tions.55
Enough to answer the question, Was there one thing called Mithraism or
Isiacism? There was not, at least not in the sense in which one can speak of
any major branch of Islam or Christianity. Too much in ancient culture,
quite aside from mechanical and administrative difficulties, opposed the
control that would have been required to produce any sort of unitary cult.
The changefulness of our subject must strike anyone who browses through
the epigraphic evidence. It calls before the mind's eye, over and over again,
the most striking pictures of dislocation and decay. Their match are com-
mon in a literary account of second-century Greece, too: roofs fallen in,
votaries departed, idols missing, the whole sanctuary tumble-down. 56 From
a third approach, the archaeological, it can be seen that Olympia enjoyed
its last bloom in the first half of the second century, while Epidaurus, after a
thin first century, recovered popularity and attention in Hadrian;s reign.
After that, it slid downhill. Inscriptions from the site come to an end in the
later third century. Artemis's shrine at Ephesus in the first century likewise
suffered from neglect. Staff had to be reduced, repairs put off, and the
second century continued on the same level. Claros, however, enjoyed good
times then and to the end of the third century, and Panamara, almost as
long.
If the catalogue of sites were extended, as it most easily might be, to
dozens of others, the conclusion forced on the observer would be a simple
one: the ups and downs in fortune of any cult had nothing to do with mis-
sionary activity or (so far as anyone can demonstrate) with properly reli-
gious forces.
Among extraneous elements may be mentioned, and set aside, the purely
art-historical, in which taste is misunderstood as faith. An instance in-
spected above is the liking for Dionysiac scenes among the patrons of fancy
marble sarcophagi. This developed in Italy in Hadrian's reign, favorin~
what were then antique motifs issuing from newly popular workshops. 5
At the same time, in the same circles, archaizing in all sorts of things like
word choice and daily manners showed itself as well in a fondness for
almost forgotten ancient rites and icons and names like Juno Sospita. 58
Needless to say, the vogue could only be cultivated by the upper classes.
Finally, they continued in their liking for Egyptian culture, a liking which
had been marked for a full half-century and which produced both the
Canopus at Hadrian's villa and the Flavians' importation of obelisks to the
capital. It produced also the fourteenth century B.c. statue of Atoum with
hieroglyphics found in a house in Herculaneum. A rich Pompeian mer-
108 Debatable

chant who commissioned Nilotic scenes in his house and Nilotic arrange-
ment of his garden was at the same time a worshiper of lsi~. Very natural-
or rather, very affected. 59
The cultivating of Juno Sospita, Liber-Dionysus, and Isis distinguished
the beau monde of Italy and its admirers but needs to be mentioned only
because of their disproportionate prominence in the surviving evidence.
The ages intervening between the present and the Roman past were of
course bound to preserve the best of literature and art more carefully than
the worst or mediocre. For just the same reasons, students today turn to the
best, first, in trying to reconstruct that Roman past. But in defense of the
resulting distortion, it must be granted that the beau monde, or their
equivalent in confined settings, made people look, conferred prestige, and
gave expression to their private preferences through all that money can
buy. In particular, they financed religion. The contrary to testimonies of
decline that can be read in Latin inscriptions or in Pausanias's Baedeker of
Greece lies in the elaboration of facilities at Lagina, typifying (except in
richness of detail) what can be known or guessed about any sanctuary:
"three porticoes" erected and the record carefully kept on stone- "three
porticoes and monumental gateway and the stoa along the front of the
temple toward the food market," and so forth (above, p. 48). Donors from
the aristocracy of nearby Stratonicea thus preempt our attention; but after
all, they paid for it.
The appearance of ruin that Pausanias mentions- half a wall here, an
unroofed building there-might be offered by a cult center rising in for-
tune. Expansion or improvement had begun, but money was not flowing
into it at the rate allowing its prompt completion. So at Athens the in-
credibly enormous space and parts of the Zeus temple under Hadrian still
awaited a finishing hand 650 years and more after they had been first con-
ceived by Peisistratus. At Claros, outside of Miletus, the shrine of Apollo on
an almost equal scale likewise had stood for centuries incomplete. It never
did get its roof. 60 It calls to mind half-done or lopsided cathedrals in Europe.
At Damascus, perhaps the biggest shrine in the ancient world still needed
part of its circuit wall done in A.D. 264-265, and work on the interior bazaar
in 286/287, and more work in 339/340; at Dougga in Africa, the Caelestis
temple was given to the city piece by piece through the generosity of the
local Gabinii (the same whose members Aulus Gabinius Datus and his son
Marcus Bassus under Hadrian paid for the complex dedicated to Concord,
Frugifer, and Liber Pater).6 1 But the special responsibility assumed by a
clan or family for a shrine, and the repair or expansion or completion of it
as their fortunes allowed over the generations, is familiar from Rome itself
 4. Conversion 109

and from a number of examples in the provinces: for Ceres, in Uchi Maius
in Africa, by the Pullaeni; for Jupiter, at Iciodurum in Gaul, by the
Petronii; above all, in Palmyra for Ba'al-shamin by the bene Ma'zin, for
Belhammon by the bene 'Agrud, and so on.
As anyone would guess, sometimes a poor man got together the cash to
pay for a religious building. More often the poor, if they were to afford that,
had to organize themselves in congregations. 62 On any rich man's land, if it
was not he himself who put up a shrine, his permission would have to be
sought-his or hers. In the Roman Campagna in the mid-second century, a
consul's wife, Pompeia Agrippinilla, headed an association centered in
Liber-Dionysus, whose four hundred or so devotees were overwhelmingly
of Greek-speaking origin mixed with a substantial minority of Roman
citizens. These latter, under Agrippinilla, supplied all the officers in the
association and were all of one clan, originating generations earlier in the
household of a certain friend of Pompey. Of Pompey, too, the Bacchic high
priestess herself, to judge from her name, was a descendant or relative. The
arrangements here recall a group encountered earlier at Samothrace, where
one rich man carried the costs of worship for his dozens of dependents. 6 ~
It is equally predictable that by far the greater part of building for cult
purposes in cities, too, would be paid for by the rich. Among them, only
two special categories need to be noted. One is made up of repeaters: men
and occasionally women of wealth whose names recur more than once.6 4
There is Marcus Aurelius Decimus, former head of the noncitizen com-
munity at Lambaesis and governor of Numidia, honoring "Jove Bazosenus,
ancestral god," plus Mithra, Minerva, Mars Pater, Fortuna Redux, Her-
cules, Mercury, Aesculapius, and Salus, "and all the gods and goddesses."
Punctilious polytheistl-outmatching in the same town even the legate's,
Cominius Cassianus's, restoration of the Septizonium, vows recorded in
the Aesculapius temple, and dedication of the Capitolium, all in his one
year of office, A.D. 210/211.65
The individuality of faith shows through the broader probabilities:
names through classes. Simply by natural inclination, some people felt
impelled to render more than conventional service to a god. They consti-
tute our second special category. They might offer themselves as priests
(exactly how, we rarely know). Their piety was registered, as we have seen,
in the dedication of their lives or in the many testimonies they have left in
one or many shrines. And when religion came under attack, it was they-in
Smyrna against the Christians, similarly elsewhere for the same cause-
who rallied support and directed public opinion. 66 They had their purely
secular equivalents, too: persons whose actions cannot be accounted for by
110 Debatable

position or material circumstances. Two illustrations will be enough.

Marcus Valerius Maximianus, native of Pannonia, learned to honor Mithra
in Poetovio, perhaps; left a dedication in Apulum in Dacia when he com-
manded a legion there; and in the same rank in A.D. 183/185 in Numidia
recorded the same faith again in the Lambaesis Mithraeum. Second, at
Pergamon in the mid-second century, Tiberius Julius Perseus recorded his
vows to Asclepius and a little later built the god a home in Utica in Africa. 67
The type of these two men may be as close as paganism gets to missionizing.
Legates referred to among the instances of religious behavior incidentally
draw attention to the links between military command and the advancing
of some given belief. The connection extends and completes the subject of
centralized evangelizing (above, pp. 103ff.). If there were such a thing as a
"state religion," it could be imposed on no group more under the emperor's
thumb, no group better suited to be the agent of further propagation, than
the army. To anticipate a longer discussion, however, there seems to be no
evidence that the government saw the possibilities latent in the army for
these uses.
A small unit of Pafmyrene troops in a Syrian station in the A.D. 220s had
and must certainly nave observed a written calendar of festivals. Suppose
all units had been issued the same: then we could see to what beliefs these
hundreds of thousands of men were bent by imperial management. But,
first, there is no saying if there was a single calendar for the whole province,
even for a whole station; second, the gods actually honored are just what
one would have expected in a Roman-citizen setting two or three hundred
years earlier, that is, the Capitoline Triad, Mars and Vesta (!),with a few old
Italian or Roman municipal holidays; and third, the Palmyrene troop
itself, off duty, worshiped its native gods and paid no attention to these
others, 68 which can only have seemed hoary, odd, and meaningless. Ques-
tion de dieu, cela manque de realite.
There is record of directions issued by Licinius about A.D. 322 to a post on
the lower Danube, ordering anniversary prayers on his behalf to be offered
in the camp annually to Sol. The ceremony need not have been prescribed
for other camps as well. A couple of years later, Licinius's communication
with an angel yielded, so he claimed, a sort of monotheistic prayer which he
had his troops join in reciting before battle with Constantine. Though the
text refers with exquisite ambiguity to "the highest god," it might fit with
Sol worship. On the other hand, Licinius was polytheist in his other poli-
cies and statements. What seems to fit the facts best is a far more conscious
dictation of worship to the army than had been attempted before, in form re-
calling Constantine's orders to his army in 312, but in content nothing
 4. Conversion 111

at odds with the long series of developments over the preceding half-century.
From Aurelian's coinage in the 270s until almost the 320s, those develop-
ments were traced earlier. 69
Quite incompatible with any theory of a mission for the army is the very
profusion of acts by not quite so high commanders and by smaller assem-
blies of troops. They make vows or sacrifices in shrines and cities, they build
temples and appoint priests within the camps themselves-not, however,
to one god or authorized pantheon but to the usual confusion of all powers:
Diana and Isis, Silvanus and Mithra, Jupiter optimus maximus of true
Romans, and Heliopolitanus, of Baalbek. 70 The most random and partial
search uncovers dozens of inscriptions from widely scattered points in the
western and northern provinces set up in commemoration of some act of
religion by officers in charge of a company within a legion, a squadron of
horse, or an entire regiment, on up to those commanders who were also
entrusted with a province like Numidia, and so to governors of Pannonia
or Noricum. Since some of the texts also indicate that the individuals were
acting as such, and none indicates action ex officio, there is no reason to
think that any of these testimonies represents a centralized policy.
Nothing, indeed, explains where that very idea of "state" or "official"
cults originates, so universally assumed in modern accounts. It rather ap-
pears that persons enjoying an important rank used it in a personal way. If
they were religiously inclined, they expressed their inclinations without
regard to their predecessors or successors, certainly without regard to the
men under them. Much as a fort commander lines them up to receive some
dignitary on a junket and leads them in hoorahs without consulting their
real views about their corpulent visitor, so it may be imagined Roman
legionary chiefs led the praise for Jupiter or Mars. The next year, another
commander would praise another god. That was one of the perquisites of
power-as, in a municipal or rural setting, it was a perquisite of wealth.
On a grander scale, the emperors behaved in exactly the same way.
Having now examined conversion from several points of view, we can
begin to see the shape of the process within pagan terms. It may in the first
place be described by a number of negatives. It was not in the hands of mis-
sionaries. People best described by the word existed, to be sure, but' they
were oddities. It was not the charge or intent of secular government, either,
though the personnel of government from the very top down certainl~ felt
and freely expressed religious convictions. In sum, modern analogie~/will
not work. The professional staff that served the larger shrines, an~ the
practice of putting proofs of the gods' power prominently on display at
even such a small shrine as Pliny described (above, p. 33), reached out to
112 Debatable

visitors and tried to impress them with a sense of awe, therefore of belief.
Periodically, parades brought images of the gods out into the city at large,
or bore them for miles around the countryside. But ordinarily initiative lay
with the unconverted, to come and look and listen, or to do nothing.
The visible structure of paganism cost money. From the beggars, exiles,
and cripples permanently camped in temple grounds, on up to the archi-
gallus of Cybele in Rome, all needed to be fed and sheltered. The shape of
support naturally followed the lines of wealth distribution in society at
large. Holy men and frauds alike hung about rich women; 71 and every brick
and block that went to make a proper residence for the deity or proper resort
for pilgrims was paid for by some member of the aristocracy-paid for or
occasionally voted for, out of municipal funds, or supplied by troops under
the donor's command. The aristocracy of course enjoyed control over all
the heights of wealth and power.
In signing their names to affidavits about miracles or to the cornerstones
of temples they had built, they added their titles: "knight" (like "Bart.") or
"Judge" (praeses) or "Colonel" (legatus). That practice did not indicate
obedience to higher or central authority. Almost the reverse: assertion of
self. It was a far better proof of one's high position to act without orders.
Testimonies of faith from great personages, without need of any added
weight, naturally made a deep impression. Plutarch, in a passage quoted
above (pp. 57-58), gives conscious expression to the pride in persuasion felt
among the nobility. Not only could they present their opinions in more
attractive, moving, forceful words; not only did they speak with the habit of
confidence in public, elevated upon thrones, titles, platforms and inherit-
ances; but the mere fact that they themselves had been won over to belief
was a son of proof of divine efficacy-a miracle. By that, everyone would be
persuaded. By that, beyond their wealth and influence and eloquence, the
aristocracy made religion in their communities their own.

5. THE DYNAMIC CULTS

This outline of the process of conversion fits well enough in the settled
surroundings of a city, village, or rich man's estate. With mention of legions
and colonels, however, we have touched on a special aspect of the subject:
conversion operating across great distances. In religious history, the more
dramatic part, in some ways also the more significant part, is the more
dynamic. By what force can we explain the appearance of entirely strange
faiths hundreds of miles from their homes, in areas that had not known
them, especially if it is true that "initiative lay with the unconverted?"
 5. The Dynamic Cults 113

In attempting an answer to that question, the last in this chapter, it is best

to start with large, easy statements. First, paganism in the Greek-speaking
provinces underwent no significant changes. 1 Individual cult centers be-
came more crowded, or deserted, or richer or poorer; but no major ones
grew up or died. Even the Roman Jupiter made almost no impression on
these lands. 2 Second, in the Latin-speaking provinces, by far the most im-
portant change was the broad establishing of the worship of Jupiter and,
after that, of a fairly straightforward, representative mixture of deities out
of Italy such as Cicero might have known: that is, a mixture heavily Hellen-
ized, with many features equally characteristic of great eastern cities like
Antioch and Alexandria; for Isis and Cybele (as Bellona) had been thor-
oughly at home in the peninsula for centuries when our period of study
begins. So, too, had Dionysus (as Liber). Only two new names, Jupiter of
Doliche and Mithra, appear, not prominently, among the top fifteen (above,
the table on p. 6).
In the ascendancy of Jupiter, one reservation was noted (above, p. 5).
The African provinces opposed to him their own Saturn. Jupiter alone or
in the Capitoline Triad was housed in more, and more splendid, temples;
but he received few prayers inscribed on stone, especially where Punic
traditions were strong. Moreover, those prayers that did address him came
from Roman citizens, town magistrates, officials. 5 The native preference for
the native Saturn proved too strong-which does not mean that Jupiter
inspired merely formal acknowledgment. In the third century and in Nu-
midia, "Jupiter Best and Greatest, of the Apennines," was invoked, whose
home lay at ancient Iguvium in Umbria.
Place names like Apenninus attached to a god point to people's lively
sense of his roots and individuality. Two other deities identified in that way
may be used to suggest the vitality of Italian exports to the provinces: Diana
of Tifata is given a temple at Intercisa on the upper Danube in the second
half of the second century, by a worshiper whose own origins can be traced
back to her neighborhood in Italy, and "Juno the Queen, of Populonia,
ancestral goddess," is invoked in Dacia by a legionary legate of the same
period. • This latter text serves also as reminder of the great tides of alien
culture flowing into the lands across the Danube in the early second cen-
tury. Jupiter or Diana or (we could add) another old Latin god like Silvanus
seemed quite as likely to turn up on new frontiers then as in the youth of
Roman conquests.5
Initially, everyone is agreed, the gods traveled only in the baggage of the
strangers settling in those regions where religion had a dynamic history-
virtually all of the west, as was said just above. The strangers both estab-
114 Debatable

lished themselves with their own faith and communicated it to others.

Those others and their visitors can be divided into a number of obvious
categories, and weight must be assigned to the two explanations for the
rising of a temple or an altar, whether by act of immigrants or of converted
natives. It is in terms of these divisions and considerations that the move-
ment of individual cults great distances around the Empire is normally
described. Each cult appealed to certain categories of persons, which it is
not our business to describe- the decurions, the lower officer class, the rural
poor, and so forth. But a few general traits may be brought out.
Once the familiar picture of traders in motion is considered, 6 it is natural
to characterize them as carriers of cults. It must certainly be assumed, and
can occasionally be proved, that they held on to their beliefs wherever they
went. Take for illustration the pair that identify themselves as Syrian mer-
chants on a votive stone to Jupiter Dolichenus in Dacia. But such a stone
explains only itself. It would go beyond both evidence and likelihood to
turn the dedicants into missionaries, even though that sort of transforma-
tion is very often implicit in modern accounts; sometimes explicit, for
example, in explaining the substantial popularity of Isis in Italy. Isiacism
was plainly not brought back by soldiers, the means whereby so many other
cults had been, and were to be, dispersed. The dates and distribution of the
evidence will not fit that possibility. Moreover, in the provinces, soldiers are
not much seen in Isiac lists. 7
But there is a third possibility far more attractive than either traders or
troops: immigrants to the peninsula brought in by force, that is, slaves and
their descendants. For the sake of focus, we may instance the Isis temple in
Pompeii, pre-dating 80 B.c. Who knelt there? But the answer imposes itself
from a familiar statistic: at the time, a third of the population in Italy was
slaves, of whom in turn the great majority came from eastern conquests and
purchases and of whom there was a far higher than average concentration
in Campania. Isis had watched over them all at birth and again when they
took passage to the west with their captors. Over the years, some got money.
Their descendants would be freed, and prosper. Yet even so, lsiacism in the
town remained mainly the choice of the servile and freedman elements. 8 In
Campania generally, as in Apulia and Etruria, three-quarters of the dedi-
cants in inscriptions to Isis are slave or freed, during the span of the Empire;
elsewhere, three-fifths, in Rome or Venetia or Sicily.
The same material that underlies these statistics and conclusions can be
approached from another direction-yielding the same results. Isiacs ap-
pear in somewhat more than three hundred inscriptions from Italy. Nearly
half of their names (43 percent) betray non-Italian origin. As we would pre-
 5. The Dynamic Cults 115

diet, there are differences according to locale, the percentage dropping

lower in the countryside and where Isis had been longest settled. There, her
worshipers included some natives as well as immigrants who passed as
natives through marriage or other means. By contrast, newcomers clustered
in ports. That explains why, in Latium excluding Rome, 85 percent of the
non-Italian Isiacs are found in Ostia; in all of Venetia, more than 70 percent
in Aquileia. 9 The true percentages perhaps were higher still, for the reason
just mentioned: non-Italian women married to natives, and their descend-
ants, might well appear as natives on the surface and so be wrongly counted,
even while holding fast to their ancestral faith. Isis evidently inspired fully
as much loyalty as any other god, but nothing suggests that she counted
much on new recruits.
Such broad conclusions can only be derived from epigraphic evidence.

Table ofFrequency ofLatin Inscriptions ofthe Empire

Number of inscriptions per year
0 5 10 15
I

Augustus I ,
Tiberius
Caligula
1\
Claudius )
Neru ~
Vespasian ~
.,.,.
Titus, Domitian
Nerva
<
.......
Trajan ~

Hadrian \
\
Antonius Pius
\
Marcus Aurelius

Commodus
Septimius Severus
Caracalla
'-- -
L,... ~
......
1- t:: joe
Macrinus, Elag-o~balus,
Alexander Severus T1-
Maximin, Gordian lii II
Philip
Dec ius
"' [>
Valerian, Gallienus

26S-284
,,
Source: Mrozek (1973) 115.
116 Debatable

By use of it, we can go on to distribute over time the inscriptions of free-

born Isiacs in Italy, so far as they are datable at all: three in the first century;
seventeen in the first and second combined; twenty-four in the second and
third combined; and eight in the fourth. And we can determine a high point
in the fortunes of Isis in the whole Latin west: under Caracalla. Thereafter,
a sharp decline. 10 So it would appear. But the evidence can be pressed too
far. In actual fact, inscriptions of every sort in the Roman Empire overall
trace exactly the same line. Offset by its rise and fall, the lines of freeborn
Isiacs and of the cult's history as a whole flatten out completely.
In the early 1900s, Jules Toutain, the first to make good use of inscrip-
tions for the study of religion in the western and northern Empire, con-
fronted the rising high priest of "Oriental" cults, Franz Cumont. The point
at issue between them was the historical proportions of those cults. Isiacism,
Cumont had said, "conquered new converts in every province"; whereas
Toutain believed that it remained everywhere "an exotic cult, taking no
root in provincial soil. It did not perceptibly modify the religious ideas or
practices of the vast majority of the inhabitants." As usual, Toutain was
right, Cumont wrong. 11 The truth appears in the table above, p. 6; even
excluding her great center of Ostia and greater, in Rome, Isis stands tall in
Italy as nowhere else. In Africa, she had only two temples. One was erected
in the completely Romanized enclave, Lambaesis- the camp and the neigh-
boring civilian settlement. From Lambaesis come all of her inscriptions. In
Gaul, though the popularity of "Oriental" cults has been defended recently
as well as by Cumont, Isiacism counted for nothing. Her only inscription
comes from the chief city, Lyon-where more than a quarter of the people
whose names are preserved for us at all are Greek, that is, from the eastern
provinces. 12 Once more, we can explain what favor the cult did enjoy by
supposing it to have been passed on within families, whose members
moved about, rather than communicated to new recruits.
We are very much at the mercy of our data. We know that Isis appealed
especially to women, but that women rarely put their names and vows on
stone. It is likely that the main part of her worshipers are hidden in silence.
Other distortions may be sensed in glancing through CIL 13, the volume of
the epigraphic corpus that deals with the northwest. Here are 125 dedica-
tions to "Oriental" deities by persons whose sex can be known. Of this little
group, 44 belong to the cult of Cybele, 81 to Mithra, Isis, and Dolichenus all
together. Of the 81, only 1 was a woman; of the44, a majority(24). But most
of the Cybele texts come from the interior of the area (35), most of the 81
from the frontier (70). Cybele is very markedly for civilians, many of them
 5. The Dynamic Cults 117

old Italian; the other gods, for soldiers; and Isis (as Toutain said) simply
not very popular ,Is
In a totally civilian area, Cisalpine Gaul, if we divide by sex the wor-
shipers of the major fifteen deities, we expect to find large numbers of
women addressing Cybele; a few, Isis. But in fact we find only one woman
honoring any "Oriental" deity against fifty-five who honor the Greco-
Roman. Men favored the exotic a little more ( 10 percent), except in the
largest city, Milan. 14 That is predictable. The data teach us, however, that
there were different kinds of civilians according to region.
If we look next at the stretch of provinces from Noricum to the Black Sea,
we can confirm these various impressions. 15 Here, women amount to 10
percent or fewer of worshipers of eastern deities. Their names are attached
only to Isis and Cybele. We find also a marked overrepresentation of dis-
tricts in which troops resided. As these latter, in Gaul and Germany, pro-
vided close to half of all sorts of inscriptions, or as in the great frontier city
of Mainz the epitaphs belong 88 percent to soldiers and another 5.7 percent
to their kin and veterans, so in the Danubian army towns the same military
types predominate-as do those towns, and their neighborhoods, over the
interior. The effect is to present the preferences of the troops as if they were
valid for the entire populations. Their weight has been doubly exagger-
ated.l6
It hardly needs to be said that, even more than women, another category
is underrepresented: that is, natives. Their names, for example, make up
about 5 percent of the total in inscriptions of Moesia Superior. Another
statistic can be used to draw out the meaning of that: only 4 percent of
Gallic inscriptions addressed to native gods are offered by persons with
nonnative names,17 The indigenolls population is far more completely
hidden from us than are immigrant women, and its own beliefs are appar-
ently barred, or of no interest, td the immigrants of both sexes.
It seems likely that the religious history of the West resembled that of the
East in its uneventfulness far more than appears on the surface. Super-
ficially surveyed, the one is sated with experiment, occasionally disturbed
only by innovations themselves developing along conventional lines, while
the other, the Latin-speaking area, is swept by floods of new belief. In
reality, however, the new turns out to be ancestral, ta patria-only dis-
placed. Underneath the lsiac in Ostia is a man or woman from Alexandria
or Antioch, or from a parent or grandparent originating in those cities.
Similarly, underneath the Isiac in Lyon: not a convert but an immigrant.
And underneath the worshiper of Mars is a worshiper really of Lenus or
118 Debatable

Segomo, changing no more than the name to be cut on stone, or changing

not even that, because inscribing prayers belonged to an alien civilization.
It should never be forgotten that the habit and the price of an inscription,
both, were needed to make a mark on the surviving record; and both were
lacking among the great majority of the population. The record on which
we depend almost entirely in forming general conclusions is itself sadly
partial.
But some conversions there were, undeniably. If the old gods were not
quite driven from people's minds, at least they made room for new at the
level which our information allows us to look at. And two cults in the west
appear to have grown to a good size from nothing-therefore, of course,
entirely through conversions-over the stretch of time here chosen for
study.
The first is that of Jupiter of Doliche, virtually unknown in the east, first
attested in the west in the A.D. 120s, seeming to attain a peak of popularity
around 200 and falling off steeply after 220 or so. The number of testimonies
of one kind or another (mostly dedications) is large enough to earn a place
on the Table of Fifteen and recognition from emperors. But for these signs
of prominence there is a common and deceptive cause at work: an intimate
association with the army. The Tetrarchs might very well salute a god so
greatly venerated by the army; and the student today must make a conscious
effort to resist the impression given by the epigraphic evidence. It is clear
that Dolichen us, however many thousands or tens of thousands of the loyal
he may have commanded in a certain sphere for a certain time, enjoyed
favor defined by persons of eastern origin or army calling and did not trace
so dramatic a rise as the apparent numbers suggest. 1s
The second cult coterminous with our period is Mithraism. In the east, a
single monument of A.D. 77 from Phrygia and an inscription and city coins
from Cilicia, much later, are the only traces of a cult thought by contem-
poraries to be "Persian" (whatever that term meant to them); but early in
the second century Plutarch knew of its current existence and prior preva-
lence in Cilicia. He speaks of it as if it might not be familiar to his audi-
ence.19 It was, in sum, then, a somewhat out-of-the-way religion whose
adherents perhaps preferred not to advertise it but which was gaining
notice in the second century.
In the west, a poet from Naples, writing early in the A.D. 80s, mentions
Mithraism. He seems to have in mind a version of an episode from Mithra's
life that was later portrayed differently. About the same time (certainly
before the end of the first century) Mithraic monuments show up on the
Danube at Carnuntum2° and, about A.D. 100, lower down the river, and on
 5. The Dynamic Cults 119

the Rhine and at Rome. 21 From then on, the signs multiply quite rapidly
all over, especially in Rome and Ostia. 22 Representation of Mithraism in
inscriptions is numerically similar to that of all inscriptions of every sort,
throughout the west. About a third of the Mithraic dedicants are to be
found in Italy; but so is nearly a third of the Corpus. A bare majority of the
Italian come from Rome; but that is just the ratio within the Corpus overall.
The civilian worshipers in Ostia include many (a half) with Greek names,
many (more than a third) with Greek names in Rome; compared to the
soldiers, their numbers are about typical in Italy (240 to 8) and in the north
and east, from Raetia to the Black Sea (247 to 65)-although, in the north-
west, they outnumber soldiers in an unusual way (in Gaul, 25 to 3; in Ger-
many, 61 to 20). 23 In chronological distribution (allowing for Mithraism's
late start, so to speak), the inscriptions follow the normal curve; and in
social distribution (making the same allowance), Mithraists include an
unremarkable assortment of one emperor, maybe; a praetorian prefect, for
certain; a good many commandants in camps and their vicinity, and other
ranks as well; from civilian settings, the highly Romanized in outlandish
places (soldiers and tax collectors) and the lowly un-Romanized in melting
pots like Ostia. 24 In excavated Ostia, there were fifteen chapels, all small
affairs: they might indicate as few as 300 votaries (out of a population in
that space of 25,000), or double or treble that, perhaps. Were impatient
larger crowds queuing up at dawn before the still-closed gates of older
gods?25
The weight of a given cult is quite impossible to guess within close
limits. Taking account, however, of a possible undervaluation in the east,
where its members may have been too poor to put their prayers on stone,
and of overvaluation in Italy and the northern frontier zones, where epi-
graphic habits certainly favored them very greatly, we shoul<t no doubt
rank Mithraism among the two or three dozen better-known cults of the
Empire.2& What elevates it above that in interest are two special features:
first, that it had no home, and second, that its chapels were of a design
ideally suited to archaeological resurrection.
Mithraism's ties with the east amount to so little that they can be denied
entirely: it was rather "created at a defined moment by some unknown reli-
gious genius," best, in the Danube provinces. So say some scholars. 27 When
it is found at Dura on the Euphrates in the third century, its presence there
is explained (perhaps rightly-nobody knows) as secondary, brought by
soldiers from Palmyra who had learned it in their service with legions from
Europe. And so forth. But beneath all such guesswork lies the firm fact that
there was a long time in which the cult was never attested anywhere; after
These twenty-four pieces of marble statuary were uncovered by
chance in 1962 in Constantza (ancient Tomis). Most or all are works
of the Severan period. At some unknown moment, most likely in the
250s or 260s, they were carefully but hurriedly hidden. They repre-
sent a cross section of local polytheism: the Fortune of Tomis with
her mural crown; several Hecates; two Graces; Asclepius; Isis; Cy-
bele; several statues of the Thracian Rider God; and so forth. Also a
snake. See Canarache (1963) 133-135. Courtesy Muzeul de Istorie
Nationala si Arheologie Constantza.
The statue of the snake found at Tomis is generally taken to be
Glycon . The identification is questioned but seems obvious as
no alternative is. If correct, it gives us the portrait of almost the
only new god to arise in our period, well known through
Lucian's essay on its high priest Alexander. Courtesy Muzeul
de Istorie Nationala si Arheologie Constantza and P. MacKen-
drick.
122 Debatable

that, scattered notices of it in the later first century; then, within a couple of
generations, full stature as the peer of Sabazios worship, let us say. For a
god with no great temple anywhere, without ta patria in which to anchor
loyalty, without a city or a people to proclaim his wonders, the winning of
prominence across a full quarter of the empire was quite remarkable.
The nearest parallel, but a mortal one bounded, or almost bounded, by
the life of a single "religious genius," was the oracle of Glycon and his
prophet Alexander. By artistic combination of Pythagoreanism, Asclepius
myths, superstitious regard for snakes, old hymns, boys' choirs, Homer and
Moon worship (or lunacy, our unpersuaded informant Lucian would have
said), the religious practices and habits of thought prevailing in the north-
east quadrant of the ancient world were formed into a brand new, artificial
whole. It even had its own "mysteries." That feature, in several older cults
in Greece and Asia Minor, took on broadly similar shapes through mutual
imitation (above, chapter2, section 1, n.20, andsection4, n.54). The genius
of Alexander of Abonuteichus consisted in sensing this and so much else
that was common in paganism across the eastern provinces.
So in the western, veneration for the east (Egypt being good but lands
beyond Euphrates better still to lead one to new heights of sacred lore) was
widespread. Further, everyone could expect some sort of more or less con-
sciously reverent meal together, whatever the name of the deity presiding.
And there should be ranks of leadership-that Mithraism had-and ranks
of initiates-which Mithraism had, too. For the more curious seekers, there
should be complicated doctrines about the return of the soul to its distant
home, perhaps explained in hymns accompanying Mithraic services.28 The
west had its ways, a little different from those of the east-less wordy, less
sunny. Mithraism summed them all up with almost commercial intuition.
In the first decades of the second century, it demonstrated an ability to
make converts throughout the very mixed Italian (and Italian export)
population, in a degree most unusual. The cult (to repeat) had no launch-
ing point, no center from which to begin a larger life-as once Bendis or
Cybele had begun. The puzzle posed by a dynamic faith that came from
nowhere has been often addressed. No solution has won general acceptance.
To touch on some of the suggested interpretations serves as a review also of
major religious trends, at least as they have been commonly perceived.
The most ambitiously overarching explanations for the success of Mith-
raism lie in the characterization of religious life in the times at large. By
contemporaries, that is rarely attempted. 29 Among modern authorities, the
best known, Cumont, begins by sketching the decline and recession of older
alternatives to Mithraism. In our period of study, "less and less is that
 5. The Dynamic Cults 121

sturdy health of character found that, unable to depart at length from the
road, felt no need of a guide and comforter. The spreading sense of decline
and frailty could be noted that follows the wanderings of the passions; the
same weakness that leads to crime urged the search for absolution ... ," etc.
etc.so For none of these thoughts does Cumont, or any adherent of the
"Spiritual-Fortitude, Spiritual-Weakening" school of interpretation, offer
any serious substantiation. The terms of description themselves are useless
as too vague; useless a second time as normative according to prejudices not
divulged, perhaps never examined; and useless a third time as applied to a
population whose moral attributes and inner thoughts are not only almost
entirely hidden from us but not even investigated through such few data as
could be used.
But Cumont continues: "In the third century, the misery of the times was
the cause of such great suffering ... that people sought asylum in the expec-
tation of a better life." This argument, making of the "Oriental" cults in
general a retreat for desperation, can be expanded backward and forward in
time to account for all the manifestations and appeal associated with those
cults.' 1 Indeed, it must be expanded, if their popularity is rightly reflected
in the only promising category of evidence available to us, inscriptions.
Such a correspondence is generally assumed. The number of epigraphic
testimonies begins its steep rise from the earliest Empire, up to a point a
little past A.D. 200. As we have seen, however, the correspondence is illusory
and the rise means nothing, because all inscriptions of every sort rise
equally.
Beyond that, what sense does it make to assign a single character to so
long an era?-as if one were to say, "in Italy, Switzerland, the Low Coun-
tries, Britain, France, and Spain between about 1400 and 1600, people were
tense and worried." The statement denies the very change and complexity
which it is the job of historians to discover, and which indeed they never fail
to reveal, wherever their sources allow them to portray events and figures of
the past full-scale. As if a century, let alone two or three, could be "an age"
-that is, a stretch of time in which just about everybody acted in signifi-
cantly different ways from other human beings before and after, but with
something close to a characteristic uniformity among themselves! It can
only be the observer's ignorance that would make all the life of a vast area
for so long a span, in the mind's eye, blur, shrink, stop. Such ignorance is
the natural condition of the ancient historian, paradoxically inviting him
to arrange his few scattered facts into grand patterns-the fewer, the grander.
Where the temptations and the hazards are so pressing, perhaps it should be
a rule that no one may generalize about ancient history until he has served
124 Debatable

an old-fashioned, seven-year apprenticeship in the teaching, or at least in

the formal study, of modern historyl' 2
If the misery of the times cannot be used very well to explain the popu-
larity of mystery cults, as an alternative it might be argued that "they pro-
vided a theodicy which legitimated the existing order by referring it to
another plane of existence, and so were able to reduce perceived discon-
tinuities."'' Such terminology scholars fall into naturally. It is, alas, only
too easy to be hard; equally hard to be easy; and translation of one's ideas
into simple words opens them not only to understanding but to question.
The ancients themselves (above, chapter 2, section 1, n.32), of course, could
sometimes see that religion helped to keep the have-nots quiet. In Mith-
raism specifically, the allotting of initiates to seven degrees, and the number
of initiates whose secular lives carried them through ranks, civil or mili-
tary, perhaps pointed to parallels that were made attractive in worship and,
from that, tolerable on the job. But was that likely to be what people
wanted? When soldiers take the day off, do they freely seek out associations
just like the ones they left in camp?' 4 The attraction of the cult lay rather in
a broad range of feelings and experiences: in roasting sacrificial hens and
pork ribs on the sidewalk or somewhere above ground, with one's friends;
descending into the barrel-vaulted dusk of the chapel, into the very presence
of the god, for a long meal with much wine; thereafter (it may be imagined)
communal chanting of a prayer, fortifying thoughts, perhaps some special
verses or paean pronounced by the priest. When and how often the priest
spoke of the god's gifts to men and drew worshipers in to a know ledge of the
soul's necessary passage to a higher home, there to abide for all eternity, we
do not know. But these were familiar parts of Mithraism.
Most modern observers of the cult who discuss how it was organized
assume that it had a rather hierarchical feel to it, because of the seven grades
of worshiper. Actually, however, more than seven terms for grades are
attested. Individual cells may have had eight or ten. Some may have had
only two or three. Whether the number produced a special sense of one's
place is unknown. The cult and burial society of Aesculapius and Hygia in
Rome (ILS 7213) had five grades of membership, receiving portions and
handouts of different sizes at their banquets. The practice recalls mention
of cult festivities at which there were to be, as a novelty, no distinctions in
seating (above, p. 47). A huge Bacchic association in the Campagna at
Torre Nova (above, p. 109) had twenty-odd ranks and offices. Entirely non-
religious associations, for example, of the builders and carpenters (fabri) in
Latin-speaking lands, often provided themselves with officers bearing mili-
tary titles, even the mass of the membership being termed (ILS 615) "regu-
 5. The Dynamic Cults 125

lar privates," milites caligati. Odd- but the worshipers of Aesculapius and
Hygia had a rank immunis borrowed from army speech. Where the variety
of evidence points in so many directions and the individual pieces are so
small, surely nothing certain can be said about the social or psychological
meaning of a cult's organization.
If it was in the promise of immortality that the secret of Mithra's appeal
lay, those who had attained it were not thought to be in any hurry to take up
their claim. "Hail to the Lions, new years and many to them!" runs the
salute for those who had reached one of the higher ranks, admitting them
among the full officiants. 55 It is certainly odd, too, that the Apologists make
no mention of any hope of eternal life, a hope to be ridiculed and under-
mined, among Mithraists. Mithraists may have seen eternity in no more
sharply defined or enticing an outline than votaries of other gods (above,
pp. 54ff.). The evidence, however, is hard to make sense of and very much
under debate.
Turning to the second feature that holds particular interest, the subter-
ranean nature of most Mithraic chapels, there the most incontestable facts
emerge, quite literally, from the earth: benches on which worshipers sat at
services or reclined at meals; paintings that show how their officials waited
on them, wearing fantastic animal masks; and various curious features of
design. The vault might be decorated with painted stars whose centers held
glass, to catch light and sparkle; gilt and jewels were used for the same
purpose on parts of cult reliefs to be particularly noticed, with bright paint
applied to the rest; a round or many-rayed hole in the vault admitted the
sun and thus dramatically illuminated some chosen point or person under-
neath; striking scenes were portrayed in relief and color on the two sides of a
stone slab rotating so as instantly to reveal a second message to viewers; and
arrangements allowed the sudden raising of a curtain from in front of such
a scene.56 Altars had cut-out crescents with glass in front, to be lit from
behind by lamps or candles; or they had cut-out halos or sun rays over a
deity's head, or whole figures of deities cut out-in short, elaborate provi-
sions for unexpected lighting effects, the more certain to impress because of
the ordinary darkness of the windowless surroundings. 57 In one German
Mithraeum, by good luck not rusted· into nothing, excavators found a trick
sword consisting of the handle and upper part of the blade, attached to a big
loop of metal, ending in the continuation of the blade; so, with the loop
around half his chest, an actor would appear to be transfixed by the weapon
but still alive. 58 That fits with other mentions or proofs of naked men,
blindfolds, helmeted men, swords, graves, and corpses real or resurrected,
assembled into what initiatory show, no one can say.
126 Debatable

Underground roused powerful associations: of death, necromancy, and a

separate world of deities and spirits. The physical fact of darkness around
you heightened expectation-and Mithraea could be very gloomy indeed.
Tertullian remarks on that; the Augustan History reports "it was the cus-
tom [in Mithraic services] to do or say something for a show of terror." 39 All
speculation concerning the doctrine taught in those services aside, all con-
troversy about cult meals aside, all doubts about the Mithraic view of im-
mortality disregarded, one certain thing remains: the subterranean rites
were carefully designed to amaze and terrify.
Of no other cult can quite the same be said. Yet of several it is known that
crypts, hidden narrow passageways, gloom, and torchlight were played
with to produce in initiates feelings of fear and awe. 40 Lucian describes the
jugglery used by Alexander of Abonuteichus, Hippolytus knew a bookful
of tricks of the same sort. There is really a lot of talk about religious deceits
in writers of the time, Jewish, Christian, and pagan. We should not be sur-
prised to find them employed in a widespread way in Mithraism. Added to
an eclectic base of broadly acceptable cult practices, the advantage that
Mithraism may have enjoyed-or rather, manifestly did enjoy-was show-
manship. Which is not to deny sincerity to the priests who decided just
where best to cut a nineteen-rayed aperture in the ceiling of their god's
house, to let in the light, or who had a fake sword discreetly hammered out
on the forge of the local smith. Sincerity and deceits may easily go together.
Nor is it to deny that the symbols freely applied to a cult scene or myth
scene, a fresco or relief, may have been linked in learned discourse to the
astrology of the time, the theosophy, and so forth. Those were a part of the
common culture of paganism. Different devices appealed to the clever clerk
on the governor's staff and to the barefoot porter in Ostia who carried grain
sacks on his back all day.
The question raised on a previous page, "How can the exceptional faith
be explained that reaches across a hundred miles, or turns up whole prov-
inces away?" can perhaps best be answered through another question:
"What made converts?" -converts of any sort, near or far. To that latter,
the answer was seen to lie in the visible show of divinity at work (above, pp.
95-97). Thereafter, to make a richer belief, other thoughts and feelings
might take over. So it all was with Mithraism-whether also of the cult of
Jupiter Dolichenus, no one can tell. To repeat: these were the only cults
whose power to draw in new recruits on a significant scale, at least for a
time, is clearly indicated.
Their success has been taken sometimes as a sign that older gods had lost
their hold on people's loyalty. The "Oriental" were demanded by some no-
 5. The Dynamic Cults 127

longer-satisfied spiritual appetite. Their title, however, has meaning only

for those who think that most of the Levant and Near East, with Egypt and
parts of the Middle East tossed in, shared religious beliefs and practices
sufficiently similar all to fit better together than any of them might fit with
the Greco-Roman-that is, that Mithra was more like Sarapis than like
Dionysus, the Syrian triad more like Isis than like the Capitoline Triad. So
stated, the unity underlying the notion of "Oriental religion" falls apart.
It is still harder to accept the decades around A.D. 300 as those in which
the "Oriental" cults enjoyed their most pronounced ascendancy. Close
before those decades, Aurelian proclaimed a special veneration of the sun;
close after, both Constantine and Licinius put Sol on their coins to adver-
tise their trust in him. But in between came the Tetrarchs, titled Jovian and
Herculian. What could be more evenhandedly, quintessentially, conserva-
tively Greco-Roman than that? Diocletian's coins most favor "the Genius
of the Roman People." Nothing could be more old-fashioned. But he also
reduced the pantheon of his coins generally. At the end, Jupiter, Hercules,
and Mars were left almost alone. Hardly a sign of paganism shrinking. The
real explanation is unknown, but Antoninus Pius's coins likewise acknowl-
edged only a small selection of the gods. 41 For that policy of his, too, no
explanation is known; from it, no weakness in paganism inferred. About
the same time, the coinage of eastern cities was becoming more crowded
with a larger selection of deities. Possibly the reason was esthetic. 42 We lack
the keys to unlock these puzzling phenomena.
Nevertheless, interpretation still insists that paganism was "pressing on
to its doom" (or the like phrase). We are invited to consider the sharply
diminishing number of inscriptions attesting to the active worship of
Mithra himself, in the third century, to say nothing of Vesta in the capital,
Saturn in North Africa, or Dolichenus in Dalmatia and Pannonia. I choose
particularly striking examples of deities fading out of our sight within
their very homes. 43 The data, however, tell nothing about people's faith.lt
is not the priest who is stilled but the stonecutter. For the last time, the
reader is referred to Mrozek's frequency table (above, p. 115). Religion, like
many another aspect of life, rises and falls on the quantity of surviving evi-
dence like a boat on the tide. Highs and lows of attestation, if they only
follow the line on the table, indicate no change at all.
The tide itself-the number of pieces of data, especially written data,
remaining from any decade in the first four centuries- has been noticed
only very rarely and casually. The neglect is surprising, in this present era
of statistics. But the dating of the pieces is problematical and the usefulness
of all the necessary work not obviously apparent. After all, everybody knew
128 Debatable

there was some approximate correspondence between the general wealth,

peace and "progress" of the Empire, on the one hand, and the richness of
documentation, on the other.
Perhaps the matter is not so simple as it seems. The custom of presenting
some thought or fact to one's community, living and to come, was generally
unnatural, at least unnecessary. It assumed literacy. It had to be displayed
to those not familiar with it and imitated; it had to be maintained at some
cost by those to whom it was a part of inherited culture. Apparently both
categories of person found they could do without it. Its decline in the west·
ern provinces was precipitous, as we have seen; in the eastern, perceptible
but never measured. In a large section of Syria, some hundreds of inscrip-
tions have been found spread across our period and on through the sixth
century. They show a gap almost complete over the years A.D. 250-325. 44
Other kinds of written evidence elsewhere in the east also are less abundant
then than fifty years earlier. The explanation that might most naturally
offer itself is poverty, during the Empire's Time of Troubles. But if that
were right, why, earlier, should the number of inscriptions have continued
to rise past Com modus, even past the turn of the second century? By then, as
everyone surely is agreed, the Empire's fortunes had long since taken a
downward turn.
It is not easy to understand why that section of Syria in the mid-third
century, or most of Egypt some decades earlier, or Proconsular Africa for a
period beginning in the A.D. 230s and Pannonia in the 220s, and so on all
around the Empire-perhaps most puzzling of all, the Arval and Vestal
cults, snug in Rome-should clearly start to fade out of the written record
as they did, at a time when there appears to be no common factor in control
of their fortunes. Not even coinage touched them all in the same way.
Imperial currency did not reach into Egypt at all; and small change was
imperial in the west; in the east, municipal-at least to a considerable
degree. Even where both the deterioration of the currency and the fading of
the written record can be noted together, unmistakable signs of prosperity
may appear (just as the reverse appears earlier: general prosperity, but
specific depressed regions like Greece). 45 I must leave these puzzles unsolved,
however, and go on to say what I think can be said.
If the fortunes of the pagan record-the number of testimonies to prayer,
invocation, praise, and so forth-run parallel to those of other phenomena
-to writing itself, in chief-so that one cannot say those fortunes were in
decline, yet nevertheless, when the worst is over and other lines begin to
rise, it is not paganism that reemerges. Rather, it is Christianity. That, at
least, is true in several places. 46 To which another fact should be connected.
 5. The Dynamic Cults 129

Whatever was needed to produce an inscription was more easily found

among Christians than pagans. During the same worst period, the curve of
epigraphic data drops down for inscriptions of some sorts more than for
others: for economic indicators and economic records of municipal life or
of private citizens, or for expenses generally. 47 It would be hard to deny,
then, that the publicity attending religion had something to do with money.
The connection seems obvious. It can be confirmed. When the pagan
masters of the world set about the destruction of Christianity in a thinking
manner, they did so from the top down, evidently taking it for granted that
only the Church's leaders counted. 48 Suppose that it was the natural but
wrong assumption of the ruling class, to think itself, or its like among
Christians, the essence of everything; and suppose therefore that the focus
of the persecutions does not prove, in paganism, the key importance of the
ruling class-yet there are still other proofs. They arise almost by them-
selves from the nature of paganism as it was portrayed in the first chapter,
above. From that portrait, if the aristocracy and their wealth were to be
removed, a very, very great deal would disappear with them. Consider a few
of the consequences. It was not universal, but it was the prevalent fact, that
the gods were decently housed. If possible, they should be housed magnifi-
cently, exhausting the resources of local rich families over generations.
Paganism, therefore, for us to recognize it in its familiar forms, required the
patrons who paid for its temples. Games, exhibitions, performances, and
public feasting likewise cost great sums. To offer certain sacrifices was out
of the ordinary man's reach, when five months of his labor were needed to
make up the price of an ox. 49 The impression of strain conveyed by these con-
siderations is reinforced by recalling from earlier pages the many glimpses
of shrines active but only half-built, or half in ruins.
The more prominent, public features of paganism no doubt did fade out
in the third century. That fact, unlike the silence that falls on record keep-
ing of several sorts, allows straightforward explanation: costs of cult had
come to seem too high. The sense might be challenged: "What sort of reli-
gion is it that takes money?" asked the Apologists in derision (above, pp. 96-
97). A fair question-if it is also fair to ask, What is especially good about
a religion that costs nothing? Why should the hollow-cheeked St. Anthony
stare down the Buddha? Eating and drinking may be a form of worship,
and dancing and singing, building and painting, writing and reading.
Paganism, as it existed in good times and as it was presented in my first
chapter, there was no reason to reject; nor was there any general disposition
to diminish pomp-not in the age that saw Rome's rebirth sanctified by the
emperor Philip with incredible expenditures, nor in the age of Constan·
110 Debatable

tine's basilicas. Only, most people who would formerly have paid the costs
of festivals and temples at the local level could now, in bad times, no longer
find the money.
To the best-attested fifteen cults, or the best-attested thirty, if their tradi-
tions had become too expensive, alternatives emerged here and there. They
were drawn from the non-Greek, non-Roman substratum to be found
throughout the Mediterranean lands. Its religious expression, so rich and
varied, underlay a great many hyphenated names, as has been pointed out:
Artemis Pergaea, Mars Segomo. Worship of such deities had continued for
many centuries beneath the very feet of Greek and Roman (or even earlier,
Celtic or Punic) conquerors in Asia Minor as in North Africa.' 0 It consti-
tuted the oldest part of living paganism-in a Darwinian sense, the fittest
to survive, since it required so little nourishment. "Higher" forms of reli-
gious life needed, and might not always find, the kind of bountiful patron-
age that made brilliant the festivals of Zeus of Panamara and Hecate of
Lagina. The "lower" unexpectedly gained somewhat in strength and prom-
inence, over the whole of the third century.
But with cults that are centered in cities or tribes, we return to our starting
point, "national" gods. They were the beginning and essence of the whole
resilient, living fabric of belief. For the most part, however, they lie below
the reach of our inquiry, at a level of the merely local and illiterate. More-
over, they were the ingredients of the Empire's culture least dissolved in the
melting pot, therefore least to be described in any general statement about
paganism as one whole. When political and economic conditions favorable
to them settled in, as was clear by (indeed before) the end of our period of
study, the point at which this study ought to end is also clear.
 Epilogue
THE MANNER OF DEATH OF PAGANISM

Paganism as one whole is most easily seen as the patterns of thought and
action, and their attendant feelings, roused by belief in the existence of
superhuman powers; such powers are most easily grouped in the mind as a
structure having a narrow summit and a broad base; and such a structure
most easily accommodates the known facts about it if it is conceived as
changing all the time, therefore alive. The metaphor of the pyramid has
been, in earlier pages, employed often enough to make it familiar, both in
regard to the superhuman world with its few great and many minor beings,
and in regard to its worshipers, among them, the few of the ruling and
literate classes and the many of the obscure and illiterate. The organic
metaphor likewise is familiar.lt underlies phrases like "dying paganism."
Metaphors have their uses, also their deceits. Paganism died, agreed-
like the last stegosaurus or like a coral reef?
Having its origins in particular localities, its upper parts generally cor-
responded with the fortunes of the natives of those localities. Athena and
Jupiter were great because they were the deities of conquering peoples;
Artemis and Saturn guarded great cities. They inevitably preoccupy our
notice. Indeed, the elaboration of their worship becomes, for us, the very
canon and definition of paganism. It is, however, in fact only the upper
parts that we are looking at, and they are vulnerable to changes in fortune
as the lowlier cults of less important peoples are not.
For one thing, if the means by which we know them are lost to us, they
themselves will appear to have receded. The appearance may be false, but
we cannot easily make correction for it. It is this difficulty that the dimin-
ishing of inscriptions creates-of coins, too, and of papyri and written
works generally. But most inscriptions, unlike these other media, served no
obvious or practical purpose. They arose from someone having something
he wanted to communicate to his whole world, from which in turn he ex-
pected approval or esteem: approval for piety or achievement-or both, as
for example in a text that might read, "So-and-so, high magistrate, erected

131
JJ2 Epilogue

this monument to Such-and-Such a god at his own expense, willingly."

Much of the epigraphic record on which we depend for an estimate of
paganism reflects not only what people believed but whether or not their
belief was likely to raise them in others' eyes. For the same reason, then, that
few Christians declared their faith during the third century, few pagans can
have cared to do so, once the emperor, and with him, those who sought his
favor, changed allegiance.
For Constantine ushered in a new age, in which for the first time there
was such a thing as a state cult-moreover, one animated at the center by
that missionary motive unique to Christianity. In a number of his state·
ments and in more still of his actions, he declared his intent to increase the
ranks of the faithful. Contrast with the adherents of other gods needs no
underlining. At the very towering peak of their appalling rage and cruelty
against Christians, pagans had never sought to make converts to any cult-
only away from atheism, as they saw it. Toleration gone mad, onemaysay.
Such were those times, cruel and violent. Under the Tetrarchs, Church
leaders in the chief eastern cities were arrested and tortured; even pagan
priests and dignitaries if suspected of fraud.l Thereafter, shrines and the
holy things within them were again confiscated or destroyed; leaders exiled;
humble worshipers beaten up. In these incidents, pagans were now the
victims. 2 Occasionally they reacted with their own violence- over the long
run, unavailingly.
They did not suffer at the hands of the governors' guards, as Christians
had done. The army remained in the fourth century predominantly pagan
-so it is safe to say, though just when the balance tipped the other way no
one knows. 3 A. H. M. Jones puzzles over the supine acceptance by the army
of now one faith in its commander, now another, from Constantine through
the 360s. There is no clear explanation; and the behavior of Arbogast's and
Eugenius's soldiers extends the problem almost to the end of the century.
One conclusion, however, imposes itself: soldiers somehow managed to
keep their religion separate from their business, even when their business
appears to us to have lain with religion. That was as true in the 390s, in
Italy and the west, as it had been in the east, in the transition from Julian to
Jovian to Valentinian. It suggests that emperors and commanders dared
not press their views upon those of their servants who bore arms.
Powerful people made their own rules. Those in Italy, even unarmed but
of high station, maintained theirloyalty to the old gods quite openly. Their
story has been told a hundred times. Little details add themselves continu·
ally, in the form of inscriptions recording the repair of temples or the
holding of priesthoods by private individuals or officials.4 So long as they
 The Manner of Death of Paganism JJJ

were left alone, they gave more than respectability to their faith. There is no
reason to suppose theirs was not the majority party until the fifth century.
Because of their family holdings and prescriptive right to governorships in
the central African provinces, those latter also yield many indications of
paganism respected, open and confident, throughout the fourth century;s
but there were deep-rooted cults there, too, which needed no support from
Italy. 6
To the north, the restoration of a long-deserted Mithra temple was adver-
tised by an inscription of the period, in Virunum; but this was under Julian
the Apostate, therefore perhaps an insincere act. What could be more genu-
ine, what could better testify to the confidence of pagans even under a Chris-
tian emperor, than the celebrations at Emona when Theodosius visited the
city? There went out to meet him "senators notable in their snow-white
dress, the flamens venerable in city-purple, the priests standing out because
of their peaked hats.'' 7 It was officially a pagan reception.
The Danube lands suffered such destruction in the third century from
barbarians and so much more, from Christian ardor, in the fourth and fifth,
that idols and temples of the late Empire are hard to find in one piece. Out-
side Pautalia, the sanctuary of Zeus and Hera, enclosing several temples
alive and well over the fourth century, at the end received such a smashing
that only two simple capitals remained intact. They were of granite. 8 Mean-
while, the inscriptions of the region change under Constantine. Pagan
gravestones and ex-votos are no longer set up, while Christian epitaphs and
increasingly rich burials come into view. All this is best explained by imag-
ining a takeover of the town senates by new converts, while humbler folk
and the countryside remained little affected.
In the east, the continuing prominence of paganism in Athens is well
known; of Asclepius's cult, too, at various shrines. 9 In Syria, in its greatest
city, Antioch, the masses appear to have been Christians for a long time, by
(let us say) A.D. 375; the city senators, mostly Christian; but the peasants
over the countryside generally a mixture, possibly half and hal£. 10 In Egypt
of the mid-fourth century, an army commander had plenty of Christians,
including priests, who were in courteous correspondence with him; but in
his headquarters' chapel an image of the goddess Nemesis presided, and his
personal servant took oath by the gods, plural. Perhaps that meant nothing.
A deacon of the Church toward the same date "swears by the divine and
holy Tyche of our all-conquering Lords," the emperors. 11 Inside people's
minds the most contrary beliefs might coexist, at some moment suddenly to
be recognized as mutually intolerable. In that same fashion Christians and
pagans lived together for generations in the cities of the Empire, in peace
134 Epilogue

disturbed rarely by spasms of frightful violence, both before and after

312.
Gaza, around the end of the century, illustrates how the religious forces
might change their balance. Being an out-of-the-way place, it had been
hardly touched by Christianity. The emperor Arcadius is imagined saying,
"I know very well that the city is full of idols. But it loyally pays its dues and
contributes much [to the treasury]. If we suddenly impose a reign of terror,
they will flee and we will lose all that tribute. Still, if you wish, we will
squeeze them gradually, removing the dignity of their idolatries and other
municipal offices." To that, an impatient missionary replied with more
direct action, performing miracles in the town. Thereupon "some of the
pagans, seeing what great wonders God had performed, now believing ... ,
shouted, 'Christ only is God, He only has prevailed!'" 12
Enough of these disconnected episodes and glimpses, picked up around
the whole Empire. Since so much of the raw fact, the archaeological, epi-
graphic and papyrological material, remains still unevaluated, to say noth-
ing of its being assembled into broad regional surveys, there is little that can
be said at this point with any confidence. Just a few broad statements may
be hazarded.
First, as Peter Brown has said, "the historian of the later Roman church
is in constant danger of taking the end of paganism for granted." u His
warning gives shape and weight to all the scattering of facts just reviewed-
a warning the more needed because what it warns against appears so thor-
oughly reasonable. We naturally assume that such wonders as were worked
at Gaza had long since had their effect in other places. It surprises us, indeed,
still to find an almost wholly pagan town in the Holy Land in the fifth
century (but there remained others in the sixth I). If wonders had not availed
everywhere, it surprises us that laws against sacrifices, seizure of idols by the
state, and so back through the crowded chronicles of violence to Constan-
tine's own reign, should not have turned everybody Christian. But pagans
survived, unterrified. To the emperor's face they shouted, "The gods save
you!" Their priests turned out to welcome him on his journeyings. The
higher reaches of his government-not only barbarian generals like Arbo-
gast, but civil functionaries-were predominantly pagan before Julian as
after, too-and not in remote regions but in the eastern provinces them-
selves. The enormous thing called paganism, then, did not one day just
topple over dead. No metaphorical description of that sort will apply, even
if the process of demise is somehow stretched out over a generation past the
great conversion of 312 (and more than a generation, it is not the fashion to
 The Manner of Death of Paganism JJ5

suggest: Julian must be a dreamer, doomed, born out of his time, or the
like).
Second, distinction must be made between two elements in religion: the
perceptible, meaning the activities and all that those activities imply in
consequence of a person's being an open participant in some belief; and the
debatable, as I have called it, meaning those feelings and thoughts that
accompany a person's acknowledgment of a god; and a person may grow
up into certain religious activities without ever believing, or come to be-
lieve but never participate, so the priority of the one or the other element is
not fixed, nor even their occurrence as a pair. It was the perceptible and its
pomp and privilege enjoyed by leaders of a congregation that the emperor
Arcadius thought to strike at, so as to turn them from their pagan ways;
while the missionary who asked for his help aimed his efforts at another
point.
To force acknowledgment of their God, Christians commanded remark-
able means. Tertullian had called the blood of martyrs the seed of the
Church. Backhandedly, the emperor Julian had shown later that he attached
the same importance to martyrdom, through what he decided not to do
even more than through what he did. Surrender of one's life for one's faith
was bound to work most powerfully on witnesses. It was quite beyond
nature, a wonder. Along with miracles of other kinds and exorcism rou-
tinely (above, pp. 95-96 and 50), it goes far to explain how thousands, not
mere dozens of the uncommitted, might be induced really to look at another
man's god and really to open their minds to a new reality. Indeed, if we
relied only on pre-Constantinian sources, we would suppose that such
supernatural acts accounted for very much the greater part of all conver-
sions. There should be nothing surprising, then, in the missionary's miracles
in Gaza at the turn of the fourth century.
But many students of the ascendancy of the Church (being thus by impli-
cation also students of the dying away of paganism) would rather erqpha-;
size the importance of the social ethics, status-structure, congeniality, morat
challenge, and psychological support offered to those who joined the
Church. 14 Perhaps it is all a matter of definition: When does the process of
conversion cease and the process of living in a faith begin? Or perhaps the
two elements of religion distinguished above cannot be known and sepa-
rated in enough instances for us to tell which best explains the fading of
pagan cults.
Anyway, Arcadius had no doubts; and by his day existed, thanks to his
Christian predecessors, many, many laws aimed against the prominent sort
116 Epilogue

of pagans. They could not legally pass on their estates, they could not enter
on profitable careers, their sanctuaries had been stripped of land and wealth. 15
As added penalty, in the good gold solidus dispensed by Constantine and
his successors to the Church, pagans could recognize the metal confiscated
from the treasuries of their gods. They were financing their own destruction.
And this they were made to do in a period that was, for quite secular and
impersonal reasons, radically inimical to the pagan "Establishment"-
that is, inimical to local urban magnates. Their difficulties had sharpened
in the third century. By the fourth, many of them had been driven into the
refuge of their villas, or into bankruptcy. In either case, they were not on
hand to pay the bills for what, some pages back, were called the upper parts
of paganism.
Upon those upper parts our account has naturally focused, because they
were the most visible in the surviving record. They were also the most
fragile. Like a coral reef, to live and grow further, they required conditions
favorable within quite narrow limits. When conditions changed, life and
growth must end. The more substantial, older, primitive parts of the pyra-
mid of beliefs, however, lying at a leve_l below the reach of our inquiry, died
more slowly-just when, no record declares.
From the triumph of Christianity, it is natural but not certainly right to
reason backwards: Christianity in its now familiar outline, Christianity as
it was "supposed to be," prevailed because it was intrinsically better.lt was
freely espoused by people who could see its superiority. But that view
should not involve the quite crude error of supposing the now familiar out-
line to have been already clear in the period of our study. In fact, of course,
the Church was undergoing constant change in its early history as in its
later. The marked prominence of exorcism in its outer face, and of demon-
ology in the inner, faded rapidly away during the fourth century-to name
two areas of development that have received mention above.
Crude error avoided, there remain several further points of doubt. First, is
it possible to define, almost a priori, major human wants to which answer
must, or can only easily, be made through religion? So, if nothing in all the
variety of paganism answered some of those wants, but Christianity did, the
rise of the latter could be explained. As one illustration (pp. 53-57-but
several others, I think, offer themselves from the preceding pages or have
been discussed in other books): no pagan cult held out promise of afterlife
for the worshiper as he knew and felt himself to be. Resurrection in the Oesh
was thus a truth proclaimed to the decisive advantage of the Church. In
making any such assertion, however, much care would be needed against
 The Manner of Death of Paganism IJ7

attributing to the third century social and spiritual needs that were created
rather than answered by Christianity.
Second, is it possible to weigh the impact of two adventitious factors, the
destructive political and economic forces at work upon the more prominent
parts of paganism after 250 and the constructive dynamic of Constantine's
reign in favor of the Church? Together, one might argue, these factors over
three-quarters of a century coincided with, and were very nearly enough to
account for, the great changes in the Church's fortunes. But the argument
must be tested.
Third, from the study and especially from the dating of pagan testimonies
post-312, is it possible todeterminejusthowmuchandhowearly paganism
retreated? The most casual reading in the late Empire suggests that there is
plenty of material to reward inquiry on this front.
Those are points that occur even to an inexpert observer. It is curious,
when one stops to think about it, that the success of the Church in taking on
historic proportions has never been described with much care. 16
 Abbreviations

Citation in the notes is abbreviated to the form "Jones (1967) 100," or,
where I cite two works by the same author published in the same year, I add
"a" or "b" to the date. The abbreviated citations can, I hope, be expanded
without difficulty from the bibliography. A few works of reference receive
their standard abbreviations, all of which may be found in the Oxford
Classical Dictionary. There too, or in the Greek-English Lexicon of H. G.
Liddell and R. Scott, may be found the usual abbreviations for primary
sources, including collections of papyri and inscriptions. The only ab-
breviation I use which may not be widely familiar is ANRW, for the
ongoing series of volumes on the Aufstieg und Niedergang der romischen
Welt, edited by H. Temporini and W. Haase.

1J9
 Notes

CHAPTER 1: PERCEPTIBLE
I. Finding Order in Chaos
I. Degrassi (1962) 995-1004, passim.
2. Collart (1937) chap. 4. Note the absence of the oriental gods Mithra, Sabazius,
Jupiter Dolichen us; and compare, almost at random, such another city as Lambaesis
and suburbs, LeG lay (1971) 129-141, with epigraphic attestations of Saturn (161 in
number), Mercury next with 13 (to be explained, p. 132, by "Mercury" as the inter-
pretation of the local deity of agriculture, esp. oleiculture); Minerva (12); Silvanus
(II); Mithra (9); Fortuna (6); dii Mauri (6); and so forth.
3. Bosch (1935) 102-130.
4. Lucian, ]up. trag. 8£. and 12; d. some of the more exotic names again in
lcaromen. 24 and 27.
5. De laud. Const. 13.4f. (PG 20.1400)-a list similar to that given by Athenago-
ras, Leg. 14, offering a dozen deities ranging from Transjordan to Sicily and Car-
thage, the rest Greek. His purpose being to show idiosyncrasies, he offers some very
odd ones, as Geffcken (1907) 187 points out.
6. Pauli Sent. 5.21.1, prophets being very common, d. Dig. 21.1.1.9£. and other
texts in MacMullen (l966a) 142-155; Latte (1960) 328£.; and Parassoglou (1976) 261
and 266.
7. Athenag., Leg. I.
8. Harnack (1905) 9£. and passim; Geffcken (1907) 186.
9. Nicolaus of Damascus, in Jos., Ant. ]ud. 16.44, to Agrippa in 14 B.c., con-
sonant with the view Cicero indicates about the Jews (Pro Fiacco 69), sua cuique
civitati religio, etc. The pagan audience replies, in Claudius's declaration that one
"should safeguard traditional customs," Jos., Ant. ]ud. 19.290, and again, in A.D.
45, Claudius's piety. urges "each to offer cult according to ta 1tlitpta;" and so to
Celsus the pagan defending the Jews ca. A.D. 178, Orig., C. Cels. 5.25, on the
grounds of their loyalty to their own native customs, ta 1tlitpta. "It is not pious to
dissolve what has been assigned from the beginning to each locality ... ," d. again,
5.34.
10. Just., I A pol. 12, setting "truth" against "customs"; Celsus, in Orig., C. Cels.
5.34, d. 8.28; Clem. Alex., Co h. ad gent. 10.73 and 12.91 (PG 8.201 and 237); Tert.,
Ad nat. 2.1, and Apol. 6.1, 6.9, and 6.10, where pagans claim to defend customs of
maiores and patres; Min. Fel. 6.1; Origen replying to Celsus, C. Cels. 5.35; Porph.,
in Euseb., Praep. ev. 1.2.1; Galerius in Euseb., H.E. 8.17.6; Licinius in Euseb., Vita
Const. 2.5, Constantine ta 1tlitpta 1tapaa1tovot\aa~, t~v d9eov eiA.eto 86~av, KtA.,

141
142 Notes to Pages 2-4

d. Orat. ad Sanctos 9 (PG 20.1253); Euseb., Praep. ev. 1.1 (5a-c); and Julian often,
e.g., Ep. 89a.453B, 89b.302B; C. Galil. 238D.
11. Praep. ev. 4.1 (130C), trans. Gifford.
12. Ad Marc. 18 and similar statements showing veneration fortd nat pta in Sext.
Emp., Adv. dogmaticos 1.49, and Dio 52.36.
13. Ad Marc. 14.
14. Plut., Moral. 756B; for his friends, d. 402E, 416C, and 1125E; good commen-
tary by Andresen (1955) 264 and Dreyer (1970) 50 n.164; and the often-quoted, pas-
sionate injunction in C/G 5041, lines 4f., "Consider first the ancestral gods, and
worship them," etc.
15. Ep. ad Diognetum 1.1 and 8.7-9.2; Suet., Nero 16.3; Pauli Sent. 5.21.2; Euseb.,
Praep. ev. l.l (5a); Julian, C. Galil. 306A.
16. Porph., De abst. 2.18, on Hesiod frg. 5.169; Tert., Apol. 19.H.; also just., 1
Apol. 23, 44, and 59£.; Theoph., Ad Autol. 1.14, 2.3, 3.1, etc.; Tat., Ad Graec. 29, 31,
and 40, and in Euseb., Praep. ev. 10.11 (495bf.), d. 10.9; Min. Fel. 6.3; Orig., C. Cels.
1.14, 4.36, and elsewhere; and Eusebius revealing his own reasoning, H.E. 5.28.2,
"they wished to make it holy, as if it were ancient." For discussion of such texts, see,
for example, Andresen (1955) 63£. and 70, and Pepin (1966) 232£. Compare Tac.,
Hist. 5.5, Jewish beliefs antiquitate defenduntur, or Sext. Emp., Adv. math. 1.203£.,
using age as the canon to determine the best usages in Greek.
17. On patrius, !tlitpto~, natpcjio~, see as a mere selection IGR 3.89 (A.D. 68), to
Zeus and Hera, tot~ !tatpiot~ eeoi~ Kai !tpOEO'trocnv tf1~ n6Aero~ at Amastris, the
latter term often by itself, e.g., Zeus and Hecate are !tpOEO'tOOtE~ eeoi tf1~ !t6A.ero~ at
Panamara, in LeBas-Wadd. 519-520 (ca. A.D. 200-cf. Laumonier [1934] 85), or
Dionysus in Teos is termed "the presiding deity of the city," in Robert (1937) 25 n.2
and 30f., as Artemis in Perge is termed livaooa and npoecm:i>O'TJ~ tf1~ n6Aero~, d.
Pace (1923) 304; further "ancestral" gods in /G 10, 2.199, Thessalonica in the mid-
third cent., "to the ancestral god Kabeiros"; Seyrig (1941) 246, Iarhibol is "ances-
tral" at Palmyra, A.D. 192/3; Rehm (1958) 117, Aphrodite at Didyma; Forschungen
in Ephesos 2 (1912) 126, "your ancestral deity Artemis" at Ephesus in A.D. 200/205,
d. Xen., Ephes. 3.5.5, her priestess calls her "the ancestral" (second cent.); Hatzfeld
(1927) 74 and 77, Zeus of Panamara; LeBas-Wadd. 2576, Helios at Palmyra; BGU
362 (A.D. 215) p. vi, "our ancestral crocodile god the great Souchos"; Men "the
ancestral" in texts of Antioch, Lane (1971-1978) 1.102-110 and 3.64; and MAMA
3.56, Helios "ancestral" in Cilicia (with editor's comments). In the west, see IRT
289, Hercules and Liber pater concealing Punic deities, like Apollo patrius in
Merlin (1908) 25, of the second or third cent.; Liber pater, lar Severi patrius at
Lepcis, in Guey (1953) 341£., on /RT295; C/L 8.21486, dii patrii et Mauri conserva-
tores; and 8.19121, Baliddiris in Numidia. Possessive epithets like "city-guardian,"
"god of our people," etc., can be easily found, e.g., in IGR 4.1571, Robert ( 1950) 68,
and LeBas-Wadd. 400.
18. IGR 1.421 =OG/S 595, A.D. 174. D'Arms (1972) 258 and 267f.,.encountering
cultores dei patri in Puteo1i, ca. A.D. 340, struggles with the strange notion that there
was only one particular deity termed "ancestral" in Italy. This example cannot be
surely identified. For other instances of gods called ancestral by worshipers abroad,
see Seyrig (1970) ll1, on /G 14.962, the Palmyrene Arsou invoked in Rome; IGR
1.45 and 46, the same, in A.D. 235; /LS4341 =C/L3.7954, Palmyrenegods invoked by
a Palmyrene in Dacia, cf. Birley (1978) 1518, adding other similar texts from Lam-
 Notes to Pages 4-7 14J

baesis addressed to eastern gods by an official; IGR 1.1293, a Roman commander

invokes dii patrii in Egypt, 29 B.c.; IG 4, 12.417, a priest in Epidaurus invokes Hera
of Argos and Zeus of Nemea in A.D. 297; IGR 1.387, Gaza invokes its gods to honor
Gordian, in Portus; ILS 4349=CIL 3.3668, in Pannonia on the Danube, to Arabian
deities by Arabians(?), d. Salac (1956) 168f.; in Augsburg, to Elagabal by a Syrian,
c£. Radnoti (1961) 385 and 401; to Egyptian deities by an Egyptian (to judge from his
name) in Miletus, cf. Vidman (1969) 147, though Isis and Sarapis, "ancestral" in
Cadyanda, are taken as naturalized there, by Dunand (1972-1973) 3.7; to Syrian
deities, 7tatptol, Atargatis and others, by a cult group in Rhodes, IG 12,3.178; at
Intercisa in A.D. 214, to Elagabal by Syrian soldiers, d. Fitz (1972) 101; to invictus
patrius, Mithra, by an eastern officialin Italy, cf. PRaum (1960-1961)899and901, on
CIL 5.5797; and last, Fronto of Cirta, in Rome, praying to "dii patrii and Jupiter
Ammon, god of Libya," in his Ep. ed. Naber 121 (Loeb ed. 2.135).
19. Examples in Robert (1937)319, atAphrodisiasapriest who "made sacrifice to
the ancestral deities while himself offering prayer for the health" of the emperors;
Vidman (1969) 164, Hadrianeia (Mysia) in A.D. 206; IGR 3.348 (Sagalassus), 664
(Patara), and 4.1497 (an obscure pagus in Asia Minor), all under Pius.
20. SHA Had. 14.3; Milik (1972) 10£.; Ohlemutz (1940) 130; IGR 4.214; Rehm
(1938) 78, 80, and (date) 83; and Ianovitz (1972) 37f. Emendations to Rehm's texts by
Gregoire (1939) 321 and Wilhelm(1943b) 168donotaffectmyuseofthem. Note also
the interesting text in Reinach (1906) 86, where the governor of Asia in the 220s
promises offerings "to your ancestral goddess on behalf of the safety and eternal rule
of our emperor," at Aphrodisias.
21. RE s.v. Tarsos (Ruge 1932) 2415; Robert (1946b) 39; Dufournet (1974) 382f.
22. IGLS 2960; Artemis Pergaea widely famed, d. Robert (1948a) 64 and Onur·
kan (1969-1970) 289 and 294; on Mothers, cf. Graillot (1912) 383, and, more gener-
ally, 381-399 and 435.
23. Laumonier (1958) 714f., and 29 Apollos in the same confined corner of Asia
Minor, ibid. 716f.; for Mars, see the Indices to CIL 13,5 (1943) and 12 (1888), listing
some dozens of local Mars versions; Suss (1972) 169 n.39; and Hatt (1979) 125-138
passim.
24. Min. Fel. 6.1; Tert., Apol. 24.7.
25. Ulp., Regulae 22.6 (FIRA2 2.285), one each in Italy, Spain, and Africa, and
the rest are Minerva in Ilium, the Mother at Mt. Sipylus, Diana in Ephesus, Nemesis
in Smyrna, Apollo at Miletus; cf. Marcus Aurelius, Ep. ad Front. 3.9 (Naber p. 47),
praying to "all the governor-gods of all peoples" and to Aesculapius at Pergamon
and Minerva=Athena at Athens.
26. I have had to make three somewhat arbitrary decisions: from the Saturn count
I have excluded the 156 inscriptions of 5 Saturn sanctuaries; from the Cybele line I
have excluded 55 taurobolia inscriptions from particular shrines; and from the
Fortuna line I have not excluded 44 Fortuna-Primigenia inscriptions from Prae·
neste, the cult center.
27. On Nemesis, whose centers lay in Smyrna, less prominently in Antioch and
Alexandria, d. RE s.v. col. 2356; commonly near theaters and amphitheaters in the
west, ibid. 2359-2361.
28. Jullian (1908-1926) 6.28£. and Hatt (1965) 90 and 106f., bothcallingTeutates
"le dieu national"; further, Zwicker (1934-1936) 177 and passim.
29. Lambaesis supplies no texts at all for Liber, but all for Dolichenus. All the
144 Notes to Pages 7-9

Gallo-German Dolichenus inscriptions come from frontier centers like Mainz,

Neuss, Bonn, KOln, and Wiesbaden. The one city Portus-Ostia, if included, would
raise by 50 percent or more the Italian totals for the five cui ts specified, of Isis and the
rest. On the other hand, one might expect to find Lyon atypical, but its deities are in
fact in proportion with the Gallo-German totals, barring a slight exaggeration of
Apollo; but note nearly half of the Cybele texts come from Lactora (Aquitania)
alone.
30. Penates in combinations, in Boyce (1937) 104-though the numbers are
small: for Fortuna, only 13 lararia, etc.
31. IGR vols. 3-4. How little the data are suited to statistical analysis appears
from the totals, 527 inscriptions for all these gods, as opposed to 5,897 Latin texts
used for the bar graph. Note also that a quarter of the inscriptions for Hera come
from the one site where she presided, Ilium, and more than half of Aphrodite's from
Cyprus, where that goddess too had her most famous shrine.
32. MacMullen (1974) 89-97.
33. Tert., Prax. 3, simplices, imprudentes, idiotae, or iotrotat and oA.tyoJ.!aOet~,
Iren., Haer. 2.26.1, quoted in Lebreton (1923) 483-488, d. Min. Fel. 2.4, vulgus
superstitiosum, and 3.1, imperitiae vulgaris caecitas, and Tert., A pol. 49. 4, caecum
hoc vulgus, etc.; Lucian, Alex. 30, totrotai nve~ ohcetat; Hermotimus 1, "the great
mass of folk of no account"; Dea Syria 8, 11, 28, etc., and Athenag., Leg. 15, Art.,
Oneir. 4 praef., o{ 1toA.A.oi contrasted with persons better informed; Euseb., Praep.
ev. 4.1 (132b), "the imbecility of critical powers among the masses, o{ 1toA.A.oi, the
weakness of reasoning and gullibility of the general populace, to 1tA.JiOo~," and
similar views expressed by Porphyry, ibid. 3.7.1.
34. Min. Fel. 12.7, with very much the same view and tone in Synes., Ep. 105 (PG
66.1488A). An egalitarian passage like Plin., Ep. 2.3£., is very rare, hardly matched
in our period by Dionysius of Alexandria in the 270s, Euseb., H.E. 7.25.26: remark-
ing on John's uneducated style, whose errors Dionysius refers to "not in mockery,
let no one think it," but solely as differentia for analysis. Other, later, passages are
discussed in MacMullen (1966b) llO£.; and exception is naturally made by Chris-
tians discussing Christ's disciples, e.g., Euseb., H.E. 3.20.5 (interesting) and 3.24.3,
or Demonstr. ev. 3.7 (PG 22.241).
35. Celsus, in Orig., C. Gels. 1.23; 3.18, 44, 50 and here quoted from Chadwick's
translation, 3.55; d. a similar violent passage in Ps.-Apul., De deo Socr. 3.122, or the
scorn in Apul., A pol. 9.1, 10.6, and 16.10, against rustici, or Diog. Oenoand. frg.l p.
3 Chilton; violent scorn matched by Clem., Co h. ad gent. 10.41 (PG 8.209), or simi-
larly Euseb., Praep. ev. 13.14 (692b).
36. Menander Rhet., Epideiktika 5 p. 337 Spengel, Rhet. graeci vol. III; an ex-
actly similar distinction in material to be presented is made by Alex. Rhet., ibid. III
p. 4, and put into practice by temple exegetes, Paus. 2.23.6.
37. Plut., Moral. 407C and 756C: Lucian, Amores 42; Juv. 6.511-591; Clem.
Alex., Paedogogus 3.4.25 and 27£. (PG 9.593 and 596); Min. Fel. 8.4; Athenag., Leg.
11; and Tat., Ad Graec. 33. Note in Juvenal and Clement the wealth of some of the
women depicted and, in Iren., C. haeres. 1.13.3 (PG 7.581), rich women are the
natural target of heresy-mongers.
38. On esoteric believers, see Plut., Moral. 352B (Isis priests); Art., Oneir. 4 praef.;
Philostr., Vita Apollon. 8.7.9; Orig., C. Cels.l.l8; Plot., Enn. 2.9.9andEunap., Vit.
sop h. 455; Clem. Alex., Strom. 5.9 (PG 9.89A-B and 88-95 passim); Porph., De antra
 Notes to Pages 9-11 145

3 and Ad Marc. 17 and 28; Firm. Matern., Math. 7.1.1 and 3; 8.33.2; Asclepius I and
10; Corp. herm. 13.13; PGM 4line 1872; Philo, Quoddeusimmut. 54; DeAbrahamo
147; Clem. Alex., Strom. 5.10.24 (PG9.100); Iren., C. haeres. 1.3.1; 1.24.6(PG 7.679);
Orig., C. Cels. 1.12 and 29; 5.15 and 29; Arnob. 5.32; Synes., Ep. 105 (PG66.1488A);
Lebreton ( 1923-1924) 492, 504 and passim, d. Au!. Gel. 19.10.14, niceties of gram-
mar treated like the Mysteries, for initiates only.
39. Athenag., Leg. 5 and 24; Just., I Apol. 4, 20, etc.; II Apol. 8; Theoph., Ad
Autol. 2.3, 8, etc.; Tert., A pol. 14.4-7; Aristides, Apol. 13; Clem. Alex., Coh. ad gent.
2.126 and 7.I (PG 8.96 and 180); Lact., Div. inst. 1.4, "poets and philosophers which
it is usual to use against us."
40. A cliche, of course, d., e.g., Dio Chrysos., Or. 18.3, and earlier texts in Kroll
( 1924) 24-34; but acknowledged in debate, e.g., by Orig., C. Cels. 7.45, or by Euseb.,
Ad sanct. coelum 10.3.
41. For the ranking of "best-seller" writers, the Iliad first, and so forth, see Pack
( 1965) passim, on papryi (-fragments) surviving: of Homer, 688; next most numer-
ous, of Demosthenes, 81; of Euripides, 76; Callimachus, 50; Hesiod, 48; Plato, 43;
etc.
42. On Cornutus, see Tate (1929) 40£.; on Seneca, Attridge (1978) 67 n.159, the
work being used by Tert., Apol. 12, and Aug., Civ. dei 6.10£.; on Heraclitus, Tate
(1930) If., and Buffiere (1962) ix and xxxii; on Philodemus, Lesky (1966) 682, with
frgs. in Diels, Stoic. vet. fragmenta; on Diogenianus, d. Theod., Graec. affect. curat.
6.8 and Euseb., Praep. ev. 4.3 (136df.); on Oenomaus, Attridge (1978) 56£. with note.
His work Wizards Exposed was used by Euseb., Praep. ev. 5.22 (213C) and Theod.
I.e., the treatise being evidently similar to another used by Hippolyt., Refut. 4.34
and (a third?) used by a certain Celsus who is quoted in Lucian, A lex. 21. On Por-
phyry, note his work on Images quoted in Euseb., Praep. ev. 3.9 and II (100a£. and
108b), etc.; and his own use of Numenius, in De antro 21 and elsewhere.
43. Porph., De abst. 2.56 and 4.16 and Deantro2and 21, on second-cent. treatises
about Mithraism; put in context by Turcan (1975) 23; on Isiacism, Nymphodorus
(?)in Clem. Alex., Strom. 1.21 (PG 8.832A); on Cybele and Attis, a section of the poet
Hermesianax, Paus. 7.17.8£., and the unknown writers Hicesius On the Mysteries,
in Clem. Alex., Coh. ad gent. 5.64.5, and Timotheus and others in Arnob. 5.5.
44. Philostr., Vita Apollon. 3.41 and 4.19; Plutarch in much of his commentary
on Hesiod's Works and Days, the frgs. in the Loeb edition, vol. 15 pp. 135, 197,217,
etc.; Porph., both in parts of his De abst., e.g., 2.56, and his Philos. ex orac., to be
found in Euseb., Praep. ev. 4.9 (145bf.) along with Apollonius's treatise again, ibid.
4.13 (150b); and Istros of the third cent. B.c., in Porph., De abst. 2.56, d. RE s.v. col.
2270.
45. On Apollodorus's use, see Geffcken ( 1907) xvii and 225, and Lesky (1966) 787;
on use of Varro, see, e.g., Clem. Alex., Co h. ad gent. 4; Tert., Ad nat. 2.1; Lact., Div.
inst. 1.6; and Cardauns (1978) 88.
46. Praep. ev. 4.2 (136b), c£. Porph., Philos. ex orac., passim, and below, p. 13.
47. Besides scenes from the later chaps. of Apul, Met., or Heliodorus, note Xen.,
Ephes., passim (dated to the second cent., p. xi of the G. Dalmeyda edition, or in
Lesky [1966] 864); Lucius of Patras whom Apuleius read, d. Cumont (1929) 97,
dated to the second cent. by Lesky I.e.; or the scenes of religious initiation in the
anon. second-cent. text in Hinrichs (1970) 29-31.
48. Paus. 1.13.8; Philostr., Vita Apollon. 4.30, a logos on Zeus, perhaps like Dio
146 Notes to Pages 11-12

Chrysostom's Twelfth Oration; and on the gigantic inscription, see Chilton (1967).
For the location of the text in the city and Diogenes' scattered friends, consult frg. 2,
V lines 12£.; frg. 15, II lines 3f. and 10; frg. 16 lines IOf.; frg. 51, III; and Chilton
(1971) xxi and xxxiii. On the religious parts, see frgs. 29-31; also Robert (1971a)
597-613; and M. F. Smith (1978) 39, bibliography to which Smith adds esp. frg. 54,
and page 54 on Rhodian correspondents, and page 43 on dimensions of the inscrip-
tion.
49. Hippolyt., Refut. 5.20.5.
50. Aug., Civ. dei 2.7,library in Cybele temples in Africa; Apul., Met. 11.7, Isiac
priest, grammateus, reads aloud from a prayer book, d. the Reader for a native
Batavian deity, Menon dvaytyvroaK(rov) in a Flavian graffito, Bogaers (1955) 241;
and further refs. in Stambaugh (1978) 587. In eastern provinces, Dio 75 (76).13, in
A.D. 200, assumes books in almost all the sanctuaries of Egypt; BGU 362 I 19
(Arsinoe in A.D. 214); perhaps POxy 1382 (second cent.) lines 19f., "the libraries of
Mercuri us" wherein is registered a miracle tale; and Orig., C. Cels. 1.12, the wise
men in Egypt learnedly investigate their received religion through many traditional
tractates. Aelius Aristides knew of vast collections of Sara pis's aretai in unspecified
"sacred archives," Or. 8.54, I p. 95 Dindorf.
51. Philostr., Vita Apollon. 7£., the Asclepius temple at Aigeai is the scene oflec-
tures on asceticism; Plut., Moral. 410 and 4120; logoi in many cities, apparently
temples (here mentioned), Aristides, Or. 25.2, d. public displays of learning at
festivals in Philostr., Vita Apollon. 8.18, and Dio Chrysos., Or. 27.5-6.
52. Theologues by title, attested in British Museum Inscriptions 3 p. 138, who
"seem to have to do with the celebration of the mysteries," at Smyrna-they are the
two sisters, mentioned with mystae, CIG 3200; BMlnscriptions, further, 4.481 line
295, at Ephesus in A.D. 104; IGR 4.353, at Pergamon; in Bithynian citiea, Robert
(1943) 184£.; C. Picard (1922) 208 and 249f.; and Plutarch consulting theologues at
Delphi, Moral. 417f.
53. Schuchhardt (1886) 428.
54. Herrmann and Polatkan (1969) 10, an endowment for a funerary cult, the
divine command for which is revealed in dreams to a man in Magnesia ad Sipylum,
date (p. 23) in first cent.; CIL 9.3513, 58 B.C., jupiter liber in the vicus Furjo.
55. E.g., CIG 3538=IGR 4.360, and Ohlemutz (1940) 61 and 77. where Pergamon
is relieved of plague by prescribed sacrifices and hymns, as the Clarian Apollo
directs, ca. A.D. 171; or in Kern (1900) 144 no. 228, late second cent., as "the god re-
vealed," cult shall be offered to Hera in a certain way, d. Ael. Arist., Or. 49.39, I p.
498 Dindorf, in A.D. 149, commands from Zeus.
56. Kern (1900) 140 no. 215; Sokolowski (1962) 59 and 67, second cent. examples
at Epidaurus and Sparta; Robert (1975) 308 and 325, an inscription from Sardis of
ca. A.D. 150 recalling a document of the fourth cent. B.c.
57. Milik (1972) 288, a lex sacra for a cult group in Palmyra; another, in Cilicia,
Hicks (1891) 234-236, for Sabazius in Augustus's reign; Sokolowski (1955) 52, a
"holy doumos" of Men in Maeonia, and ibid. 55, a similartext; idem (1969) l06f., on
the Men regulations just referred to.
58. Milik (1972) 287, in Palmyra, and 216, on CIL 6.52=ILS 4335, in Rome, ne
quis velit ... triclias inscribere; d. Sokolowski (1955) 55, rules for members of the
first-cent. B.c. cult group directed against poisons, philtres, abortions, or spells.
 Notes to Pages 12-14 147

59. Rules evidently posted, to be noted by entrants, e.g., at Olympia, Paus.

5.13.3f., or Blinkenberg (1941) 871, a lex sacra on a stele "just outside the entrance to
the shrine" of Athana Lindia, ca. A.D. 255=Sokolowski (1962) 159f. Typical prohi-
bitions: worshipers should "not stare around too much," not wear a helmet or shoes
or anything of goatskins, etc.; idem (1955) 186f., a text also discussed by Nock
( 1958b) 415-418, where, in the second- or third-century Dionysus s~rine at Smyrna,
abstinence is enjoined "from babies' food," i.e., milk, etc.; IG 22.1365-6=Sokolow-
ski (1969) 106f., Oliver (1963) 318, and Lane (1971-1978) 3.7f., all three scholars
discussing the Severan Attic inscription that bans garlic, pork, etc. Compare Sext.
Emp., Hypotyposeis 3.220f., on pork, goat, and other meat sacrifices, suitable or
not, and the learned conversation in Plut., Moral. 670E; further, SIG 3 736, of 91 a.c.,
addressed to the cult of Demeter and Kore in Andania (Messenia), women to be
unshod, their clothing white and not transparent, etc. Pausanias (4.33.4f.) justifies
my use of so early a text, since he found the shrine active in his day.
60. Sokolowski (1962) 43, the date being first or second century.
61. Above, n.59. For some good pages on the philosophy of sacrifices etc., see
Sodano (1958) xif.
62. As in the directives offered in inscriptions of Magnesia, cf. Kern (1900) 140
and 144, Oeo~ !xpnaev; Porphyry taking dictation from Apollo, in Euseb., Praep.
ev. 4.9 (145b-146b) and in De abst. 4.18 crediting leges sacrae to the gods; Lucian,
Dea Syria 15, where cult forms are given by Attis; and Soc., H.E. 3.23 (PG 67 .204),
where an unnamed oracle dictates Attis liturgy to Rhodes.
63. Sokolowski (1955) 52, "by the gods' command"; ibid. 53, "to Dionysius in a
dream"; IG 22.1365, cit., "the god choosing him"; Sokolowski (1969) 224, directions
by a worshiper at Dardanus, "which he inscribed by the goddess' command," her
name lost; and in Africa, Merlin (1916) 263f. =ILAfr. 225, iussu domini Aesculapi
... , quisque intra podium adscendere volet a muliere, a suilla, a faba, a tonsore, a
balineo commune custodial triduo; cancellos calciatus intrare nolito.
64. Sokolowski (1962) 2011£., a third-cent. text from Ephesus beginning, "Head-
ing of ancestral law." For the government to intervene was only a police measure,
seen in Calder (1935) 217, where the proconsul forbids molesting sacred birds at
Attouda in A.D. 77.
65. Robert (1968) 568f., 576,584, 589f.; idem (197la) 597 and602-609: the datable
oracles are Severan, but thought and wording are in part Hellenistic. Professor
Robert's discussion and the work of original discovery by G. E. Bean and others are
enough to restore one's faith in scholarship.
66. Milik, above, nn.58f., and Gawlikowski (1973) 57, at the temple of Bel; a ban
on entering shod into the Saturn temple, etc., in Charles-Picard (1954) 130f. and
138, cf. other rules in ILAfr. 225, cit. above; in Italy, CIL 9.3513, cit. above, and
5.4242 (Brescia, to Dolichenus) and 3.3955" (Siscia, to Jupiter Heliopolitanus), both
the latter texts banning sacrifice of pigs; and CIL 3.1933 =ILS 4907, Salonae in A.D.
137, cf. CIL 12.4333 II lines 21£. (Narbonne, A.D. 12), "the same laws in otherrespects
for this altar and notices as for Diana on the Aventine," and CIL 11.361, at Ari-
minum, h(aec)a(edes) S(alutis)A(ugustae) h(abet) l(eges) q(uas) D(ianae) R(omae)
in A(ventino), the expansion of the abbreviations accepted by Borghesi, Bormann,
and Wissowa, in RE s.v. Diana col. 333.
67. CIG 2.2715 (with the term dpeta{) cited by Longo (1969) 1.17, d. Strabo
148 Notes to Pages 14-16

17.1.17, "some writers record the aretai of the oracular answers" at the Sarapis
temple of Canopus. Of the long bibliography on aretalogy, I need cite only Engel-
mann (1976) 97-108; Grandjean (1975) 8-ll and 102; F. Dunand (1972-1973) 3.83-
87; Malaise (1972) 173£.; M. Smith (1971) 175-177; Witt (1971) 101£.; Des Places
(1969) 164-168; Vidman (1969) 40f.; Festugiere (1949) 209f. and 231-233; Robert and
Robert (1946-1947) 343£.; Harder (1944) 8, supplying the text which I translate
immediately below, and much else, passim; finally, Cumont (1929) 232f. n.6. Note
POxy. 1382, a narrative of the arete of (Zeus-Helios-) Sara pis in the second century.
68. Apul., Met. 11.25, where the hymnic nature is well seen by Tran Tam Tinh
(1964) 109 n.3, the passage "crystallizing devout beliefs expressed in aretalogies and
hymns," best printed by setting off successive lines, each to begin tu ... , te ... ,
tibi . .. ; Poxy. 675; PS/844; and Mesomedes, in Powell (1925) 198 and Lesky (1966)
811, the poet being a freedman of Hadrian whose works occur in manuscripts with
musical notation.
69. Apul., Flor. 18 and Apol. 55.
70. Riese (1894) l.26f., identified by line 15, merito vocaris Magna tu Materdeum;
Page (1962) 520£. no. 129, third cent.; Powell (1925) 196, Hymn to Fortuna; Heitsch
(1961-1964) 2d ed. l.l57, hymn to Cybele, in Hippolyt., Refut. 5.9.9; and Meso-
medes' To Helios, Lesky (1966) 811.
71. Diog. Laert. 5. 76, Sarapis hymn, evidently centuries old but still being sung;
Lucian, Demosth. 27, Isodemus's hymns to Aesculapius; Ael. Arist., Or. 47.30, I p.
453 Dindorf, "the ancient hymn beginning, 'I hail Zeus ... "';and Laumonier
(1958) 402, at Stratonicea "the customary hymn to the goddess" Hecate.
72. /G 42, l.l32-l34; Maas (1933) 148-155 and 160£., the stone of the late Empire;
and various ancient testimonia to the Sophocles text, cited ibid., e.g., Philostr., Vita
Apollon. 3.17, it is sung (present tense) in Athens to Asclepius; Oliver (1936) ll2-
ll4, mentioning also (ll5-ll7) another widely diffused but anonymous hymn to
Asclepius of the same period, appearing in inscriptions from Roman times and
from many locations-Dium in Macedonia, Ptolemais, Athens, etc.
73. Wiegand (1932) 53£.; Ohlemutz (1940) 148f., the date in the late 170s; on the
attribution to Aristides, ibid. 149 n.78. Aristides collected his hymns in a lost book.
Cf. Or. 26.36, I p. 153 Dindorf, where Asclepius tells him to compose hymns for that
god himself and Hecate, Pan, and others also. "And there came a dream from
Athena, too, containing a hymn to the goddess with this beginning: 'Youths, come
to Pergamon ... ,'and another to Dionysus, of which the chorus was 'Hail, Lord
Dionysus, ivy-crowned.' "
74. Many texts too early for my period of study, e.g., Edelstein and Edelstein
(1945) 1.328, on /G 22.4473, first cent. s.c. or A.D.; butfrom later times, Edelstein and
Edelstein 1.331-334; Kaibel (1878) 433£. no. 1027 (second-third cent.); /G 22.4533
(third cent.); C/G 2342=/G 12, 5.893 (second-third cent., to Apollo); Mitford
(1971) 197, second cent., to Antinous; and in the western provinces, the brief
verses in C/G 6797=/G 14.2524, Autun; RIB 1791, on Hadrian's Wall, to the dea
Syria (pace Wright, ibid.); and GIL 8.9018=Buecheler (1895-1897) l.l2l no. 253,
Mauretania in A.D. 246 to Jupiter Ammon, Pluto, and Panthea.
75. Quandt (1962) 44, for origin and date of the collection; cf. Clem. Alex.,
Strom., 6.4.96 (PG 9.253), the Egyptian cantor typically has at hand a hymnal; and
note a good deal of excerpts and influence from hymns in magical papyri, cf.
Nilsson (1947-1948) 60£. and Heitsch (1961-1964) 2d. ed. 1.171.
 Notes to Pages 16-17 149

76. To discuss the interior face of the Mysteries would carry me beyond my topic,
but d. Hippolyt., Refut. 5.9.9, cited again below, n.92, and Firm. Matern., De errore
profan. relig. 22, cited by many scholars of mystery cults, e.g., Cumont (1929) 226
n.46.
77. CIG 3538, on the Zeus temple wall at Pergamon, ca. A.D. 170; IG 22.4533, for
relief from pestilence, Severan date, at Athens; Dain (1933) 67-69, for a thanksgiving
ceremony at Heraclea near Latmos; and the implications i~ the description of some·
one as "victorious poet, composer of verse and rhapsode of the divine Hadrian," in
Clerc (1885) 125, from Nysa.
78. Parade paeans in, e.g., Sokolowski (1962) 203f., third cent. Ephesus; morning
paean to Dionysus, in Robert (1937) 20£., in Teos=LeBas·Wadd. 90; toAsclepiusin
Athens, DesP1aces (1969) 168, second cent. or Aelian frg. 98(ed. R. Hercher,2.233f.);
or to Asclepius in Pergamon, Ael. Arist., Or. 47.30 (p. 383 Keil).
79. Ael. Arist., Or. 48.2 and 50.50, I pp. 465 and518Dindorf; Clem. Alex., Strom.
5.8.47 (GCS 15 p. 359); and in inscriptions, Latte (1931) 134 and Robert (1955) 86
and 89 (also on gems).
80. Ps.·Lucian, Lucius 37f.; for Kore in Alexandria, Epiphanius, Panarion haer.
51.22.10; for any paean, RE s.v. Paian (Blumenthal, 1942) col. 2345.
81. "First-harper and carrier of the flame for De1phinian Apollo," honored in an
inscription, Pekary (1965) 122f., comparing also Inschr. von Didyma 182.
82. Quasten (1930) 63-65; Malaise (1972) 122£., 147, and 150f.
83. De die natali 12.2 (pp. 21£. Hultsch), written in 238. The passage is cited and
compared with other contemporary evidence for the use of instruments, in Quasten
(1930) 4 and 7f. Note SIG3 589 line 46 (Magnesia on the Maeander), advertisement of
a "concert with piper, whistle-player and harpist," at a lectisternium in the second
cent. B.C.
84. Aside from much scattered evidence for tibiae in old Roman worship and for
much more music of several sorts in Isiacism in Italy or anywhere else, the lines in
Firm. Matern., Math. 3.5.33, may speak for the Latin west as a whole in their
mention of hymnologi et qui laudes deorum . .. decantent. A collegium cantatorum
serving Apollo is once attested in Rome, cf. Marchetti Longhi (1943-1945) 81£.
Practices in Syria appear in Lucian, Dea Syria 44 and 50£., at Hierapolis, cf.
hymnodes serving a Syrian deity (though in Athens, SIG3 1111 ), and, in Palmyra,
Milik (1972) 153. In Africa, note the pipe accompaniment to sacrifices honoring
Saturn, in LeGlay (1966) 345.
85. Robert (1945) 105, evidence unknown to recent works that deal with the two
Delphic inscriptions of the later second cent. B.c.: Pohlmann (1960) 80 and Neu·
becker (1977) 149£.
86. Hymnodoi in I. Uvy (1895) 247f., at Smyrna and Ephesus, and in Nilsson
(1945a) 66, who adds other cities and deities; serving the imperial cult alone, IGBulg.
666; IG8ulg. 2 15ter and 17; IGR 4.1608, cf. Mellor (1975) 192; but doubling with
another cult, IG 14.1804 (A.D. 146, imperial cult plus that of Sarapis, at Rome); sub-
ventions to a guild of imperial cult hymnodes, in Keil (1908) 101-104; imperial cult
hymnodes at Smyrna shared with the gerousia, ibid. 110 concerning CIG 320lline
5, just like the sharing of hymnodes of Artemis at Ephesus with the city senate and
gerousia, C. Picard (1922) 251.
87. For the god Men at Athens, see Oliver (1936) 108, third·cent. evidence; maid·
ens and matrons in second-cent. Olympia, Paus. 6.20.3; ephebes in Ephesus, Keil
150 Notes to Pages 17-19

(1908) 107 n.l7; ephebes singing to Dionysus in Teos, LeBas-Wadd. 90; in Perga-
mon, to Zeus and other gods, CIG 3538 of ca. A.D. 170; Laumonier (1958) 402 gives
the evidence for a thirty-boy choir, cf. Keil (1908) 109, a twenty-four hymnode choir;
Lucian, A lex. 41, boys recruited to a shrine from neighboring cities, and boys at the
Pergamene Asclepieion in Ael. Arist., Or. 47.30 (p. 383 Keil); for Claros and Pana-
mara, see below, the Clarian choirs being of seven or regularly nine boys, Robert
and Robert (1954) 118 and 215. Whenever we find a hymnodidaskalos it is likely we
are dealing with an amateur choir, e.g., at Pergamon, Festugiere (1935) 205.
88. From Tabai, Heraclea, Laodicea, etc., ca. A.D. 120-185: C. Picard (1922) 262
supposes delegations to have hired a choir "trouve sur place," but Robert and
Robert (1954) 115, 203-210, and 215 offer a better view; also Robert (1969) 299-301;
and Wiseman (1973) 168 shows consultation from Stobi without a choir used.
89. Clerc (1885) 124, an inscription from Nysa mentioning a "poet, rhapsodeand
theologue," who contributed books to the guild ofhymnodes; OGIS513, Pergamon,
with note ad loc. on hymnodes/theologoi in Smyrna and Rome; and C. Picard
(1922) 249.
90. Nilsson (1945a) 67, citing inter alia SIG51109 line 115.
91. Menander, De hymn. If., ed. Spengel, Rhetores graeci 3.333, and Epideiktika
4, ibid. p. 390; cf. Alex. Rhet., ibid. pp. 4-5, like Menander distinguishing between
two accounts to be offered, one for the many, the other for the few, and detailing
what characteristics of a god a speaker must cover: derivation, age, identity with
other-named manifestations, powers and deeds, etc.
92. Hippolyt., Refut. 5.9.8f., where the setting is somewhere in western Asia
Minor; for the type of occasion and large audiences, d. the inscription in Keil ( 1908)
107 n.l7, "the ephebes singing the hymn in the the[ater were graciously] heard" by
Hadrian at Ephesus; and Plut., Moral. 417F, "the Delphic theologoi believe that a
serpent there once fought the god over the place of the oracle, and they let poets and
other writers [i.e., rhetors) compete in recounting these things in the theaters."

2. Attracting Crowds
l. The best discussion is by Schleiermacher (1966) 205-212; the most recent,
Hatt (1979) 131-133; d. also Ternes (1973) 7£. and 15 n.73. These three concentrate
on the Rhine and Moselle area. For Gaul, see G. Picard (1970) 185; C. Picard (1955)
231-234, on Vienne, the interpretation controverted, d. Vermaseren (1977) 134; and
Formige (1944) 70-72 and 92.
2. Golfetto (1961) 48f., cf. Charles-Picard (1954) 161: "on imagine de veritables
drames liturgiques."
3. Tert., De spect. 10.2; Apol. 15.4.
4. "There were no Roman games, even theatrical ones, that were not in some
way religious" -Gage (1955) 398. Compare Ps.-Cypr., De spectaculis 4, quod enim
spectaculum sine idolo, quis ludus sine sacrijicio . .. ? Bardenhewer (1912-1932)
2.494£., ascribes the essay to Novatian.
5. Roussel (1916) 259, referring also in 260 n.6 to Butler (1915) 373,379, and Fig.
324, concerning Si'; ibid. 429£. and Illustration 371, Sur, and 442-445, Sahr, the
theater dimensions being 20m. in diameter with seven rows of seats. Further, on the
Delos theater, see Will (1951) 62, 68f., and 78.
6. McCown (1938) 159-162, 165, and 167; below, p. 21.
 Notes to Pages 19-21 151

7. Robert (l953b) 409, the theater not yet fully excavated.

8. The theater belongs to the first half of the first cent., d. Cumont (1926) 185
and 202; Frezouls (1952) 83-85, for certain features of design referring to a (noncult)
theater in Palmyra, pp. 87-89. Hanson (1959) 66 refers to still another cult theater in
Seleucia on the Tigris, unpublished then and still, to my knowledge. He notes,
ibid., the names of occupants inscribed on the seats of the Dura theater. Compare
the practice also at various Moselle sites, in Ternes (1973) 8, and Sch1eiermacher
(1966) 209, instancing sites at Altbachtal near Trier and Pachten.
9. Goh. ad gent. 2.12.2 (GGS 1.11 line 20).
10. Lehmann-Hartleben (1939) I35and 138; Fraser(l960) 117-119;andSokolow-
ski (1962) 135f. C. Picard (1955) 236 compares similar buildings at a numberofsites,
and Bookidis (1973) 206 describes a sort of theater for "at least 85 initiates" of
Demeter and Kore at Corinth, in active use during the Empire.
11. So I estimate from Heyder and Mallwitz (1978), the stone seats built in the
early Empire (pp. 68 and 71 ), the activities observed by Pausanias (9.25.5-8), and the
dimensions to be surmised from p. 27 Abb. 18 and p. 32, plainly for hundreds but
probably for up to 2,000 people.
12. The Pergamene theater in Boehringer (1959) Beilage 1 and p. 155 with Abb.
25 (second cent.), and Behr (1968) 27£.; on Epidaurus, Herzog(1931}63, citingPaus.
2.27.5, and Kotting (1950) 18 and 31.
13. GIL 13.1642, at Forum Segusiavorum near Lyon, a priest theatrum quod
Lupus Anthi filius ligneum posuerat de sua pecunia lapideum constituit.
14. That was the standard practice in Rome well down into Cicero's time. But
note SIG5 589line 46, where, at a lectisternium in Magnesia, there is to be "a musi-
cal performance, pipe, whistle and harp" in the agora. St. Andrews in Scotland and
Siena do the same every year, though no longer does the Church preside. For sacri-
fices being offered in the agora by private citizens whenever they felt like it, see Ael.
Arist., Or. 28.39 (I p. 498 Dindorf) and Ps.-Clem., Homil. 7.3 (PG 2.217); a sym-
posion offered for a local god on the town square at Philippi, in Collart (1937) 475.
15. On Ephesus, see Xen., Ephes. 1.2, cited in C. Picard (1922) 331£. and 338; on
performers, d. Robert (1940a) 284, using evidence partly for the east (Galen), partly
for the west, too: the senatusconsultum of 176/177 (FIRA 2 1.299, which speaks ofthe
vetus mos et sacer ritus in Gaul, to stage gladiatorial shows, and how sacerdotes of
the imperial cult hired the troupes). Laumonier (1958) 383 and 400£. explains the
9eatpucoi of Stratonicea, published by Hatzfeld (1920) 89 and 92, and the key text
in Diehl and Cousin (1887) 158.
16. Laumonier (1958) 347.
17. Job. Chrysos., Homil. in Matth. 7.6 (PG 57.79) and Mala!. 284£., and discus-
sion in RE s.v. Maiumas (Preisendanz 1928) cols. 610f.
18. Paus. 7.18.11-13, d. Herbillon (1929) 64-67, on coins of Amasia.
19. SJG5 1115 lines 30f., Si1ens at the theater at Pergamon; on kinaidoi, c£. Ber·
nand (1969) 2.28, second cent. evidence in Egypt, cf. Firm. Matern., Math. 7.25.4 and
esp. 14 (dancing to Cybele with drum accompaniment), and Robert (1964) 185; on
ballatores, c£. GIL 6.2265 and Isid., Etymol. 3.22.11; on cb:popatat, C. Picard (1922)
256, at Artemisia in Ephesus and Magnesia, in one corps numbering 20; ibid. 300,
on Couretes.
20. "Danced Hymns" in Wild (1963) 51-53 and 107 n.68, with much additional
152 Notes to Pages 21-21

material, from the reign of Tiberius to Hadrian, on dances (professional) offered to

other Egyptian deities at various shrines, accompanied by tambourine and sistrum,
ibid. 6lf., 72, and 77-82.
21. The quotations come from Tert., A pol. 15.4, Min. Fel. 37.2, and Apul., Met.
10.30f. (the author African, of course not the scene); Clement's views are in his Co h.
ad gent. 2.42£. (PG 8.76); 2.60 (PG 8.77); 4.58.3-4, hymns and masks, both, on the
stage; for histrio=dancer, cf. Charpin (1930-1931) 579, whoalsoaddssomescandal-
ized passages from Cyprian et al., pp. 580f.; and Fronto, Ep. ed. Naber 156 (Loebed.
2.105), histriones, quom palliolatim saltant . .. (again, the author not the setting
African). And the same equivalence appears, too, in Arnob. 4.35 (PL 5.1071£.), ...
quid pantomimi vestri, quid histriones, quid illa minorum atque exoleti generis
multitudo1 ... Saltatur Venus . .., saltatur et Magna sacris compta cum infulis
Mater, etc., and Aug., De doctr. Christ. 2.25.38 (PL 34.54).
22. Dig. 38.1.26 praef., the Pantomimus vel archimimus is opxnatt\~ ii J.liJ.lo~.
cited by Mommsen (1868) 462 n.2, who goes on to equate the archimimus promis·
thota of an inscription with a locator scaenicorum. For a Severan star actor and
priest of Apollo, parasitus et sacerdos Apollinis and others like him, cf. Muller
(1904) 342-346 and Gage (1955) 401-407. Ammianus (14.6.20) later describes mimae
as dancers in Rome who "are borne about in flying gyrations while expressing
numberless likenesses invented by theatrical fables."
23. Gaur (1963) 322-325, on South Indian temple dancing, the only kind I have
seen; but compare well-known Balinese arts, in McPhee(l966) 6, the dubs or guilds
of performers, and p. 16, "dancing and acting are so closely interwoven that it is
impossible to define where the one ends and the other begins."
24. Ps.-Apul., De deo Socr. 14.149; Lucian, De salt. 79; d. the epigraphic mention
of Dionysiac priests, xop&uaavt&~ ~OUK6A.Ot, at Pergamon in A.D.l06, d. Conzeand
Schuchhardt (1899) 179£., the eds. drawing attention to Lucian. For upper-class
enthusiasm for dramatic religious shows (but perhaps only plays, not dances), cf.
Tat., Ad Graec. 22 and 35 (the audience will own slaves, indicating wealth). Pollux
4.103, perhaps speaking not of the Roman period, distinguishes a number of
regional dances, e.g., "Ionic to Artemis, mostly in Sicily."
25. Hyporchemata, in Lucian, De salt. 16; cf. dancing at sacrifices in Syrian
fashion with cymbals and drums, by priest and women, introduced to Rome, in
Herodian 5.5.9.
26. Attested at Ancyra, cf. Franchi de' Cavalieri (1901) 70, the Mart. S. Theodoti
14 (women dancing in the procession to Artemis and Athena), cited by Hepding
(1903) 133; at Stratonicea, an opxnatt\~ hired along with dance choruses, in Lau-
monier (1958) 303, and above, n.24.
27. Above, chap. 1.1 n.92. The inquisitive Naassenes are present: 1tap&op&uouaw
olitot to~ l..&yoJ.lEVot~ Mf'jtpo~ J.l&'Ydi..Tl~ J.lUOtTlpiot~. Compare the action of the
authorities at Delphi which initiates (Katopyui~ouaa) all Greeks of the peninsula,
Plut., Moral. 418A. The ceremony is in the open and openly described.
28. SJGS 736 (Andania in Messenia) on the administration of mysteries, in·
eluding the hiring of dance choruses (J(Optt&iat) and instrumentalists through
l..&ttoupyt\aavt&~; CIG 3200, supplying of the c5PXT10t~ for the mystae (of Demeter?
-CIG 3 p. 722) at Smyrna; Joh. Chrysos., In Col. homil. 12.5 (PG 62.387), telling us
that in Greek mysteria there are dancing and instrumental music, the dancers to be
 Notes to Pages 23-26 153

dismissed as "prostitutes," i.e., professionals; Clem. Alex., Coh. ad gent. 2.60 (PG
8.77), women act out the mysteries in various cities, surely professional scaenicae
since decent women could not perform thus-d. Aug., Civ. dei 2.4, Cybele parades
through Carthage accompanied a nequissimis scaenicis. Notice, too, that Herodian
(3.8.10) witnessed in Rome in 204 the shows staged in the theaters "both for offerings
and all-night affairs, celebrated just like 'Mysteries.'" He betrays no sense of holy
secrets having been parodied or violated.
29. Cybele cult dancing described by Graillot (1912) 302-306, noting (p. 258)
professional performers in third-cent. Italian inscriptions. Add AE 1940, 131, an
Ostian tympanistria m. d.; and scenes in frescoes, Spinazzola (1953) 1.223 Fig. 250,
and pp. 224-242 and Plate XIV; Vermaseren (1966) 42f.; itinerant priests of dea
Syria=Atargatis in Apul., Met. 8.27, drawing on Ps.·Lucian, Lucius 37f.; and Isiac
dancing described by Tran Tam Tinh (1964) 27f. and 102 with Plate XXIV and by
Paribeni (1919) 107-111 with plate facing p. 106.
30. All instruments used, including the hom and bass hom, cf. Plut., Moral.
364F, GIL 6.219, Sokolowski (1962) 204, and Qu::sten {1930) 15 and 22. But not the
organ, apparently.
31. There is secular music of the masses at the service of Caesar's soldiers, Suet.,
Divus Julius 49.4 and 80.2, and at the service of the Church, Philostorgius, Eccl.
hist. 2.2, cf. SHA Firmus etc. 7.4-two references too early or late for our period.
Also Ovid,. Fasti 3.523-542.
32. As an arbitrary sampling from very abundant numismatic material, see, on
Ma, Bosch (1935) 122, third cent. from Heraclea, and Price and Trell (1977) 98 no.
176, Comana (the whole book a model for clarity and interest); ibid. no. 265, Men in
Antioch, cf. also Lane (1978) 541, third cent. Prostanna; RE s.v. Limyra col. 711,
under Gordian, cf. Plin., N.H. 31.22, and that other oracle also on coins, the
Cumaean Sibyl, in Engelmann (1976) 195£., the issues dated to Antoninus Pius's
reign; Ohlemutz (1940) 126 and 150; Graillot (1912) 357-360 (notes); Syrian deities,
too, e.g., the temple of Jupiter Heliopolitanus on coins of the emperor Philip, cf.
Rey-Coquais (1967) 43. For panegyriarchs, see those of the third century at Apamea
and elsewhere, in Ramsay (1895-1897) 2.442, cf. Sylloge numis. Graec. von Au lock,
Nachtrii.ge 4 (1968) no. 8348, where the panegyriarch is eponymous, indicating the
importance of the office.
33. Euseb., H.E. 5.1.47; Strabo 5.3.10; in Britain, at Viroconium, cf. Wilson
{1973) 27, shops around the second-cent. temple; and at two other sites, one of them
a temple flanked also by a theater, "sufficient to prove the periodic attendance of
large crowds," ibid. p. 31; Wheeler and Wheeler (1932) 50£., shops in the shrine of
Nodens; in other western provinces, cf. MacMullen (1970) 337, Spain, Switzerland,
etc., and Italy also; further, in central Italy, Evans (1939)41, a mercatusat the time of
sacrifices to Liber, and Stambaugh (1978) 585, citing inscriptions from Rome,
vestiarius de Dianio, etc., and also SHA Aurelian 48.4, wine sold in porticibus
templi Solis. Cf. the beer shop in the Sarapis temple at Arsinoe, PLon. 1177 (A.D.
113), and the "cafes and gardens" attached to the Adonis sanctuary in Laodicea and
rented to entrepreneurs, explained by Haussoullier and Ingholt (1924) 336, the date
falling in the third or fourth century.
34. Ibid. 333; IGR 3.1020, at Baetocaece, "those who come to worship" buy
"slaves, livestock and other living things" every fortnight; at Damascus, similar
JH Notes to Pages 26-28

arrangements, MacMullen (1970) 337 n.25 and Dussaud (1922) 224-233, the date of
construction being ca. A.D. 200? and expansion being commissioned in 339/340
with use of funds "of the Lord God" (Hadad). That makes clear the connection of
cult and commerce.
35. On Tithorea, cf. Kahrstedt (1954) 11 f., who, however, omits a few vivid details
from Pausanias (10.32.15): that "the peddlers make tents of reeds and other materials
on the spot, and on the last of the three days hold a fair, selling slaves and livestock
of all sorts, clothes, silver and gold." Further examples of the practice and facilities
in the third cent. in Just., Epit. hist. 13.5.3, and similar texts from Greece cited in
MacMullen (1970) 336.
36. Note complaints against governors, procurators, and soldiers in the well-
known inscription from Scaptopara (Thrace), IGBulg. 2236=CIL 3.12336, recall-
ing the precautions against forced billeting (napoxTJ) in IGR 3.1020 line 35.
37. On special coin issues for a panegyris, see Mattingly (1928) 202; Franke (1968)
28f., on the navT)yuplKO~ d.yopav6~o~. see Robert (1966) 19 and 24f., cf. idem (1968)
573, "panegyriarch, that is, superintendent of the commercial sector of the festi-
val .... " Note the indication in IGR 4.144 (Cyzicus), where a priestess is thanked
for attending to the "merchants and strangers come to the festival and tax-free mar-
ket from Asia."
38. Dio Chrysos., Or. 27.5f; cf. 77.4, a whoremonger circulating with his troupe
in central Greece, from one festival to another. In the period of this study, temple
prostitution survives only at Byblos (Lucian, Dea Syria 6) and at the festival of Zeus
Larasios near Tralles every four years, d. Robert (1937) 406f.
39. Ramsay (1941) 76; d. IGR 3.1020, the fair atBaetocaeceatleast three, perhaps
five, centuries under the same regulations.
40. Polemo on Perge in Foerster, Scriptores physiognomici graeci et Latini 1
(1893) 282, a Hadrianic source brought to notice by Robert (1948a) 66. The Pergaean
festival is noted by Strabo 14.667 also. As to its wide appeal, compare Roussel and de
Visscher (1942-1943) 179, an inscription recording the assertion in A.D. 216 that
"this Zeus-shrine [of Dmeir in Syria] is famous ... and certainly visited by ab-
solutely everyone from around, who come here and send processions." On the
Ephesian festival, the quotation is from SIG' 867, with a similar description of
the attendance in Xen., Ephes. 1.2.3.
41. Witt (1971) 98 and 162; for bathing of images, see the Mart. S. Theodoti 14
(cit. above, n.26); Cybele washed in the Almo river, Ovid, Fasti 4.337£., andRE
s.v. Pompa (F. BOmer, 1952) col. 1950; for Artemis's image taken to the sea off
Ephesus, see C. Picard (1922) 314; for Adonis's trip to the sea off Alexandria and
Byblos, see Will (1975) 100; and other lavationes in Stocks (1937) 6, the deaSyriaand
her pond, and Graillot (1912) 137.
42. Daremberg-Saglio s.v. Circus 1193; RE s.v. Pompa col. 1908; above, nn.5 and
15.
43. Hill (1897) 87f., concerning types of cart; Paus. 7.18.12, deer-drawn cart for
Artemis at Patrae; elephants, Daremberg-Saglio Joe. cit. and Fig. 1528.
44. Diod. Sic. 1.14.3; Plut., Moral. 365B; Xen., Ephes. 1.2; Robert (1975) 324, on a
variety of "-bearers" (-<popoi), in Latin, c(a)ernophori, cannophori, cistijeri, den-
drophori, hastijeri, pastophori, rhabdouchisa (IGR 1.614), and Anubofori (GIL
12.1919). For a hilarious picture of articles carried in a religious procession, cf.
Athen. 201C.
 Notes to Pages 28-29 155

45. Ovid, Fasti4.179-186; Mart. S. Theodot. 14; Paus. 2.7.5; IG 7.1773 (a compe-
tition at Thespiae in the second cent. for the best marching hymn written); Epi·
phanius, Panarion haer. 51.22.10f.; and Xen., Ephes. 1.2.3.-1.3.1.
46. Sokolowski (1955) 28 and 30-but both inscriptions pre-date our period.
47. RE s. v. Pompa lists over 350 entries, of which about 100 are once-only affairs.
Evidence for the great majority of the remainder pre-dates Roman times, but most
no doubt went on into the Empire. Patrai (items nos. 46-48 and 131) and Cos (27, 99,
126, 141, 188, and 229f.) are cities of several parades. Plut., Moral. 39A, indicates the
keen interest in parades, Macrob. 1.23.13 indicates the high status of participants.
The same passage shows pompae in Syria and Italy, c£. in Rome, lecticarii for
Jupiter-Dolichenus cult, in Merlat (1951) 186; also Isis parades, often mentioned.
For parades in Syria and Phoenicia, see Hill (1911) 61, and Epiphanius loc. cit. (at
Petra); for Cybele parades, cf. Aug., Civ. dei 2.4, and in Autun, in a text brought to
notice by Hepding (1903) 72, the Vita S. Symphoriani 3, which is best consulted in
Zwicker (1934-1936) 2.163, with a dramatic date in the late second cent.; Sulp.
Severns, Vita S. Martini 12.2, pagan deities paraded at Tours still in Martin's day;
and Isis parades at Vienne implied in CIL 12.1919, through the Anubiforus.
48. Dea Syria 10. Pilgrims were male only, and shaved their heads and eyebrows
and prepared for their trip by continence, ibid. 55 and Macrob. 1.23.13.
49. Ohlemutz (1940) 177, the hostel at the site, and 180, the coins from Cyzicus,
Sardis, etc., up into the fourth century.
50. In Boeotia, at an oracle, Pausanias describes the hostels (9.39.5) for the
suppliant sick at Tralleis; for the sick finding lodging in a village near the shrine,
close to Nysa, see Strabo 14.1.44; at Panamara, cf. Laumonier (1958) 227, inns
serving the Zeus shrine; at Olympia, cf. Kunze et al. (1958) 33-41 and 55-67, two inns
adjoining the sanctuary, one of second-cent. date (p. 59), the other ca. 200 (p. 67); at
Jerusalem, an inscription describing full facilities for needy pilgrims, in Safrai et al.
(1974) 192, better dated (Augustan?) by Clermont-Ganneau (1920) 193; and lodging
for the pregnant and dying at Epidaurus in a special building, Paus. 2.27.7. In the
West, see the evidence in Ternes (1973) 8, hospitalia at many sites; idem (1965) 415,
emending CJL 13.4208 of A.D. 232; Agache (1973) 52, sites in France; Weisgerber
(1975) 20 and 83 (dates of active use), a hostel at holy springs, twenty miles northeast
of Trier; Binsfeld (1969) 247-252 and 266, a healing shrine by springs, active ca. A.D.
150-275, the hostels having rooms with fireplaces; Formige (1944) 48, very big
buildings at Sammy; and Wheeler and Wheeler (1932) 44-50, a very large building at
the healing shrine, active in and after A.D. 250.
51. More than two hundred known Asclepieia, cf. Kotting (1950) 13.
52. On Vicarello, cf. Friedlaender (1912-1923) 1.329f.; on Samothrace, Fraser
(1960) 15f. and 47f. (visitors from Beroea) and 58£. (from the Strymon, Thessalonica,
Thasos), in second and third centuries; travelers from good distances to other
shrines attested in inscriptions, in Capello (1941) 99-137, IG4.953 and 956 (merely
illustrative for Epidaurus), and Laumonier (1958) llO, for the Zeus shrine near
Mylasa. On women pilgrims, see Paus. 10.4.3 and Plut., Moral. 249E, referring to
old times, but the shrine at Delphi was still active, witness Plutarch's friend Clea the
priestess. Note also an Isiac on pilgrimage from Rome to Alexandria, in Tibullus
1.3.31.
53. The poem in Bernand (1969) 2.128, discussed pp. 135£.; other texts, pp. 7£.,
most but not all of pilgrims, discussed pp. 23£.; and second-cent. visitors from
156 Notes to Pages 29-JO

Greece and Jordan, also from Alexandria, ibid. 167, 176, and 368; for pilgrims to
other shrines in Egypt, see Nock (1934) 58-67; Bataille (1952) 165; Yoyotte (1960)
55£.; and Hohlwein (1940) 262, 264£., and passim.
54. Lucian, Dea Syria 55, cf. Dio 36.11 implying guides at Baalbek, "they recount
stories and point out ... "; Ael. Arist., Or. 25.2, 6 1t&puiyc.ov, d. Ps.-Lucian, Amores
12, the ~aKopo~, the text being of the early fourth cent., says Helm in RE s.v.
Lukianos col. 1730; Paus. 1.13.8; 1.31.5; 1.35.8 (in Lydian temples); 1.41.2 and 4;
2.9.7; 2.23.6; 7.6.5; and 10.28.7; Athenag., Leg. 28; priests do the explaining, in
Lucian, Dea Syria 1 and 4; Plut., Moral. 354A and 410B.
55. Ohlemutz (1940) 143 describes Aristides' lecture on the local spring; Strabo
14.1.44, on the inquiries at Nysa; d. Art., Oneir. 1 praef., Artemidorus's search for
knowledge on his subject "in Greece at cities and festivals"; on learned visitors, see
Lucian, Heracles 4 and Dea Syria 15, and Paus. 7.23.7; for Cynics loitering at the
temple gates, cf. Dio Chrysos., Or. 32.9, and the Cynic who appears from nowhere to
comment on a trial for irreligion, Acta S. Apollonii 33, at Rome.
56. Plut., Moral. 385£., cf. 417F, reference to the theologues of the shrine, and
Paus. 7.23. ?f., already referred to.
57. Min. Fel. 2.3, d. Favorinus in Aul. Gel. 18.1.2.
58. Justin, Dial. l.l, where the setting is a gymnasium either in Ephesus or
Corinth, cf. VanWinden (1971) 21; Lucian, ]up. trag. 4 and 16 (in Athens-the
passage quoted).
59. Aside from the familiar but fictional Octavius of Minucius Felix, one must
note Ps.-Clem., Homil. 3.29 (PG 2.129), the dramatic date underTiberius, pitting
Peter against Simon Magus before a throng, for three days, 3.58 (PG2.148); Euseb.,
H.E. 4.16.1, Justin against Crescens the Cynic before a throng, d. II Apol. 3;
Theoph., Ad Autol. 2.1; and Scherer (1960) 13.
60. Dio Chrysos., Or. 12.44£.; above, chap. l.l n.49 ..
61. Paus. 1.21.4 and 4.31.11£.; Plin., N.H. 35.17; and Lucian, Heracles 1-3,
puzzling over a painting of Hercules depicted in Celtic fashion, and interpreted for
him by a native. Idem, Toxaris 6, Scythian religious beliefs studied in local
paintings.
62. Paus. 5.10.7, at Olympia-the same assumption in Lucian, Toxaris 6, and
animating a modern viewer, e.g., Laumonier (1958) 349, on the painted reliefs of the
Hecate temple at Lagina. Inscriptions: Tr.Rul£. 1592=/GR 1.743; IGBulg. 679
=IGR 1.567=Vermaseren (1956-1960) 2.355 no. 2265; CIL 2.4085, the restorer of
Minerva's temple calls himself perfector et pictor; CIL 8.25520=Merlin (1908) 21,
Bulla Regia under Diocletian; CIL 8.8457=20343, Sitifis in A.D. 288; Flusser (1975)
13, Samaria (third cent.?); and Reinach (1908) 499, Aphrodisias second cent.
63. .f<'or trescat:s with scenes (leaving aside all walls merely colored, plus all the
rich Mithraic material and Dura), as a sampling from abundant remains, see, in
Rome, Graeve (1972) 319, temple for Syrian cults; Gauckler ( 1899) clx, in Carthage;
Tran Tam Tinh (1964) 102, in Herculaneum. As to painted reliefs, for a sampling,
see P. Girard (1878) 92, the Athenian Asclepieium; Cintas (1947) 58, 73, and 78,
stelae of the first cent. in the Hadrumetum tophet; Brommer (1973) 22, in the Vulcan
shrine at Alzey; and Kern (1892) cols. 113£., Eleusinian Demeter stele of the second
century. For a painted theater, at Stobi, see Wiseman (1971) 402. Fora sampling of
striking statues, see 0. Deubner (1977-1978) 234 and 248, a Hadrianic Sarapis
 Notes to Pages JO-Jl H7

statue; gilded ones in CIL 9.3146, Cybele in Corfinium; Plin., N.H. 33.61, Fortuna
at Praeneste, and 33.32, Hecate at Ephesus is dazzlingly bright (=gilt?); Goodhue
(1975) 36 and 55, a first·c~nt. Dionysus in Rome; Merlat (1951) 176, Dolichenus has
an inaurator in Rome arld, 183, statues are gilt (or painted?); Reutersward (1958-
1960) 2.186, in the Pompeian Isis shrine; and Audollent (1901) 379, the Caelestis
shrine at Carthage of the early Empire, with gilt-faced goddess. Oenomaus esti·
mated idols at 30,000 in round numbers, to which more were ever added: 120 Nikai
and Erotes in the Ephesus theater, cf. Pekary (1978) 727; 29 gold statues to be
paraded in the theater, all of Artemis, OGIS 480 lines 6-9, cf. TAM 3, 1.136, the
4yaAJ.la 7tOJ.lmK6v 9&~ of Termessus dedicated ca. A.D. 203. In Africa, statues of
deities were dedicated regularly in thanks for municipal office, e.g., CIL 8.858, 1887
( =ILA lg. 3066), 2372, 7098, 7983-84, 15576, AE 1968 no. 609, and Leschi (1957) 227.
But the practice is known elsewhere, e.g., ILS 9416=IGLS 2716, a Jupiter statue
given for one's decurionatus at Baalbek.
64. Paint in letters of inscriptions: Roebuck (1951) 39; Nock (1934) 58; Bernand
(1969) 2.192; Rostovtzeff (1937) 204; and Vermaseren (1956-1960) 1.69 no. 53, Dura;
Kabbadia (1885) 199 and Herzog (1931) 43, Corinth, with traces of gilding also-
these examples from the east; and from the west, Vermaseren (1956-1960) 1.96 no.
148, Lambaesis, and 312 no. 908=CIL 13.1771, Lyon, with gilt; Vermaseren (1956-
1960) 2.366 no. 2308; Kan (1943) 69, the Hadrianic dedication in the Dolichenus
shrine at Carnuntum. These items of evidence (not methodically sought out) for
practice of a fragile sort seem to me to indicate that the letters of all inscriptions were
routinely painted.
65. Examples of inscriptions reconstructed or refreshed in Hanfmann and Wald·
baum (1975) 68, in the Artemis shrine, Severan setting-up of old stelae on antique
bases; IG 10, 2.255, cf. ibid. p. 540 n.2, a first-cent. recutting of an earlier testi·
monium; Kern (1900) 140 no. 215, decree recut; the same process in Gawlikowski
(1973) 71 and Sokolowski (1962) 67, a lex sacra at Sparta in the second cent.;
compare a cult relief repainted, in Forrer (1915) 69, and an idol renovated, sigilum
renovavit, Albertini (1931) 198-200 no. 3, A.D. 209-211, in a Sol shrine.
66. On Oenomaus's estimate, quoted by Pekary, see above, n.63; the passage,
surviving in Euseb., Praep. ev. 5.36.2, goes back toHesiod, Works252f., who speaks
of "daemons" (also quoted by Max. Tyr., Phi los. 8.8). On pruning out, see Coarelli
(1974) 45f., at the Capitolium in Rome.
67. IG 12, 2.108; Buckler and Robinson (1932)97;and,onM.AntoniusPol~o's
inscription, for which the setting is given in Phrynicus a little later in the second
cent., by Phrynichus 395 p. 494 Rutherford, see Ohlemutz (1940) 134. On directions
in dreams, see further AE 1941 no. 106, near Doliche itself: "the god Jupiter
[Dolichenus] commanded" So-and-So who "set up a statue for the god for the
emperor's salvation"; CIL 6.30998, image of Sarapis at Rome ex viso; a city obeys
the Clarian oracle to "set up now a Phoebus at the gates, bearing his bow, to avert
the plague," in the emended text of Mordtmann (1881b) 261£.; IGBulg,2 370=IGR
1.767, from Anchialus near the mouth of the Danube in the second cent., the
erection of "icons suitable to the deities according to the oracle of our lord the
Colophonian Apollo," by municipal decree (in time of plague?); ILS 3526, at
Tibur, a Sarapis signum erected ex viso; Plin., N.H. 34.19.58, Augustus, being
"directed in his sleep," restored a pillaged Apollo statue to Ephesus; similarly,
158 Notes to Pages JJ-J2

altars set up by the imperium of the god, vel sim., in /LS 3534, and 3973, iussu
Proserpinae; Schwertheim (1978) 2.792, a Cybele statue set up Kat'btttayrjv (the
most common Greek formula); ibid. 796, Kat'CSvap (that formula too being very
common); an aedicula Kat'dvetpov, C/G3947; C/L3.990, ex iussu deiApollinisfon-
tem Aeterni . .. NN restituit; a relief showing Zeus with eagle, Kata KtA.eucnv toO
9eoii, MAMA 5.17; a similar relief Kat'btttayrjv, to Zeus Brontaeus, in Schmidt (1881)
135; reliefs showing two deities, dedicated Kat't?tiTtVotav ~to<; KtA.A.aJ.LEVllVOii, in
Schwertheim (1975) 357; a sanctuary of Venus renovated ex visu, CIL 10.5167; and
/LA lg. 2132, a temple built ex praecepto. On merely conventional erection of statues,
most easily documented in Africa, see above, n.63.
68. Paus. 10.32.13, referring to invitations by dream also in another area, the
Maeander valley; 10.32.17f., on punishment; and compare the cures in the Sara pis
temple, accompanied by threats against any impiety or blasphemy, in Aelian, De
nat. animal. 11.31 and 33.
69. Conveniently, a few in MAMA 4.279-285; earlier, in Ramsay (1889) 218,
giving a good summary; idem (1895) 1.147-149; Wright (1895) 72; Keil and Premer-
stein (1908) 16; Steinleitner (1913) passim; Buckler (1914-1916) 169-177; idem (1933)
7f.; Longo (1969) 1.158-166, citing inter alia SEG 4 (1930) 647£.; Robert (1948b) 108;
and inscriptions in Attica, Sokolowski (1969) 106f. and Lane (1971- 1978) 1.46 and
3.20-25.
70. Pettazzoni (1939) 198-200 stretches little evidence very far, to find sin/guilt in
all Syrian cults; better, Gawlikowski (1973) 57, the fragmentary lex sacra," ... qu'on
ne vole pas et qu'on ne peche pas ... il a jure parYarhibol ... etprendra son gage";
and Hermann and Polatkan (1969) 58, "Medon made a crater for Zeus Trosusand
the ministrants ate the [parts] not sacrificed and [Zeus Trosus] struck him dumb for
three months and abode with him at his dreams, that he might set up a stele and
write on it what had happened to him,'' etc. At Epidaurus, see Herzog(l931) 37 and
passim, pointing out that Pausanias (2.27.3) saw six stelae, and on the three
surviving (plus fragments) there are 66 cases, therefore ca. 130 originally. But in
Doric, says Pausanias-which the extant are not.
71. Herzog (1931) 8 (d. another scoffer, p. 10), 27 (d. another cheater, p. 31), and
28, with comparison drawn (pp. 43f.) to the stele with case history, /G42.126=S/G3
1170, a few years before Pausanias's visit to Epidaurus, or (p. 45) /G 42.127 of A.D.
224, goiters cured in the dormitory, and the inscription dedicated Kat'dvap. Note the
case in /G 42.956, second cent. (Epidaurus) where the concluding injunction from
the god is to publish (or perish, presumably), avaypci\jlat taiita.
72. For people seated in a sanctuary in the Isiac shrine in Pharos, Rome, or
elsewhere, see Tibullus 1.3.27£.; Martial 2.14.7£.; Ovid, Pont. 1.51£. and Amores
2.13.17; and Tran Tam Tinh (1964) 110; at the Lerna Asclepieion, Paus. 2.4.5; and
other texts for the city of Rome in Stambaugh (1978) 580 n.190.
73. Herzog (1931) 59, "volksti.imlich," d. in the Tiber island inscription, "sim-
plicity of style" remarked on by Longo (1969) 1.85.
74. Plin., Ep. 8.8.5 and 8.
75. Of course, even from the three great shrines, most is lost; but the fourth,
Aigeai, had its cure inscriptions (Liban., Ep. 695.2, addressed to the governor of the
area in A.D. 362) and its fame (Euseb., Vita Const. 3.56 and Soz. 2.4f.).
76. The Tiber island shrine is well known, if not truly Roman. See cure in-
 Notes to Pages JJ-J4 159

scriptions in, e.g., Weinreich (1909) 30, Longo (1969) 1.84f., and SJGS 1173=JGR
1.41; from Africa, a similar text in Beschaouch (1975) 112, in somnis monitus,
Saturni numine iussus ... pro comperta fide et pro seruata salute votum solvit
libens animo, A.D. 238; and from Egypt, compare, e.g., the stele with many inscrip-
tions on it, therefore a little resembling the big six at Epidaurus, here thanking
Sarapis, SEG 8.464. Cf. Aelian, De nat. animal. 11.31-35 and the next note.
77. POxy. 1381 (second cent.), praising Imhotep=Asclepius, esp. lines 90f. and
144f.; Anthol. Palatina 6.231, a poem of thanks if the goddess saves the poet from
suffering, "in spirit and intent ... closely similar to inscriptions," as F. Dunand
(1972-1973) 3.214 points out; for collections of cures wrought at the Sarapis temple
of Canopus, d. Strabo 17 .1.17; and note Tibullus 1.3.27£., "that Isis can heal, many
tabellae in your temples assert," closely echoed by Juv. 12.26-28.
78. Veyne (1967) 738 n.2, brings to notice Philostt., Heroikos p. 141, l9K,
tmocppa'Y\~O~EVO\ ta~ &\)x,a~. meaning pieces of papyrus; as to wooden records (as I
take them to be), see the 1tttta1C\a in Lane (1971-1978) 3.28, the nivalC&~ in Strabo
8.374 and Paus. 9.39.14, and the tabellae in the preceding note. Note also Roebuck
(1951) 39, a dedicatory "inscription," painted, the letters not cut at all.
79. Besset (1901) 326, fromApolloniaon the Rhyndacus, "theearsand[=on?] the
altar in the shrine" (oflsis?), probably not similar to the inscription (pace Besset 326
n.l) found at Epidaurus, Cutius has auris (=aures) Gallus tibi voverat olim,
Phoebigena, et posuit sanus ab auriculis, in Weinreich (1912) 64=JG 42.440, who
(for no good reason) calls it falsified. The ears indicate the suppliant's organs cured,
d. parallels in Bieber (1910) 5-8. Milik (1972) 104f. cites from Palmyra the inscrip-
tion thanking Zeus, "for he turned to him his eye and his ear," explaining the
symbol on the cippus. Charles-Picard (1954) 119 describes Baal Hammon=Satum
being often portrayed holding the lobe of his ear, listening, or the ears shown alone;
for the latter at Philippi, see Collart and Ducrey (1975) 179. In Egypt, "artisans
manufacture ears and eyes of precious material, offered to the gods of the temple,
symbolizing that the god sees and hears all," says Clem. Alex., Strom. 5.7.9 (PG
9.69). For addresses "to the ears" of a deity, d. CIL 3.986 (Apulum in A.D. 180), to
Aesculapius; CIL 5.759 (Aquileia), to the Bona Dea; to Isis, in Vidman (1969) 40f.,
the "Karpokrates" text given above at section 1, n.67; F. Dunand (1972-1973) 3.208 n.2,
ears in a relief from Aries and elsewhere; and "ears"=listening posts in Asclepieia,
in Wolters (1914) 149 (Epidaurus) and 151 (Pergamon).
80. 'Em\Koo~, a term attached to dozens of deities, see Weinreich (1912) 5-25; for
geographical distribution, see occurrences common in Thrace, IGBulg. 682, Col-
Ian and Ducrey (1975) 245, and Poenaru-Bordea (1964) 103; in Asia Minor, Rehm
(1958) 126 no. 137, Keil and Premerstein (1908) 25, Lanckoronski (1890- 1892)2.220,
and Lane (1971-1978) 3.78f.; in Greece, Weinreich (1912) 26 and Baur (1902) 61; in
Rome, IGR 1.33 and 35-36 and AE 1913; 188; of the statue of any deity anywhere,
Athenag., Leg. 26; and in Latin exaudiens, exauditor, in CIL 10.4553 and ILS 3744
and 3980.
81. Feet made of stone, e.g., CIL 6.572, inscribed deo Sarapi ... ex visu, or
uninscribed, as at a cluster of shrines in the Dijon area, with other organs, Toutain
(1907-1920) 1, 3.383, or at the Ephesian Artemision, d. K.Otting (1950) 49; but feet
carved in relief merely as a sign that the inscriber had been present are common, too,
and sometimes mean that the deity addressed had been present. For much discus-
160 Notes to Pages J4-J5

sian of the three meanings, see Mordtmann (188la) 122, Amelung (1905) 159f.,
Lanckoronski (1890-1892) 2.220 and 232, Hatzfeld (1927) 106, Deanna (1924) 32,
Guarducci (1942-1943) 308-315, giving a full collection from all provinces, Yoyotte
(1960) 59, and Malaise (1972) 106 n.3.
82. In addition to the works cited in the notes just preceding, on parts of the body,
see the inscriptions mentioning them in Robert (1955) 163; lnscriptiones Creticae,
ed. M. Guarducci I p. 171 no. 24, "I have set up to you, Savior [Asclepius], two
[statues of] Dreams for my two eyes, enjoying [again] my sight" (third cent.); and
Greg. of Tours, Liber vitae patrum 6.2, brought to notice by Kotting ( 1950) 13 n. 46:
at a shrine in Cologne, simulacra ... membra, secundum quod unumquemque
dolor attigesset, sculpebat in ligno.
83. Girard and Martha (1878) 420-440 passim, e.g., 436; Girard (1881) 16, the
inscription from Athens of the earlier third cent. B.c.; IG 7.303, Oropus ca. 240 B.c.;
and Homolle ( 1882) 5, the Delian Apollo's inventory of ca. 180 B.c., with extraordi-
nary variety of objects, e.g. (pp. 130f.), weapons, anchors, sports equipment. For a
Latin example, see CIL 8.12501, inventory of a Caelestis temple.

3. Displays and Accommodations at Temples

I. Paus. 2.32.1, Troezen; Lucian, Dea Syria 60, at Hierapolis; at Panamara,
Deschamps and Cousin (1887) 390, and Roussel (1927) 126, the locks of hair offered
in stone receptacles; and at Paras in the third cent. post-Caracalla, dedication of
"the first-cut hair," "the hair of childhood," to Asdepius, IG 12, 5.173- 176. The
most striking is Hygieia's temple near Sicyon, almost hidden under the offerings of
women's hair, says Pausanias (2.11.6).
2. IRT 231 and 295 (Oea and Lepcis); Paus. 5.12.3 and CIL 10.3796 (Capua,
third cent.)=Buecheler (1895-1897) l.l23 no. 256, "I believe that never as a gift
before ... was revealed a head so vast, born amid the shadows of the hills and for-
ests .... "
3. R. M. Peterson (1919) 152, at Capua, to the Arabic god Dusares in A.D. ll,
"Because he heard them" (the dedicants); Forrer (1915) 41, stone painted lions; and
Lucian, Dea Syria 41.
4. Paus. 2.11.8 and 2.28.1.
5. Lucian, Dea Syria 47; Aelian, De nat. animal. 12.30, instancing several loca-
tions; and Plut., Moral. 170D. On Labraunda, d. Laumonier (1958) 47 and 96-98,
instancing sacred fish at many Greek shrines, e.g., Attica, Paus. 1.38.1 and Smyrna,
SIG 8 997 =Sokolowski (1955) 49, the lex sacra of the first cent. B.c. cursing those who
hurt or steal the goddess's fish (of Atargatis). Du Mesnil du Buisson (1943-1944) 325
and Lambrechts and Noyen (1954) 262f. and 274 discuss Astarte's sacred fish, d.
Athen. 8.346, worshipers "bring her offerings of fish made of silver or gold, but the
priests bring to the goddess every day real fish, which they have fancily prepared and
served on the table"; also Cornutus, Theol. graec. compendium 6, and Clem. Alex.,
Coh. ad gent. 2.238 (PG 8.121), fish worship in Syria. Note the survival of the
custom till recent times, Glueck (1937) 374 n.4.
6. Lucian, Dea SyJia 14 and 54, d. Clem. Alex., loc. cit.; Philo, De providentia,
in Euseb., Praep. ev. 8.14 (PG 21.673 ), d. Phoenician worship in Africa producing a
Tanit= Caelestis to whom doves were sacred, Merlin (1909) ccxxxvii and LeGlay
(1966) 352; at Attouda in A.D. 77, note the proconsul's decree, in Calder (1935) 217,
 Notes to Pages J5-J7 161

and Robert (1971b) 92-95; further on sacred doves, Buckler (1914- 1916) 169-171,
second half of the second cent.; MAMA 4.279, Lairbenos, second/third cent., and
Cameron ( 1939) 155-158; the Achilles shrine in Arrian, Ep. ad Traianum 32 (p. 398,
C. Muller, Geogr. graeci minores I); and Plin., N.H. 10.51, geese fed in Rome.
But Aristides notices the geese wandering untouched (sacred) in the Isis shrine at
Smyrna, Or. 49.49.
7. Plin., N.H. 16.242 and 16.162, d. Reinach (1906) 107, Aphrodite's palm grove
cared for by a second-cent. liturgist at Aphrodisias.
8. Plin., N.H. 16.213, and other texts in C. Picard (1922) 62.
9. Graillot (1912) 122, d. Monsieur Themanakis (1893) 208, "the festival of
wood-cutting" for the deity at Cos; Vermaseren (1977)63, theinscriptionofA.D.l47
referring to the collegium dendrophorum with its sanctum, including a "plot of
land ... ad pinus ponendas;" a first-cent. lex sacra prohibiting cutting in Hera's
woods on Samos; and IG R 4. 48 and Robert and Robert (1958) 298, a copse sanctified
to Silvanus by Romans in Mytilene.
10. CIL 6.2033, anno 21 and passim, the Dea Dia's lucus; compare the Greek
reverence for groves like those at Dodona and elsewhere, seen for example in Zeus-
of-the-Shadows, in Cook (1903) 414, on Paus. 3.10.6. In Egypt, famous holy groves,
Orig., C. Cels. 3.17; at Palmyra, Caquot(l954-1957) 78andMilik(l972)7, on groves
with gardens attached.' Generally, see the good collection of evidence in DE s.v.
Lucus (A. Pasqualini, 1975).
II. Ps.-Lucian, Amores 12. A matching glimpse from another region and an-
other kind of source: CIL 10.6073 =ILS 6284, at Formiae, where a benefactor stages
some ceremony for Jupiter, "at which dedication he distributed 20 sesterces to each
of the city senators dining publicly in the grove."
12. l'ish meal, above, n.5; Fratres Arvales in their cenatoria and triclinia, e.g., in
Acta aro. A.D. 218 A 1; d. the delicious meal of the Salii, in Tert., Apol. 39.15 and
Suet., Claud. 33,or thechoicewinesofthepontifices, inHor., Od.2.l4.28. For other
priestly meals, d. Robert (1940b) 18 n.S, Apollonian in Cyrene; IG 5, 2.269, As-
clepian in Mantinea; Milik (1972) 150f., for priests of local baalim in Palmyra.
13. Examples of dedicatory banquets in Stambaugh (1978) 567; see above, n.ll.
14. CIL 6.2273, cocinatorium at Rome for Venus; AE 1975, no. 197, a culina at
Casinum given by four freedwomen (explicit evidence of women in cult dining
rooms being rare in any province; for the Venus shrine at Casinum, d. also CIL
l0.5166f.). For Cupids on earrings, cf. E. Levy (1968) 530, comparing Attisea,rrings
in Vermaseren (1966) 49, and Philostr., Vita Apollon. 5.20, Demeter or Dionysus
figurines worn as charms. For other culinae in Italy, cf. CIL 5.781 and 815 (mensa)
and 6333 (mensae 1111); 6.2219; 10.3781; and 14.3543.
15. CIL 3.4441, porticum cumaccubito;Tudor(l969-l916)2.11 (painted plaques)
and 257 (many showing a mensa tricliniaria).
16. 1 Cor. 8.10; cf. a similar situation, of Christians dining with pagans at the
tpa1tt~a of 3aiJ.lOV&~, envisaged in the Pseudo-Clementine Homil. 9.9. (PG 2.248)
and 9.15 (PG 2.252), dated by Quasten (1950-1960) 1.62 to the early third cent. For
Paul's scene, c£. archaeological evidence from Corinth: the dozens of dining rooms
found on a terrace of the acropolis in the temenos of Demeter and Kore, in Bookidis
(1973) 206; Roebuck (1951) 51, three Hellenistic dining rooms with red painted
walls, each with eleven couches, painted red and yellow, repainted, and regularly
 162 Notes to Pages 17-JB

used "throughout the Roman period" (p. 55); and equivalent facilities at the
Troezen Asclepieion (couches in wood) and elsewhere in Greece (pp. 52 and 54);
other examples in inscriptions, cf. Launey (1937) 402, at Thasos, each o{JCoc; con·
taining seven JCA.ivat.
17. Launey loc. cit. citing rooms at Assos and Aphrodisias; further, on the 3t:t7tVt-
at1lptov for Aphrodite at Aphrodisias, Reinach (1906) 242; Robert (1945) 1.49f.,
GJCt1vat; Hicks (1891) 232f., a Jlayt:tpt:iov at a Hermes temple in Cilicia; dv3provt:c;
in the Zeus temple at Labraunda, in Laumonier (1958) 50-55, one ofthem 22 x 12m.,
another with built-in couches; at the Ephesian Artemision, a vast tanat1lptov,
Philostr., Vit. soph. 2.23.2 p. 605, a building meant "to be bigger than all others
elsewhere combined" (hence, the architectural form common). Similarly in Plut.,
Moral. 146D, though unhistorical: a party, far larger than the usual, sacrifices at the
Aphrodite temple and dines in the adjoining hestiaterion.
18. Starcky (1949) 62, the rooms filled with sherds of wine cups, and Milik (1972)
141 (Khasr Semrine) and 142 (IGLS 584, A.D. 129). Note at the temple of the Baal at
Damascus a chief chef, archimageiros, in Perdrizet ( 1900) 441, and a temple south of
Tripolis equipped in A.D. 184 with "a shelter and in it aanpac;," an outdoor dining
arbor with portico and cistern. See Rey-Coquais ( 1972) 87 and 93; and for the Greek
term, SJGS 1109, lines 47 and 70, serving a Thracian cult group.
19. For Dura, notably the temple of Aphlad, see Cumont (1926) 396, thinking
participants in group dinners are pilgrims (corrected by Milik [1972] 122). In
Starcky (1949) 55-57 and 64£., the dpxt9taattt:orov of the Syrians at Delos justifies
the term thiasos for the Syrian cell (also called hatpt:ia) which meets in its dv-
opc.Ov or o{Koc;. The amphorae sherds found there explain the use of the rooms; cf.
also Milik (1972) 134-140. For Palmyra, seeCumont(1929)256f., using mentions of
a symposiarch in LeBas-Wadd. 2606a and IGR 3.1045 (the two, however, are the one
same inscription; and Cumont has lost a part of the former text) and IGR 3.1533, the
symposiarch of the priests of Bel, A.D. 203; Starcky (1949) 52-54 and 58-62; Milik
(1972) 121 and 157 on the coalescing of workers (likewise at Hatra); 144-148, GUJ.t-
noatov=meeting room for the cells consecrated to the deity; 151-154, the serving
staff for the rooms; 184£., tesserae of invitation; and 188-192 and 287, conduct of
meetings; also, briefly and broadly, Teixidor (1977) 133-135.
20. CIL 3.7954=1LS 4341, at Sarmizegethusa; at Rome, Milik (1972) 216, citing
CIL 6.52, a sacred triclia guarded ex imperio Solis, i.e., by Malakbel, cf. also 6.712
and 31 036; Goodhue (1975) 8, on the deipnokrites of the Syrian shrine; and perhaps
CIL 6.8750f., the archimageirus. At Philippi, note the inscription of a Syrian cult
group in Collart (1937) 475, d. Starcky (1949) 66 n.6.
21. The three known inscriptions of this cult that mention a triclinium (one,
iussu dei) or cenatorium come from Virunum (Noricum), Bononia, and Rome.
They are handy in Kan (1943) 113 and 127, with description of the Virunumarchaeo·
logical remains, p. 77 (one room with hypocaust for winter meetings).
22. At Cologne, Soli Sarapi cum sua cline, ILS 4394=CIL 13.8246; also Tert.,
Apol. 39.15; Vidman (1969) 55, the second· or third-cent. dedication of a mensa to
Isis at Am phi polis; and Ael. Arist., Or. 45.27, addressing an eastern audience, in·
stanced by Milne (1925) 8. This evidence makes me think the Isiac cult meal was
widespread outside Egypt, pace Vidman (1966) 115, even if direct testimony is thin.
23. Collart (1944) 136-138, updated by Gilliam (1976) 316, who gathers the
(second- and early third-cent.) texts and (p. 320) explains the use of non-Sarapis
 Notes to Pages JS-40 161

quarters. The "Dionysius" quotation is from PYale 85, to which editors add that
the invitations were usually oral, hence few survive. Milne (1925) 9 draws attention
to the 8t:utV11t~p\ov of two local gods at Karanis, in SEG 20.652 (A.D. 69/79) =OGIS
671 =IGR 1.1120; also to the beer bar in a Sarapeion, PLon. 1177, to the word ol1Co<;
in one of the invitations, and to PFouad 76, an invitation to an Isis meal. Note also
POxy. 2976 (second cent.), mentioning that "I was dining with my friends yesterday
in your [the goddess Thoeris's) most fortunate precinct," and Orig., C. Cels. 3.17,
remarking on "the magnificent tents, CJ1Cllvai, around" Egyptian temples. On the
word similarly used in an Asian shrine, see Robert (1945) 1.49.
24. ILS 7215; CIL 6.10234 (=FIRA 2 3.106), CIL 12.2112, and other inscriptions in
Waltzing (1895-1900i 1.210f., adding CIL 10.6483 (Hadrianic: Axo[ranus?]=of
Anxur?) and CIL 6.2219, a gift of porticus cella culina ara to a city ward, for an
unspecified cult. For the Greek equivalent, found rarely, see Poland (1909) 454f.
Greek cult groups generally built or owned their own chapels.
25. Apol. 39.15; cf. the oo~xoa\ov and 1Catdd.l1CJ\<; ofthe cult dancers at Ephesus,
and there, too, the "procession that carried the picnic," 8t:\xvoq~op\a1Ctlxo~~. in
Artemis's honor, in C. Picard (1922) 300 and 312£. The dancers met in the pry-
taneion, on whose columns their names were inscribed each year, in the first cent. to
ca. A.D. 200, cf. Keil (1939) 126f.
26. For example, Cumont (1896-1899) 1.68f.; idem (1915)206, bones including of
boars, in Roman and German Mithraea; Vermaseren (1956-1960) 1.185 and 2. 79 and
213. bones of birds being most common; and Bull (1978) 79, at Caesarea.
27. To the better-known evidence from the west, CIL 8.20780, l>epe (1964) 826,
CIL 2.266, a tomb equipped cum munitionibus tricleae, and Gallavotti (1972) 360,
add Negev (1971) 111 and 127-129, the practice accommodated in much of the
Levant by tombside stone tables, esp. in Nabataea of the first cent. to ca. A.D. 300.
28. "Recent studies," and good ones, beginning with Nock(l964) 72-76; Gordon
(1975) 235f.; Kane (1975) 321 and 332-349 passim; Turcan (1978) 148, declaring
"presque tout ce qu'on lit sur ce sujet dans Les Mysteres de Mithra de F. Cumont est
honnetement purement conjectural"; and Meslin (1978) 296-305 passim.
29. Hinrichs (1970) 29-31 quotes from a novel in a late second-cent. copy,above,
chap. 1.1 n.47; compare Dio 71.4, the Boukoloi in Egypt kill Romans "in sacrifice
and swear an oath on the entrails"; the cult society of Priapus, worshiping his
genius in his sacellum with its drunken and erotic orgies, pervigilia, in Petron., Sat.
17 and 21; also Robert (1943) 190, on the festivals typical of villages around Nicome·
dia, where "il s'agissait essentiellement d'une beuverie." The poor rustics of the east
find a match in those that Ovid observed, Fasti 3.541£., senem potum pota trahebat
anus, or in the scene, surely meant to be familiar, of the half-drunk farmer coming
from worship with his family (Tibullus 1.10.51; cf. Propertius 4.4.76-8).
30. E. Josi in Enciclopedia Cattolica s. v. Refrigerio (1953) 631 and Fevrier (1977)
35, citing fourth· and fifth-cent. texts.
31. Clementis Recognitiones 4.13 (third cent.?); a similar scene in Tert., Apol.
35.2, connected, however, with emperor worship, which I exclude from this study.
For the open-air eating at such occasions, see also Acta Marcelli 1, cum omnes in
conviviis epularentur, at Tingi.
32. Haussoullier (1899) 317, emending and commenting on CIG 2883c, which
gives many details; cf. Laumonier (1958) 572-576 on the festival in question. Big
givers to such occasions might give often. See, e.g., the priest of Aphrodisias bon·
164 Notes to Pages 40-41

ored "for many l..moupyia~ Kai tcr-ruia&l~,'' in Reinach (1906) 115. In Palmyra and
Syria in general, see cult feasting in Milik (1972) 156£.; Teixidor (1977) 114, A.D. 114
in Palmyra; and Lucian, Dea Syria 49, the carcasses to be eaten being hung in readi-
ness from the trees in the Hierapolis shrine.
33. For secular or semisecular (that is, imperial cult) purposes, citywide parties
are very amply attested in the west. For a cult feast as an exception that proves the
rule, see ILS 6328, Misenum in A.D. 169, a feast offered to the city senate on the
occasion of "the all-night celebration for the ancestral god." Both the practice of the
nawuxi~ and, probably, the deity are eastern.
34. Greg. Nyss., Vita Greg. Thaumaturg., PG46.944. The Apologists offer many
scattered phrases similar to this passage, but in bits and pieces.
35. Lucian, De sacrificiis 13.
36. Firm. Matern., Math. 3.10.6, 4.10.8, and 4.14.5, constitutes a very good source
to prove the substantial profits normally to be expected from serving in a temple,
not only as priest but in other capacities. A. H.M. Jones (1940)228and Magie(l950)
545 and 1404 provide good discussion of sale of priesthoods and revenues therefrom,
some very large. Add PTeb. 298 (A.D. 107/8) lines 32-46; Herzog (1931) 31 no. C55
and 34 no. 068, grateful patients must give a gift to Asclepius; and Roebuck (1951)
28, describing the offertory box to receive the gift, here and at other shrines (this one
found filled with coins of the second cent. B.c.). All of this finds confirmation in the
accusation by Christians that Asclepius's cures had to be bought, cf., e.g., Acta
lohanni, in Acta Apostolorum Apocrypha 2 (1898) 206, eds. R. A. Lipsius and M.
Bonnet. Forthewest, noteTert., Ad nat. 1.10.24, more compendiously inApol.l3.6,
"you [pagans] exact a profit for a place at the temple, for access to the rites ... ,"and
more to the same effect, 42.8, speaking in broad terms; a narrow example in CIL
9.3513, the priesthood to be sold, the hides to go to the temple of Jupiter at vicus
Furfo, by rules posted in 58 B.C.
37. Plin., Ep. 96.1 0, condensed; hides reserved, cf. the preceding note and the very
fragmentary second-cent. Phanagorian lex sacra, Sokolowski (1969) 176f., with the
late second cent. Attic inscriptions, in Lane (1971-1978) 3.8 and l2f. =IG 22.1365f. =
SIG 5 1041 =Sokolowski (1969) 106f.=Oliver (1963) 318, defining the date; other por-
tions of victims specified in the (undated and unpublished) Didyma lex sacra in
Robert (1945) 1.49, also specifying "sale of the whole by weight ... the cooks shall
sell the heads of the animals ... ";Sokolowski (1962) 203£., containing the third-
cent. Ephesian regulations. For ridicule of the gods' scraps, see Clem. Alex., Strom.
7.6.89f. (PG 9.441) and Tert., Apol.14.1. For mention of "the god's table," see 1Cor.
21; Ps.-Clem., Hamil. 9.15 (PG 2.252), the text of the early third cent.; IGBulg. 1592
=IGR 1.743, Beroe; Lane (1971-1978) 3.13, Antioch in Pisidia; and the mensae
above, nn.l4f. The most specific information lies in GIL 6.820 (Rome), a fragmen-
tary price list of fees for wreaths, hot water, various sacrificial animals in parts or
whole, etc.
38. No part of sacrificial victims to leave the precinct: Paus. 2.27.1 and 10.32.14
and Sokolowski (1969) 106. My quotation is Clem. Alex., Strom. 7.6.1 05 (PG 9.445 ).
Compare Lact., De mort. persecut. 11.1, Galerius's mother "used to offer sacrifices
with almost daily sacrificial feasts, and gave banquets to her fellow villagers"; also
Dio 51.1.2, "sacred" games are so-called as having distributions of meat.
39. Sacrificial horses in Koethe (1933) 18, without comment, speaking of Ger-
many, and in Dell (1893) 185, at the Carnuntum Dolichenus shrine, also without
 Notes to Pages 41-43 165

comment (though note that Scythians=Celts were known to sacrifice donkeys, cf.
Clem. Alex., Co h. ad gent. 2.29.4). At the Pompeian Isis temple, bones of fowl pre-
dominate, cf. Tran Tam Tinh (1964) 37. On Mithraea, above, n.26. But all of this
represents a small and poorly informed sampling from much ' evidence.
40. Carton (1908) 28, and details in 112-116, the period of activity going down to
the late third cent.; concerning painted stelae, temenos arrangements, etc., at other
sites, cf. Merlin (1910) 8, 19 and 36f. ("repas sacres''); Toutain (1892)4, 13, 17 and 22
(A.D. 182, cf. other dated stelae at pp. 38f.); sacrifices capite ordinato, p. 22, on CIL
8.16749=1LA lg. 2977, different from capite viso, where it is the god's head referred
to, cf. LeGlay (1966) 339, also AE 1942-1943, no. 31, genio Amsige ex viso capitis.
Further, on stelae and their scenes, see Cintas (1947) 58, 73, and 77; and Foucher
(1964) 36-39; but cf. LeGlay (1966) 305 and 345-349, holocausts for Saturn, wor-
shipers eating nothing.
41. Ael. Arist., Or. 48.27 (I p. 472 Dindorf); long stays, Kotting (1950) 25 and
Henog (1931) 33 no. C64 (four months). For accommodations, see above, chap. 1.2
nn.49f. and, on the lack at Pergamon, Behr (1968) 30. Pausanias says "people live
round about [the Sicyonian Asclepieion], mostly suppliants of the god" (1.21.4).
42. Philostr., Vita Apollon. 4.18; Xen., Ephes. 1.5.6 and 9.
43. The term is ot KatOtlCoOvn:~. at the shrine of Hecate at Lagina, Hatzfeld
(1920) 75, where they have some corporate existence and speak as a body (like the
Isiacs at Pompeii?-who support this or that candidate for office with their votes, cf.
Castren [1975] 115); also at the sanctuary of PlutoandKoreatNysa,cf. Radet(l890)
233, and of Apollo at Didyma, CIG 2879, in Haussoullier (1902) 282 n.l, with whole
villages embraced in the circuit of the temenos. Note at Panamara and Lagina the
distributions "to the poor," below, chap. 1.4 n.20, and the reference to "the thief,
brigand, kidnapper, and every sort of criminal and sacrilegious person'' to be found
in the Ephesian Artemision, Philostr., Epp. Apollonii 65 (1.363). For crowds of
fugitives at temples throughout the Greek East, see Tac., Ann. 3.60-63 and 4.14.
44. Acts 19.24; Philostr., Vita Apollon. 5.20; cf. the plumbarius at Hierapolis
discussed by Rey-Coquais (1967) 57; the !!OPE'If6c; at Philae, discussed by Bernand
(1969) 2.97; in Africa, the third- or fourth-cent. maker of lamps who inscribed on his
product, "Buy lamps, sieves (?), statuettes," CIL 8.22642, 6; perhaps the "mer-
chants" in the agora at Andania (SIG 5 736, but cf. above, chap. 1.2 nn.32 and 37, on
temple markets defined by priests but overseen by secular officials); and a great deal
of indirect evidence of mass-produced, stereotyped objects to be dedicated, e.g., in
Merlin (1909) ccxxxvii (figures ofTanit plus dove), or Bessou (1978) 204-206, terra-
cotta Venuses.

4. Routine Staff and Administration

I. Wilcken (1912) l, 2.65, A.D. 232 at Elephantine; idem (1885) 436, A.D. 215 at
Arsinoe; cf. Constantine received at Autun in A.D. 310, where a spokesman says, "we
decorated the streets leading to the palace with ... the images of all our gods,"
Paneg. vet. 8(5).8.4.
2. The external j usticiars to cities of the Lycian koinon registered their names in
the temenos of Zeus Osogos, LeBas-Wadd. 352...;358; cf. emperors' calls on local
deities above, section 1, n.20.
3. Suet., Aug. 72.5, the emperor persaepe presided over trials in a temple at
166 Notes to Pages 43-45

Tibur; at Miletus, "the senate-house was officially dediCated to the Didymaean

Apollo," cf. Laumonier ( 1958) 557, just as at Ephesus a cult group held its symposia
in the prytaneion, cf. Keil (1939) 127, and the prytaneion of Elis was built right
inside Zeus's sacred grove, Paus. 5.15.8; and note at Cosa north of Rome, a Mith-
raeum "constructed in the cellar of the curia in the late 2nd century and continued
in use through the 4th," cf. Collins-Clinton (1977) 23 n.43. In Dardanos, the god-
dess's parade started from the prytaneion, cf. Sokolowski (1969) 224, and a similar
role for the prytaneion at Stratonicea, below, n.22.
4. Priests or devotees of Ma =Bellona in, e.g. Tibullus 1.6.43 and Tert., A pol. 9,
and Lact., Div. inst. 1.21; of Cybele, d. e.g. Ps.-Lucian, Lucius 37£., and, on howl-
ing, Graillot (1912) 124 n.6; of Atargatis, Lucian, Dea Syria 50£.; of Isis, Plut.,
Moral. 364Ef., and Firm. Matern., De errore profan. relig. 2.3.
5. Ibid. 4.1£., Aug., Civ. dei 7.26, and ILS 4168, in Cybele cult; compare Clem.
Alex., Coh. ad gent. 10. 74, on Isiacs "with dirty long hair, disgraceful in unwashed,
tattered clothing, knowing nothing at all of a bath, like wild animals in the length
of their claws, and many castrated ... "; Lucian, Alex. 13, long hair for a new faith;
and often in plastic depiction of holy or inspired men.
6. Tert., De idol. 18.1 is the most general statement; on the other two cults, d.
Vermaseren (1977) 97 and Cumont (1929) 241 n.78, with idem (1975) 183 n.l74.
7. IGLS 2928, in a temple ten miles from Baalbek, Hocmaea virgo dei Hadara-
nis, qui annis xx panem non edidit (I) iussu ipsius dei (second half of the second
cent.), comparing Hier., Adv. lovinianum 2.17 (PL 23.316), fasting lsiacs, and
idem,.Ep. 107.10 (PL 22.875£.), IsiacsandCybelevotaries, with IGLS2929, showing
that H ocmaea virgo deae Syriae lived 100 years. For a second example, see Audollent
(1890) 534, Saturninus sacerdos. Si quis possit observare vinum non bibere annos
triginta octo, menses septem, ipse possit sacerdos esse. On fire-walking, see Strabo
12.2.7 p. 537, with Magie (1950) 1151 concerning the locationandDupont-Sommer
and Robert (1964) 62 on the rite. They compare Iambl., De myst. 3.4. In Macedonia:
New York Times, June l, 1956, with photo.
8. Sen., De superstit., in Aug., Civ. dei 6.10, trans. Loeb ed. The date when the
work was composed is uncertain: A.D. 31 up to the 60s, cf. Attridge (1978) 67 n.l59.
9. On Sarapis cult, see Apul., Met. 11.20, Porph., De abst. 4.9, and Arnob. 7.32;
on Asclepius's cult, cf. Aelian frg. 98 and Ael. Arist., pp. 378, 453, and 458 Dindorf,
in Pergamon, with Nilsson (1945a) 67f. and Latte (1931) 133; ibid. 134 on Dionysus
cult in Teos.
10. LeBas-Wadd. 90, Teos; IG 4, }2 pp. 173f. on Epidaurus; and pyrphoros is an
official found in many cults.
11. Sokolowski (1962) 60f.=IG 4, 12.742 add., hymns to Asclepius; Tibullus
1.3.30, to Isis.
12. Corbett (1970) 150£., temples always closed (and others always open, cf. also
Stambaugh [1978] 571); Ps.-Lucian, A mores 12 and 16, the Aphrodite temple freely
entered in daytime but closed at night; Min. Fel. 22.8; Rebuffat (1975) 501, benches
inside the Mars temple southeast of Lepcis; sacrifices inside temples in Greece,
Corbett (1970) 150; lattice-work barricades around temples shown in Price and Tre11
(1977) nos. 8, 165, and 262-266, cf. Sa1viat (1963) 260f., on pre-Empire barriers.
13. Dedicatory lamps common, e.g., in Africa, cf. Toutain (1892) 17 and idem
(1907-1920) I, 3.64; DE s.v. Lucerna (M. Manni 1975) 1953; in Rome, GIL 6.368 (a
 Notes to Pages 45-49 167

bronze lamp), Goodhue (1975) 28, and Vermaseren (1956) 1.192 no. 473; in Greece,
Laue (1931) 134 and Nilsson (1950) 105. For statues with torches, cf.IGLS 2716=
ILS 9416, and Rey-Coquais (1967) 52 and n.3; Robert (1964) 155 and 157.
14. Piganio1 (1946) 250, pointing further, p. 251, to the effect intended by the rite,
to "developper aux yeux du peuple des tableaux frappants."
15. Even legionary eagles were anointed on feast days, Plin., N.H. 13.23; for kos-
mophoroi and the like, see C. Picard (1922) 242 and Cumont (1929) 89.
16. Reddite cantu, in Vermaseren (1956-1960) 1.99 no. 484; d. Vermaseren and
van Essen (1965) 187-232, esp. 223, hymns composed by a priest?
17. Not mentioned by Ferguson (1970), Wissowa (1912), or Latte (1960); in
Nilsson (1941-1955) 2.216, mention only of the miracle of ?40 B.c., the cult other-
wise dismissed in a few lines (1.127): "der Kult is nicht griechisch."
18. 'Evepy&ia\, in Hatzfeld (1927) 62, the text partly restored. It would be need-
lessly cumbersome to document every point of description separately. Laumonier
provides full, dear treatment of Stratonicea; very briefly, on Panamara alone, Cook
(1914-1940) 1.18-24. I also draw on Roussel (1931) 85, a text praising Zeus "since [he
has shown us many great and wonderful manifestations] for the salvation of the
city, from of ancient times ... ," the date (pp. 90-92) lying in the late first cent. B.c.;
and a similar text in Hatzfeld (1927) 63, "since the greatest and most wonderful Zeus
of Panamara saves the city ... ," (the date, p. 64, lying in the late first cent. A.D.). For
more on the shrine, I draw also on Hatzfe1d (1927) 69 (remains evidently of a lex
sacra) -109; Deschamps and Cousin (1887) 373-378; idem (1891) 209; Cousin (1904)
26 (theoriae); and Roussel (1927) 124, "the theater which is the i£p6v itself of Pana-
mara.''
19. On the Panamaran "mysteries" see Roussel (1927) 126 and 132-134; Lau-
monier (1958) 247, 255-259, and 323-329; and Kane (1975) 331£.; on the Lagina-
mysteries of Hecate, see Deschamps and Cousin (1891) 174 and Cousin (1904) 241.
20. Laumonier (1958) 365; on Lagina, further, pp. 370-382 and 392-401; Diehl
and Cousin (1887) 158; Hatzfeld (1920) 75-77 and Laumonier (1958) 346, actions
taken jointly by boule, demos, gerousia, and oUv t4i lep<j) (ortcj) 7t£putoi..iq>) KatO\·
Kouvte~; assistance to the poor, pp. 74 and 89, d. at Panamara, Laumonier (1958)
267.
21. Laumonier (1958) 402, on C/G 2715a-b=LeBas-Wadd. 519-520.
22. Above, chap. 1.1 n.67; Laumonier (1934) 85 and idem (1958) 276and 388 on
LeBas-Wadd. 519f.; d. Laumonier (1958) 275, the not-fully-explained Zeus-wrought
miracle whereby (as the inscription records) "the ox in piety toward the priest,
during the parade, of its own accord leads the priest to the senate-house and then,
after the sacrifice, went off right away."

CHAPTER II: DEBATABLE

1. Needs and Answers
l. For arotr\p as Asclepius's most frequent characterization, see Bieler (1935-
1936) 1.121; examples in Herrmann and Polatkan (1969) 46f., near Pergamon;
Weinreich (1909) 30, at Rome; Robert (1973) 161£., at Messina; /G4, 12.127 (A.D. 224)
and 415 (A.D. 258), at Epidaurus; no. 417, the god addressed without name, simply as
168 Notes to Pages 49-51

"savior," as also in Crete, above, chap. 1.2 n.82; his function implied in all the testi-
monies of suppliants "saved," <no9&i~. aco9etaa, at Insula Tiberina, Longo (1969)
1;84, or at Pergamon, IGR 4.279 and Ohlemutz ( 1940) 155 and 158. For healing by
Isis and Sarapis, c£. Strabo 17.1.17, Aelian, Hist. animal. 11.35 (with a welcome
pun), SEG8.464, and Vidman (1969) 188and 159. Forothergodsalsohealing, c£. for
example Syrian ones in LeBas-Wadd. 2343 and Gawlikowski (1973) 94; Apollo,
CIG 6797=/G 14.2524, at Autun; Sol, CIL 3.5862, at Augsburg; and Hestia and
other deities at Ephesus, in Robert (1955) 100.
2. Philostr., Vita Apollori. 4.10 and 8.7.9, invoking Hercules Apotropaios, c£.
Tat., Ad Graec. 16 (PG 6.811B), demons cause disease, and the same view held by
superstitious people, Plut., Moral., 168Bf., and by Gnostics, according to Plot.,
Enn. 2.9.14.
8. Philostr., Vita Apollon. 4.20, c£. exorcism practiced on sick animals, ibid.
6.43 and Veget., Mulomedicina 3.12; exorcism also in Lucian, Philops. 16, the
reference to the specific exorcist judged historical if not identifiable by Betz (1961)
12; other nameless practitioners in Firm. Matern., Math. 3.4.27 and 3.8.9, and in
Orig., C. Cels. 1.68. As a youth, Marcus Aurelius was taught skepticism about
pagan exorcists, Medit. 1.6, but Tatian, Ad Graec. 18, had no doubts, nor the pur-
chasers of home remedies, such as Cyranides 1.3.10, 18.25 de Mely.
4. Just., II A pol. 6. Cf. Dial. 76.6, and various contemporary Apologist passages
cited by Archambault (1909) ad loc.; Iren., Adv. haer. 2.32.4 (PG 7.829A), comparing
the Jewish exorcists at 2.6.2 (PG 7.724f.); Tert.. Apol. 23.4 (quid hac probatione
jidelius?)-21.1; idem, De spect. 26; and Orig., C. Cels. 1.46, c£. 1.6; also Theoph.,
Ad A utol. 2.8, "up to the present day the possessed are sometimes exorcized"; Cypr.,
Ad Donat. 5; Tatian, loc. cit.; Acta Petri et Pauli 33, p. 193 Lipsius, written in Asia
before A.D. 200, says James (1924) 300 and 304£. Later sources include Greg. Nyss.,
Vita S. Greg. Thaumaturg., PG 46.916 and 944£., for the mid-third cent.; Athanas.,
VitaS. Anton. 40, 63, 64, and80; Or. de incarnat. verbi 19, 81, and48(PG25.129, 149,
and 181); Euseb., Demon. ev. 3.6 (PG 22.238-236); C. Hierocl. 4 (in the Loeb ed.,
with the Hist. eccl., vol. 2 p. 493); Arnob. 1.43-45; Lact., De mort. persecut. 10.2f.
and Div. inst. 2.15(16) and 4.27.1; and Firm. Matern., De errore prof. relig. 18.5f.
Exorcism was to Christians "a deliberate and official activity and not ... a private
trade pursued for profit," as Nock (1933) 104 sees.
5. Epigraphic equivalent of dejixiones is very rare. An apparent exception may
prove the rule: IGR 4.93, fragmentary, the curse likely invoked on some criminal.
6. Lact., Div. inst. 5.20, with an echo in Tert., De spect. 25.102. Lactantius
emphasizes his case by joining a brigand to the gladiator, and a poisoner, etc. But
such criminals could only invoke demons, as he well knew. Similarly, Tertullian
rebukes the admiration of scaenici, De spect. 22.101; but open inquiry of Apollo
may be made by an acrobat about his upcoming performance, c£. Vidman (1969)
147, or Philostr., Gym. p. 283 Kayser (=Jiithner 41), where a nameless god sends a
vision that saves a wrestler's career.
7. Baldness, ILS 3135; fertility, Hall (1977) 193, but probably before our period
of study; Zeus 1tA.outoMt11~. Vidman (1969) 196, under Caracalla, and on coins of
Nysa under Nero, BMC Lydia p. 175, and Men with the same epithet in Lane ( 1971-
1978) 1.91; to Silvanus, ob libertatem, GIL 6.663; to Mars, ob immunitatemo£ a cult
guild, RIB 309, A.D. 152; to Isis, ob remissa exacta inlicita populo a maximis im-
 Notes to Pages '1-'2 169

peratoribus (Severus and Caracalla), GIL 11.1585, Florentia; lightning, in Robert

(1949) 31 and Taus and Berardi (1972) 86-88; to Poseidon for safety during an earth·
quake, Perdrizet (1900) 431, at Gerasa and Byblos; drowning, GIG 2716=LeBas-
Wadd. 516, IG 14.997 and 10, 2.67,cf. BGV 423 (second cent., toSarapis)andArrian,
Ep. ad Traianum 34, ed. C. Miiller, Geographici graeci minores 1.399; unspecified
"dangers" in IGR 1.107=IG 14.1030, GIG 3669 (Cyzicus, second cent.), and other
texts in Robert (1945) 1.21 and 24, curiously formulaic; life saved, in Gawlikowski
(1973) 64 and ILS 3160, Marti Augusto conservatori corporis sui; danger on a jour-
ney (the stele showing a trotting horse in relief), LeBas-Wadd. 515, Stratonicea, and
DeRicci (1934) 257-259 (Severan, in Egypt); safe return from danger, to Hera, for the
dedicant's sons, lnscr. ltal. 10, 3.115, and for troops, Milik (1972) l97f. (A.D. 316,
correcting ILS 8882, Coptos), Rebuffat (1975) 504 n.28=IRT 920 plus another,
clearer inscription; Winnett (1957) If. (dates) and, e.g., nos. 9llf. and 714f. (safety)
and 253 and 700 (booty); for the municipal slave, cf. GIL 11.4639, Tuder. These
thirty-odd examples of inscriptions which name the cause of thanks, excluding the
Safaitic, could no doubt be doubled-but not quadrupled, it is my impression. A
small corpus to work with.
8. A sampling in MAMA 7.243; Lane (1971-1978) 1.60£., children, that word
alone, or "of his own" or (1971-1978) 4.18f., "wife, children and slaves"; IGR 1.44,
"of wife and children"; "of his own [kin]," in Rey-Coquais (1972) 87; "of my
masters, self, and all that are mine," GIG 3792, in Robert (1955) 32, with other
prayers for "masters," pp. 29-31, or "men" and "possessions," idem (1939) 203;
prayers for "all who are his" vel sim., in Drew-Bear (1978) 43 and 47; "for his son,"
in Kalinka (1933) Beibl. col. 62 and Engelmann (1976) 108 no. 42; "for wife, children
and kin," IGLS 2728; Merlat (1951) 142; and Rostovtzeff (1937) 204, at Dura in the
Atargatis temple. In the west, cf. Merlat (1951) 35 and 249, GIL 12.403, 8.2625, and a
variety indicated by the abbreviation p(ro) s(alute) s(ua) s(uorumque), 3.1133, cf.
7505.
9. "For the stock," ~fttp -rillv ~cprov, in Schwertheim (1978) 793, "to the god
Prietos"; pro armento, in GIL 12.4102, to Silvanus, cf. Cato, De agr. 83, how to offer
sacrifices pro bubus, uti valeant, to Mars and Silvanus; for a mule, LeBas-Wadd.
686, to Men; for the oxen, LeBas-Wadd. 1192 (Pisidia, A.D. 163); Hasluck (1904) 23,
from Thrace; in Phrygia, Drew-Bear (1978) 48f. and Robert (1939) 204,citingmany
of the texts in MAMA 5 (add no. 50 of that vol., an anepigraphic relief); for hunting-
dogs, see Robert (1955) 29 with Arr., Gyneg. 33, where one prays to Artemis for a
good dog. For prayers for crops in Phrygia, d. Odp ICcipncov or sOOsvdac;, in MAMA
5.79 and pp. xliii, and Drew-Bear (1978) 38; in Bithynia, Sabin (1978) 775; in Pisidia,
ibid. 780; in Pamphylia, Robert (1955) 240, "to Zeus Galactino according to Apol-
lo's command"; in Bithynia, "for the vines," Hasluck (1904) 23 and IGR 3.36 (A.D.
161 ), and in Thrace, IGBulg.2 374; in Phrygia, MAMA 5 p. 42, the editor referring
also to 1.8 (Phrygia) and Dionysus's epithets "of good harvest," vel sim. For con-
scious discussion of the gods' role in providing good harvests, d. Paus. 2.25.10, RE
s.v. Zeus, I:Epiklesen (H. Schwabl, 1972) col. 344, Iambl., De myst. 5.6, and Max.
Daia in Euseb., H.E. 9.7.8.
10. No fertility deity-in our period (whatever may have been the origins and
scope of deities in darker centuries). On Dionysus in that role, texts are not numer-
ous, but they are roughly in proportion with the total of those that link any deity
170 Notes to Pages 52-5J

with any sphere of activity. See examples in the preceding note, plus CIL 5.5543, to
Jupiter and Liber vinarum conservator, and 8.24520, to Liber from oenopolae cum
meraris omnibus ( =tabernarii?) of Carthage; but be warned by Firm. Matern., De
errore profan. relig. 3, showing that Attis cult serves the fertility of the fields; and
PO xy. 2782 (second/ third cent.) showing sacrifices to Demeter (here= Isis?) for good
crops and weather is the kind of text that should be common and is not. The more
familiar spheres of deities may not be the real or chief ones-as indeed of Liber pater
himself, d. above, chap. 1.1 n.I 7.
11. For Mercury=Hermes guarding peddlers, cf. Finke (1927) 7, Mercurius pere-
grinorum; Barkoczi et al. (1954-1957) 1.326, Mercurio lucrorum potenti, NN nego-
tias(! ); and Robert (1966) 25, Hermes statues often preside over weights and scales in
the market; for Silvanus, guardian of beasts in the woods, d. preceding note and
Tourtain (1907-1920) 1, 1.265, dedications by ursarii vel sim.; Tert., De idol. 10.2,
the schoolmaster thanks Minerva; Poseidon, earthshaker, in the preceding note and
in Philostr., Vita Apollon. 6.41; and generally on the appropriateness of deities, the
discussion in Plut., Moral. 757C-758E. For deities of women and children, above,
chap. 1.3 n.l and Hatt (1979) 131, in the period A.D. 150-225.
12. Tran Tam Tinh (1964) 117, referring tojuv. 6.522; d. Juv. 6.531£. and passim,
Min. Fel. 22.8, the disapproval and ridicule that pagans nudi crude hieme discur-
runt, and Bickerman (1973) 11, on Lactantius's views and a pagan's probable re-
sponse. For explicit examples of do ut des, as easily found in Greek as in Latin
inscriptions, cf. Keil (1939) 119, in Ephesus, the wish that Hestia and Artemis may
serve as the dedicant's advocate since she had spent her wealth freely on them (date,
under Marcus Aurelius). This may be compared with the broad, self-comfortable
statement about worship in Art., Oneir. 2.33. In the west, see lnscr.ltal. 10, 2.194, ex
voto quot (I) a dea petitconsecutus; CIL5.6505,Mercuri ... , utfacias hilaressemper
tua templa colamus; CIL 5.3221, .quot (I) se precibus compotem fecisset (sci!. Deus
Aeternus), cf. 5.3321, or the same thought so much a cliche that it is abbreviated voti
c(ompos) d(at), CIL 6.47, and Cagnat (1898) 440, v.c.l.m. For more examples of the
cliche, see LeGlay (1971) 130.
13. Dubois (1902) 27, on "a renascence of Dionysiac mysteries in Italy." But the
evidence does not exist: Dio 76.16 (no word of mysteries), cult colleges at Ostia (CIL
14.4, the date too early by nearly a century) and at Rome (CIL 6.641, no word of
mysteries). Charles-Picard (1954) 203-205 detects the same renascence, based on sar-
cophagi reliefs, the symbolism of which may show the promise of immortality. The
inference seems quite subjective, at least not well supported. See Nock (1946) 140,
critical of the inference in its best-known version, Cumont's, and further, Andreae
(1963) 81 n.445, supporting Nock and citing F. Matz to the same end. On the ques-
tion whether Dionysus cult could insure immortality, Bruhl (1953) 309 professes to
lean on Cumont (1942), a work in which I can find nothing solid, and(cf. Bruhl 314
n.20) also Cumont (1949) 256f., again not useful. Bruhl himself (p. 312) offers a
number of assertions while admitting scanty evidence beyond the sarcophagi: CIL
6.30122, which provides no support, and Plut., Moral. 61IE, depicting the soul as
undergoing modified metempsychosi,s, and without individuality. Plutarch tells us
that he relies on teachings partly ancestral, partly Dionysiac; but they appear rather
Neoplatonic.
14. The "mysteries" are new features within the period of the Empire, d. F.
 Notes to Pages 53-54 171

Dunand (1972-1973) 3.244f. and 253. The evidence lies in Plut., Moral. 351Ff. and
364E, Heliodorus, Ethiop. 9.9, and Hippolyt., Refut. 5. 7. The arguments of Merkel-
bach (1962) 12 and passim, that initiatory rites can be found hidden in Apul., Met.
4.33f., seem quite unsubstantial and can be set aside in the light of Williams (1978)
194 and Kane (1975) 342. InApul.,Met., noteesp.ll.6, vivesbeatus, life may even be
prolonged a bit; but then, ad inferos; 11.16, ter beatus for winning Isis's patro·
cinium, ut renatus quodammodo-meaning, for Lucius, spes futura beneficiis
praesenti bus pignerata ... ; 11.24, beneficia pigneratus; and 11.30, Lucius earns
money liberali deum providentia. The real benefits are earthly.
15. Pie(?) rebus renatum, etc., in Gordon (1975) 236, a far less full text inVer·
maseren (1956-1960) l.l99f. no. 485 (A.D. 202); idem, (1969) 254, offering a new
version. Franz Cumont, after more confident statements in the past, finally con·
fessed in 1947 to "our well-nigh total ignorance of the mysteries," in Mithraic
Studies (1975) 198. For the third text, CIL 6.736 (A.D. 391), arcanis perfusionibus
(not profusionibus) in aetemum renatus, c£. Meslin ( 1978) esp. 303-305, apparently
unaware that itisa forgery (thus omitted from Vermaseren's Corpus and specified as
false by Duthoy [1969) 66). The date and the echoes of Christianity and of non·
Mithraic taurobolia, cf. below, n.23, would in any case deprive the text of value for
our purposes. Finally, we have CIL 3.686 (third cent., Philippi)=Vermaseren (1956-
1960) 2.382 no. 2343, in part reading, sic placitum est divis aeterna vivere forma qui
bene de supero lumine sit meritus, "gifts promised to you in your pure course of life
by the natural innocence once commanded by the god.'' The verse epitaph is pretty
poetry, no more. It portrays afterlife "as a Satyr" summoned to Bacchus, or to the
Naiads. It has no reference to Mithraism. As to the "dead" man shown in Mithraic
frescoes, recalled by SHA Commod. 9, that need not be the initiate, about to be
resurrected; pace Meslin 306, that item in the rites may have many meanings un-
related to resurrection.
16. Orig., C. Cels. 6.22, with bibliog. ad loc., H. Chadwick(1965) 334; Porph., De
abst. 4.16, but these "statements ... about metempsychosis are not reliable," says
Gordon (1972) 115 n.26. He gives no reason for saying this, but Porph., De antra 6, is
compatible with Celsus-still more so, the statement ibid. 29, if that passage were
tied by Porphyry to Mithraism (it is not).
17. Nilsson (1949) 765f. on the fresco called by Kane (1975) 343 "the most con-
vincing instance known' to me of an earthly banquet foreshadowing the blessed
feast of the future life"; but even here (Nilsson, p. 764), the inscription next to the
dedicant contains common and quite opposed pagan sentiments of the sort, "eat,
drink and be merry, for tomorrow you die."
18. CIL 6.32316, date, A.D. 180, says Goodhue (1975) 9. Cumont (1917) 280-282
and others thought this text was proof of belief in immortality; but (Goodhue 111-
114) still others disagree. Cumont (1917) 282 n.2, in his efforts to support his view,
offers three perfectly useless citations, to Diog. Laertes (speaking only of Pythago-
ras), Mnaseas (nearly five centuries too early), and S/G2 584 (not mentioning im-
mortality). The inscription, in its tone, seems to grin at death, in the use of the
diminutive of "soul," animula, and in the sentiment, tcp Oavcitcp JJT)Stv 6<p&li..6JJ&·
vo~.
19. On two kinds of initiation at Samothrace, see Fraser (1960) 91, where, of
twenty-nine people initiated on a single day in A.D. 19, only three received the more
172 Notes to Pages 54-56

advanced epopteia. One was the person heading the whole group. "Schinas paid
the bill for all his party," says K. Lehmann, ibid. The circumstances seem to contra·
did Lehmann (1954) 93, that selection for epopteia "was conditioned neither by
financial nor by social status." Of the Eleusinian mysteries (the Great-the Lesser
being at Agrae), Otto (1955) 23-30 offers a powerful vision; but it must be set against
the icy sarcasm of Hippolyt., Ref. 5.8.39 (PG 16.3150B ), specifically speaking of the
Great Epoptikon (therefore including the more advanced, if there were several
grades). I do not agree with Kerenyi (1955) 34 n. 7 that Great rites at Cyzicus (known
only in one inscription, BCH [1890] 537) necessarily imply Lesser. Nor does the
seating in the show-theater of Demeter's shrine at Pergamon seem to me adequate
evidence of two levels of initiation-contra, Ohlemutz (1940) 211.
20. Theon Smym., Math. p. 23 Dupuis, an interesting look at a generalized
initiatory experience, suggesting that the same procedures were found in a number
of rites. For the date, see Lesky (1966) 879, or OCD s.v. Similarly, Plutarch in
Stobaeus 4.52.49 (Loeb ed., vol. 15 pp. 316-318) mayofferageneralizeddescription.
He calls them onlyteA.etai ~eyciA.at; cf. Moral. 1105B, asserting that "lots of people,"
1tavu 7toA.A.o{, feel no fear of death, believing that the initiatory rites will assure them
a delightful hereafter, described just as in the preceding fragment. Tertullian, De
praescr. haer. 40.1-5, receives most recent comment from Gordon (1975) 234-236.
The punctuation of the passage is controverted, but I agree with Gordon that
diabolus rules all verbs save signat (my disagreement on 40.1.5 alone is irrelevant).
Echoes of word choice recall idem, De baptismo 5, per lavacrum initiantur in both
Isiacism and Mithraism. But the meaning of the rite may lie (5.1) in the Apollinian
and Pelusian Games, "where people are anointed, tinguntur, and believe by this
they are born afresh and unpunished for oath-breaking" (regeneratio to impunitas
periuriorum, best explained by chap. 1.2 nn.69f., above, and n.34, below).
21. Charles-Picard (1954) 146 n.2; Festugiere (1945) 123, Corp. Herm. X, 20.
22. Meslin (1974) 290 is only the latest in a long series of scholars to make the
assumption. He speaks of worshipers of "the Oriental cults," undefined, as identi·
fying themselves with the suffering and death of their god and his resurrection. But
of course Jupiter Dolichenus, Sabazius and Mithra are Oriental but never suffered;
and resurrection in divine stories more often calls to mind fertility (planting and
growing) than immortality, e.g., in the story of Attis, cf. Firm. Matern., De errore
profan. relig. 3, or of Kore or Osiris.
23. As pointed out by Lambrechts (1952) 163.
24. Cf. Duthoy (1969) passim, esp. 106 and 110. Meslin (1974) 310 thinks that in
CIL 6.510 (A.D. 376)=1LS 4152, renatus in aetemum through taurobolium, "Chris·
tian influence is here manifest." For the duration of the effect of the rite, usually
twenty years, cf. Carmen contra paganos 62, and note CIL 13.511 and520, in which
the emperor's salus is sought by the ordo of the city through taurobolium-hardly
his eternal life.
25. Just., I Apol. 18. He adds ref£. to Pythagoras, Plato, Homer, all remote and
literary. Cf. also the view meant to be representative (and negative) in Min. Fel. 8.5.
26. Lact., Div. inst. 5.19 and 7.13.5. For the collection of such oracles, cf. above,
chap. 1.1 n.65.
27. Orig., C. Gels. 7.32; Min. Fel. 11.2-4. Apologists in many passages indicate a
defensiveness about the idea of resurrection in the flesh.
 Notes to Pages 56-58 17J

28. Ad Lucilium 24.18 and 36.10, cf. Min. Fel. 11.9.

29. LeBas-Wadd. 2343, "saved by the savior-gods from illnesses," CJG 3827 add. q
(Rome), "Hecate savior" (hardly of souls!), or Harper ( 1972) 225, "for the preserva·
tion, arotllpia, of the priest NN, to Men," are typical of usages too common to need
further parallels. Cf. above, n.I, or, in literary sources, Firm. Matern., De errore
profan. relig. 22, "take heart, initiates, the god is saved; there is salvation for you,
out of your pains," or Orph. hymn. 67 line 8, Quandt2 p. 47, Asdepius hailed as
"savior, bestowing a good end to life" (i.e., only a good death).
30. Lattimore (1942) 74-86, esp. 84f.; Cagnat (1898) 262, n(on) f(ui), n(on)s(um),
n(on) c(uro), copying the equivalent in Greek, ibid. 262 n.4.
31. For the (still continuing) "old" view of mystic hopes in banquets, cf. the
representative conjectures of Collins-Clinton (1977) 19: "those [Bacchic] meals
would also have been considered an earthly anticipation of the continuing joyful
banquets of the blessed afterlife in which only the initiates could share." By con·
trast, c£. the writers cited above, chap. 1.3 n.28, adding Laumonier (1958) 322f. and
329, on Bacchic and Panamaran cult meals.
32. Moral. 822B; for similar statements that religion deserves support or was
invented because it helps to control otherwise antisocial behavior, cf. Critias,
Sisyphos, ad init. (A. Nauck, Tragicorum graecorum fragmenta, p. 771); Polyb.
16.12.9£.; Poseidonius reflected in Diod. 34/35.2.47; Diodorus's own views, 1.2.2;
Cic., Rep. 2.2.4 and Leg. 2.7.15f.; Plin., N.H. 2.26; Athenag., Leg. 1.2, saying all
cults are lawful "in order that men may forbear from wrongdoing through their fear
of the divine"; and Hermeneumata pseudodositheana p. 39 Goetz (later second!
earlier third cent.), Aesopic fables promote moral behavior.
33. Paus. 8.15.2, in Arcadia, "most of the citizens of Pheneus take oath by the
Petroma [a sort of betyl] in the most major affairs"; Philostr., Vita Apollon. 1.6,
near Tyana a spring of Zeus god of Oaths, where perjurers are scalded or touched by
a wasting disease, etc.; Achilles Tatius 8.11-14, testing by the sacred spring at Ephe·
sus; Plio., N.H. 31.22, in Bithynia, the river Alcas. In the west, Juv. 13.34f., with
both believers and doubters at Rome; Paneg. vet. 7(6).21.7, Apollo's spring (at
Grand and elsewhere); and testing "at the memorial to the saints" in Milan, pre·
sumably derived from pagan beliefs, Aug., Ep. 78.3 (PL 33.269).
34. IGR 4.1376; further, in Phrygia, Ramsay (1895-1897) 2.656, the violator "will
have angered the gods in heaven and beneath the earth"; 700, "the subterranean
demons"; Robert (1960a) 430 (Jewish); and IG 9,2.1201, Thessaly.
35. On the deity insuring justice in Isiacism, see the "Karpokrates" hymn quoted
in chap. 1, above; Ovid. Ex Pont. 1.1.52£.; and the late third cent. B.c.IG 11.1299
lines 26-83; on Helios, see Cumont (1923) 65£., from Rome but in Greek; Arrian on
interesting practices in Nicomedia, cited ibid. 67 (the passage cited from Vett.
Valens seems to prove little), and 69, citing Aurel. Viet. 39.13; Celsus, in Orig., C.
Cels. 8.48 (and less clearly at 3.16 and 4.10); the Ps. Plutarchan (but early second
cent.) Amatoriae narrationes, where divine vengeance enlivens fiction, Moral. 7740
and 775B and E; and background in Nestle (1948) 568-596 passim. The material in
Judaism and Christianity is obviously abundant-a sampling in MacMullen (1968)
90 n.29.
36. Robert (1958b) 112-122 passim, on The Just Divinity embodying "concep·
tions religieuses plus epurees que les paganismes classiques de Ia Grece ou de
174 Notes to Pages 58-59

I'Anatolie.'' In Mithraism, Cumont expatiated beyond measure on two fragments

of evidence, namely a Dura graffito, cf. Cumont (1975) 204 and the inscription in
Vermaseren (1959-1960) 1.49 no. 18. Gordon (1975) 231 n.64 seems a little harsh in
calling the latter text "the only evidence yet published that Mithras was conceived
as just in the Graeco-Roman world," and further depreciates even that evidence.
But his dismissal of Cumont is not far off the mark. Note Cumont' s main statement,
(1929) 147, that Mithra presides as judge against the evil-a statement supported
(282 n.68) by reference top. 83 (on Sarapis-irrelevant) and top. 92 (on Egypt-
irrelevant) and to Cumont (1942) 89£., where is offered a dualistic picture of Mith·
raism, but nothing on Mithra as judge. Here, Cumont also refers to Philo, De
praemiis et poenis 6, but the passage, contrasting heaven and Tartarus, has no rele-
vance to Mithraism. Tertullian, De praescr. haer. 40.2£., refers to the initiate's sins
(in an unspecified cult), expositionem delictorum de lavacro repromittit, which
might perfectly well be violations of ritual, not ethics. Porphyry, De antro 15, does
not decide the ambiguity, though explicitly on Mithraism. He is cited for other ends
by Gordon (1972) 114 n.24. The passage in Orig., C. Gels. 3.36, evidently regarding
Isiacism, describes oracular or healing gods that also punish people for sins against
"some rule about impure food, or touching a corpse"; similarly, Juv. 6.520f. and
535 on Isiacism show that only ritual purity (i.e., regarding sexual taboos or the
like) are in question, for expiation; so likewise PI ut., Moral. 168D, transgressions of
diet or, against promptings of one's guardian spirit (daemon ion), taking the wrong
fork in the road; but Corp. herm. 10.19-21, ed. A.·J. Festugiere ( 1945) 122£., specifies
purity through "not doing wrong to any man." The text is late for our purpose,
fourth century.
37. Above, chap. 1.1 at n.91; cf. ILS 4326, priests praise a priest/magistrate iussu
Iovis, and Seyrig (1941) 246, third-cent. inscriptions testifying to Iarhibol's approval
or support of this or that local magistrate.
38. De lege Manilia 14.42. Such statements are naturally easy to find in emperor
worship or anything close to it, e.g., Plin., Paneg. 1.4f.
39. A village in Asia Minor is the beneficiary of prayers "on behalf of the demos,"
in Lane (1971-1978) 1.69; villages pray for themselves, cf. Sabin (1978) 775; Buckler
et al. (1926) 88, "in obedience to an oracle"; Drew-Bear (1978) 38; Robert (1955) 240;
and Hasluck (1904) 22, Bithynia, and 23, Cyprus. In the west, cf. GIL 5.8208=/LS
3980, pro salute sua et suorum omnium et viciniae. Addresses to a deity by a city
corporately lie in LeBas-Wadd. 2627=C/G 4500 (A.D. ll4), at Palmyra; other ex-
amples in Hatzfeld (1927) 63£., Stratonicea, and IGR 1.568, Nicopolis ad Istrum,
"by command from a dream." Cf. above, chap. 1.1 n.l7.
40. C/G 3538=/GR 4.360, Pergamon, cited by Wiseman (1973) 178; ibid. 179 on
IGR 4.1498, Trocetta; but Keil and Premerstein (1908) 10, doubt the connection
with A.D. 166; further, Wiseman (1973) 180-182, possibly Stobi; Robert (1973) 162,
possibly Messina; above, chap. 1.2 n.67, further instances; Philostr., VitaApollon.
6.41, earthquakes in Asia under Domitian; and consultation of Apollo's oracle by
Caracalla (?),in Birley (1974) 511£., by Diocletian, in Lact., De mort. persecut. 11.7,
and by Asian cities on a regular basis, above, chap. 1.1 n.88.
41. Ael. Arist., Or. 28.39, 1 p. 498 Dindorf; SEG4, 1 (1929)467, d. Prosopography
of the Later Rom. Emp. s.v. Festus 4; compare Ambrose, Ep. 15.5-7 (PL 16.956),
thanks to S. Acholius's prayers, saevienti lue et ardenti perstilentia perturbati Gothi
ac territi sunt, at Thessalonica in A.D. 380.
 Notes to Pages 59-62 175

42. Athenag., Leg. 23; Origen in Rufinus, PG 12.789B; Ps.-Clem., Homil. 9.14;
Firm. Matern., De errore profan. relig. 13.4; Acta Xanthippae et Polyxenae 2 (mid-
third cent., says the·ed. M. R. James), the idols in a Spanish town tremble at the
sound of Jesus's name; Euseb., Praep. ev. 5.2 (182C); Basil, Comment. in Isaiam
10.236 (PG 30.532C); the same among pagan sources, Ps.-Apul., Asclepius 37;
Schermann (1909) 12; on statues moving, winking, etc., see for background Wein-
reich (1909) 145£., and examples of such miracles in POxy. 1242 III 51£., Dio
Chrysos., Or. 31.95, and Euseb., Praep. ev. 5.34.6f., on the statue of Theagenes in
Thasos; also Athenag., Leg. 26, on Nerullinus in Tralles and Peregrinus at Parium,
and Paus. 4.32.2, on Aithidas in Messenia. For pious kisses and touches perhaps
wearing away parts of statues, cf. Jucker (1960) 91, credible but inconclusive.
43. To the refs. in the preceding note, add testimonies totheworshipaccorded to
Plotinus, in Eunap., Vit. soph. 455, with Porph., Vita Plot. 22, where Apollo in an
oracle declares Plotinus after death to be a "daemon, though yesterday a mortal"
(lines llf. of the hymn). On Antinous, "his" miracles after his death can only mean
"through his statue," Orig., C. Gels. 3.36, Origen attributing the deeds to mere
magic. For the worship of a deceased person as "hero," cf. CIG 3936 (early second
cent.) and Ramsay ( 1895-1897) 2.603, a text of A.D. 245; ibid. 566, a text much im-
proved and discussed by Robert (1937) 132f., regarding a priestess heroized who
prophesies from her tomb.
44. Marc. Diac., Vita Porph. 60f., the Aphrodite statue at Gaza inhabited by a
demon; Euseb., H.E. 7.18.2f., at Caesarea.
45. The bulk of the evidence is epigraphic. I withhold most that I have gathered.
It seems to prove nothing. But there are useful discussions in Robert (l940a) 72,
Weinreich (1909) 7, Gaertringen (1935) 369£., and Nock (1925) 95f., all these on
Greek usages; on Latin, cf. ILS 3006-384 7 passim (apparently nothing of value in
secondary literature). I have tried to scatter illustrations generously through the
notes to the first chapter.
46. Art., Oneir. 4.2, "after seeing [=in a dream], make sacrifice and offer thanks";
Achilles Tatius 4.1.5£., Kat'lSvap or tvtinvtov; Marc. Diac., Vita Porph. 60, Kata
!CtA.euatv; ex monitu deorum, Ulp., De off. procons. 7.6, FIRA 2 2.580; imperiis
deum, Apul., Met. 11.29; ex imperio, Sen., Ep. 108.7 and De vita beata 26.8. In the
Levant, oddities: t~ tvKeA.euaE<O~ in Gaza, IGR 1.387 (the stone in Rome); "directed
in a dream," in Aramaic, Milik (1972) 388.
47. Herrmann and Polatkan (1969) 10, first cent., and, on "heroes," above, n.42;
and AE 1971 no. 208, Genio leg. VII Gem.fel.... ex iussu Geniivotum,secondcent.
48. IGR 1.568, ... Kata t7tttayf!v 6veipou, 1'1 ~ouA." ... ; 1.387; and other examples
to be sought in or through n.39, above.
49. Ep. ad Traianum 34, C. Muller, Geographici graeci minores 1.398£.
50. Orig., C. Gels. 1.66 and 2.60.
51. Bemand (1972) 133; Petron., Sat. 17.
52. Examples in Strabo 14.1.44; Gsell (1931)254and257,ona third-cent.African
inscription and its parallels throughout various western cults; and AE 1975 no. 722,
a Dacian inscription (restored) in which a debt to Liber is discharged: secundum
interpretationem sacrorum Solis in tabulam scribendum curavit.
53. Collart and Ducrey (1975) 239, comparing CIL 3.11137, i.d.f.
54. Moral. 414A, cf. Juv. 6.555, at Delphi, oracula cessant; and Moral. 574E.
55. Lucian, Cone. dear. 12, where, as Betz (1961) 58 says, Momus "complains
176 Notes to Pages 62-6J

about a sort of oracle-inflation" (and n.5, refs. to other Lucianic passages on oracles;
note esp. Alex. 29). For Amphilochus, cf. Lucian, Alex. 29 and Philops. 38, Dio
73.7.1 (indicating that "it foretells through dreams," an unusual method), and
Paus. 1.34.3, praising its veracity. For Claros, note esp. Porph., Ep. ad Anebonem
2.2d, Sodano pp. 9f., "far-famed and inspired" in his day, and below, p. 63. Euseb.,
Demonstr. ev. 5 praef. (PG 22.337), in declaring the oracles, indiscriminately, to
have been "quenched," is simply wrong. So tooisCumont(l929)285, for"thescep-
ticism regarding oracles" in the early Empire. He cites Strabo (who speaks obiter
and obscurely of his own views), Livy (skeptical of portents two or three generations
earlier), and Lucan 5.111 (who actually laments the silence of Delphic wisdom and
justice). Dodds (1965) 57 likewise errs in speaking of "the increasing demand for
oracles" in the second half of the third century. He adduces oracles quoted by Por-
phyry (but these predate the crisis, cf. above, chap. l.l n.65) and POxy. 1477 (but
that also is not relevant, cf. POxy. 2832f. andRE s.v. Astrampsachus col. 1797).
56. Euseb., Praep. ev. 4.2 (136a-b), 4.3 (136bf.), and 5.22 (213c), esp. citing Oeno-
maus's view that all oracles were human impostures, "the deceits and sleights of
wizard-men." Against the challenge that paganism was authenticated in its oracles
(e.g., Celsus, in Orig., C. Gels. 8.45), Christians could thus draw on a rich skeptical
literature. The Christian arguments appear compendiously in Passio SS. Carpi,
Papyli et Agathonicae I7f. (H. Musurillo, Acts of the Christian Martyrs p. 24), also
often in the Apologists, e.g., Lact., Div. inst. 2.17.

2. The Vitality of Paganism

1. Lucian, A lex. 38-and 44, where one of the school appears as active chal-
lenger. Chilton (1971) xxivf. finds Epicureans on the rise in the century, but Drach-
man (1922) 121 finds that "an essentially anti-religious school like that of the Epi-
cureans actually dies out at this time," the first half of the second century. The
evidence, either way, is scant.
2. The charge of atheism against the Christians was extremely serious, and
seriously repulsed. The evidence, e.g., Just., I Apol. 5f., or Athenag., Leg. 1, has
been often discussed. Harnack ( 1905) 11 shrewdly notes the counterattack made by
Apologists Greek and Latin through the pagan atheist Diagoras, as reproach against
all pagans. For atheism being allegedly illegal, cf. Dio Chrysos., Or. 43.11 and 75.5
(obscure)-most likely a matter of public scandal but not actual crime, as also in
Athens (Lucian, Demonax 11, with danger of stoning).
3. Euseb., H.E. 4.15.6; 4.16.3; and 5.1.9.
4. Athenag., Leg. I; Aelian, Varia hist. 2.31-"like Euhemerus or Diogenes or
Epicurus."
5. Lucian, ]up. trag. 53, "a few men" accept atheistical arguments.
6. Lucian, Timon 4; above, chap. l.l n.48; and Juv. 13.34(.
7. Tran Tam Tinh (1964) 108 and 115 n.2.
8. G. Picard (1954) 420-422.
9. Cumont (1929) 285, quotingTaramelli. Taramelli in fact ventures no indica-
tion of date for the inscription, and it is actually not of the second century. Cumont
goes on to claim, here adducing no facts at all, that oracles "in the 3rd century fell
under complete disbelief."
 Notes to Pages 63-66 177

10. Lact., Div. inst. 5.19; Porph., Ad Marc. 14, with more to the same, high moral
effect, ibid. 15£., 18, and 2!J.
11. Ovid, Fasti 6.249-252, stressed by Wagenvoort (1966) 965£., a scene to which
one may compare Plut., Moral. 270D, "just as today, still, men who have offered
their prayers and made their obeisances in shrines are accustomed to linger and sit
about," or Tibullus 1.3.27f. The second quoted passage is from Dio Chrysos., Or.
12.60f. It is noteworthy that Mars Len us, who should be a god of war, saved a wor·
shiper by his love: servatus Tychicus divino Martis amore, IG 14.2562=Buecheler
(1895-1897) 2.394 no. 850-explained by Drexel (1922) 25: the deity locally is a
healing one. In POxy. 528 a man prays for his wife to "Thoeris who loves you."
12. Sen., De superstit., in Aug., Civ. dei 6.10 passim. But, for a later audience,
also one meant to be more general, an Apologist presents identical views, ridiculing
winter acts of expiation (borrowed from Juvenal? above, chap. 2.1 n.l2) and the
fanatic who slashes his arms and body, and thus vulneribus suis supplicat, Min. Fel.
22.9, cf. Sen., I.e., vulneribus suis ac sanguine supplicant; compare Min. Fel. 23.1,
ridiculing lsiac lamentation for Osiris, with J uv. 6.534, presenting the priest derisor
of the plangens populus.
13. Most methodically, mythology is destroyed by Heraclitus, De incredibilibus
(first cent.?), in Mythographi graeci !J, 2 (1902) 80-82, indebted to Palaephatusof the
fourth cent. B.c., e.g., ibid. p. 6 echoed by Heraclitus, De incred. 2; on Pausanias, see
Casson (1974) 293; on Lucian, Drachmann (1922) 125; in the Latin tradition, d.
Cic., De nat. deor. 2.5 and Plin., N.H. 2.14-27.
14. Phrases quoted from Tran Tam Tinh (1964) 116 (whose references, Bayet and
Beaujeu, offer no footnotes); Bieler (1985-1936) 1.2 (whose references lead to literary
sources only) with Drachmann (1922) 121 offering identical views on second cent.
religiosity; Dodds (1965) 9, actually seeing "a wave of pessimism sweeping over the
West" (p. 18) in a period of much broader boundaries-for criticism, d. MacMullen
(1976) 14-16 and notes; and Nock (1937) 112, referring to "the anxious questing
mood of the time" when he evidently means the second to later fourth centuries (in
the context, adducing a second-cent. inscription, Artemidorus, Heliodorus, Julian
and the SHA); finally, Geffcken (1920) 20.
15. Dodds (1965) 3, a statement challenged by Gordon (1972) 94.
16. Apul., Met. ad fin., well-known chapters; Aelian, frg. 89, in R. Hercher,
Aeliani opera 2.230f., a highly colored little tale, pro-Asclepius, anti·Epicurus, very
closely recalling the sort of aretai that Pausanias read at Sicyon (above, chap. J.2 at
n. 70), and full of conviction. ·
17. Chap. 1.2 n.44 (Xenophon), chap. 1.3 n.29, and chap. 1.2 n.72.
18. Nock (1933) 112.
19. Liebeschuetz (1979) 214.
20. Oribasius 45.30.11-14; Report of the XXll Congress (1962) quoted by Bicker-
man (1973) 5 n.2. Lazurkina had known Lenin personally. Lenin spoke against
Stalin, on this occasion. She explained to my friend V. Rudich that she often com·
municated with Lenin in dreams.
21. In all regions: CIL 10.1560, servitor deorum ex viso; ILS 3513, publicus . ..
pontificum, his sight restored; CIL 9.3146, ministra; aedituiin CIL 12.2215 and AE
1971, 31; a ~aKopoc; cp{A.toc; of the god, IG 22.4514; tepeuc; the normal office, e.g., IGR
1.106, Katd dA.t:ucnv, or in M. Dunand (1939) 562; also vero1C6poc;, IGR 1.107=/G
178 Notes to Pages 66-69

14.1030; in Latin, normally sacerdotes, e.g., ILS 3831, ex responso; Albertini (1943)
377-380; LeGlay (1971) 126; C/L 3.7728 and 6.659.
22. /G 42, 1.380-430 (notice 386 and 415, "by dreams," "by command," A.D. 148
and 258).
23. CIL 6.31181, late Antonine, according to Speidel (1978) 18; CIL 3.7760,
priests plural, "by command of the god," in Merlat (1951) 20, A.D. 202; and, in
general, I. Toth (1971) 24.
24. Below, chap. 2.3 nn.65-67.
25. Marcus Aurelius, Medit. 1.6 but d. 1.17.20; Art., Oneir. 4.22, p. 257 Pack, on
Fronto; Max. Tyr., Philos. 9. 7; the often-cited example of Galen, in his Sub fig.
empirica 10, K. Deichgriiber, Die griechische Empirikerschule2 ( 1965) 78£.; also in
De libris propriis 2, I. Mueller, Galeni scripta minora 2.99, and In Hippocr. de
humor. p. 222 Kiihn; compare another famous physician acknowledging his debt to
Asclepius, in Robert (1973) 184. For belief in divine healing, see further Cic., Ad
jam. 14.7.1, crediting Apollo or Asclepius; Just., I A pol. 54.10, d. Athenag., Leg. 23
(but it is really demons, he says, not a god); distinguished devotees of Asclepius, like
the rhetors Antiochus, in Philostr., Vit. Soph. 568, and Polemo, ibid. 535 and AE
1933, 275, or the consul of A.D. 147, d. Habicht (1959-1960) 112; d. also notable
votaries of the fraudulent Alexander of Abonuteichus, in Stein (1925) 259f.
26. Above, chap. 2.1 nn.45 and 71; in chap. 1.1 nn.48, 65, 67, and 72.
27. Above, n.5.
28. In Palestine, the inscription, "Isis the true!" in AE 1948 no. 145, or Robert
(1955) 88, at Ephesus, the inscription consisting solely of the words, "Great is the
name of the god! Great is the Holy! Great is the Good! Obedient to a dream." Cf.
POxy. 1382 (second cent.), "as you stand there [in the Sara pis temple], say, 'One is
Zeus Sarapis,"' discussed by E. Peterson (1926) 217£.; Marc. Diac., Vita Porph. 31:
struck by a miracle, bystanders cry out, "Great is the god of the Christians! Great is
the priest Porphyry!"; above, chap. 1.1 nn.70-72, 75,and 79; chap. 1.4 n.10; and
below, chap. 2.3 n.54.
29. /G 4, !2.131 col. II (Epidaurus, third cent.) is the only instance I can recall
from inscriptions, of a story being told, however briefly:" ... to the Mother of the
Gods: Maidens of memory, come down from heaven and, with me, hymn the
Mother of the Gods. As she came wandering through the mountains ... "(Zeus spied
her, etc.).
30. In Pack (1965) nos. 455, Favorinus, On Exile; 1325, Musonius Rufus on
infanticide; 2583, anon. second cent. frg. of only a dozen words containing some
speculation about the gods (surely not to be asserted as a treatise On the Gods); a few
scraps of aretalogy, some (as Pack indicates) perhaps rather from romances (add
now West [1971] 96, on no. 2468); a third cent. B.c. Orphic text, no. 2464; some
hymns, nos. 1933-1936 and 2481; thus in total, from over 3,000 items, we have the
items just cited plus a scattering generally in nos. 2466-2494. For the popularity of
individual authors, see above, chap. 1.1 n.41.
31. Moral. 263E-29IC, 113 customs, the vast majority concerning religion: first,
266F (or 271F, etc.); second, 272D (or 275A, etc.); third, 274A£. (or 266C, etc.); fourth,
267D (or 271D, etc.); fifth, 279Cf. (or 281D, etc.); and sixth, 278Ff. (or 287B, etc.)-
two or more examples for each mode of argument. The essay on Isiacism uses the
same categories. For the first, d. Moral. 356D or 359B; second, 357F, d. also 675F-
 Notes to Pages 70-71 179

676B, favorite line of argument for Plutarch; third, S75A; fourth, 358D; fifth, 362Af.;
and sixth, S62Ef. In the first century, Cornutus, Theol. gr. compendium, makes
abundant use of the first and second devices, e.g., in chapters 5, 6, 28, and 30.
S2. A difficult text. I hope I present it correctly. See Merlat (1951) 38S on GIL
S.7756, with suggestions by von Domaszewski and Hirschfeld as well.
SS. Above, chap. 1.1 nn.S2-37.
34. Timoleon 26.1£; similar incidents similarly described, Pericles 6.1 and Ser-
torius 11.7-12.1, examples cited by Erbse (1952) 297; c£. Koets (1929) 59 on supersti-
tion and a solar eclipse, Nicias 23; ibid. 74 on Cleomenes 39; ibid. 32, the word
superstition in a bad sense "explained by the everwidening chasm that separated the
educated elite from the masses." In Lucian also, Phi lops. 9; but Latin illustrations
are hard to find in our period. Otto (1909) 532-544 is superficial and looks only at
earlier authors.
35. For a survey of testimonies to magic, see MacMullen (1966a) 100-108. The
increase in superstition is generally sensed by modern students. P.R. L. Brown
(1972) 122 doubts it. "Thus it is far from certain that there was any absolute increase
in fear of sorcery or in sorcery practices in the Late Roman period''-but he offers
nothing in support of his "thus." Further, Brown (1971) 82dismisses the older, and
"too often" the modern, view that the increase arose from the "democratization" of
culture and "dilution of the ideas of an enlightened minority ... intrusion into the
upper classes" of lower class ideas. "To take issue with such a view would involve
rewriting the social and religious history of the Roman world." This, he does not
attempt. For the moment, we must be content with conventionally documented
interpretations. For lead defixiones, besides the standard but old corpus of Audol-
lent (1904), see more recently examples inMouterde(1930-1931) 110£. and 118, from
Beirut and Damascus; Wortmann (1968) 60, from Oxyrhynchus; Elderkin (1937)
384, from Athens; and D. M. Robinson (1953) 172, from Macedonia and Pisidian
Antioch. For orators hexed, see Cic., Brut. 60.217, the consul of 76 B.c., and Liban.,
Or. 36, in Festugiere (1959) 102 and 453£. Emperors resort to sorcery, MacMullen
(l966a) 104 and further exploration in idem (1968) 82-95.
36. The summary of an unpublished text, POxy. 3008, in POxy. vol. 42 (1973) 30
n.l. Inquiry of one of the editors of the series, for more information, went unan-
swered. For the onset of the decline, one may choose (of course, somewhat arbi·
trarily) Sextus Empiricus, cf. MacMullen (1966a) 109-111 and idem (1972) 12-15;
for the course of the decline, briefly, Nock (1926) xxiiif.
37. Ps.-Ciem., Homil. 1.3 (PG2.57), thefourthcent.editionofa third cent. docu-
ment; dramatic date under Tiberius, 1.6 (PG 2.61 ); d. the similar picture in Rufinus,
Vita S. Eugeniae 3 (PL 22.1108£.), the supposed date being Severan; Tert., Apol.
47.5-7 (all schools invalidate each other) and 47.9, condemning mere cleverness,
ingenia, cf. the contemporary texts in Min. Fel. 5.3 and Parassoglou (1976) 269,
"dangerous 1t£p18pyia" banned by the emperor in A.D. 199. Foreshadowings of the
mutual destruction of the schools are seen in passages of Plutarch, c£. Cherniss
(1976) 371; and the condemnation of abstract thought-curiositas-by Christians
can be traced from Tertullian on, in Labhardt (1960) 216-218. Add CT9.16.4 (A.D.
357).
38. Const., Or. ad sanctos 9 (PG 20.1253 Bf. and 1256Af.). It is only for brevity's
sake that I call these words Constantine's, without a question mark. They represent
180 Notes to Pages 71-75

in any case what could be asserted in the period with unembarrassed confidence. For
a good discussion of the integrity of the source, see Baynes (1931) 48-54. For similar
passages a couple of generations earlier, see MacMullen (1972) 13; later, many more.
39. The change in the status of philosophy and rational thought was so gradual,
it had still to prevail universally as late as Julian's reign. Liban., Or. 16.47, com·
paring Plato and Pythagoras with the (Christian) wisdom of women and menials,
and doing this before an open audience, proposes views that could have been
written by Celsus, nearly two centuries earlier. Cf. Orig., C. Cels. 3.44. And there are
figures like Synesius still to come.
40. Rehm (1958) 299-301 no. 496, discussed by Robert ( 1960a) 544f., the fragmen·
tary reply indicating it was right cult that induced favor; at Stratonicea, Laumonier
(1958) 275; at Cyrene, Robert (1940b) II£., the indicative epithet for the priest being
KaA.A.t&tT]~. "who brought a good year"; idem (1946) 142£., the same epithet at Olbia
and Rhodes and Lindos, that last yielding an inscription that Blinkenberg (1941)
882 dates to the second cent., Robert to the third.
41. Dio Chrysos., Or. 1.54, near Olympia.
42. Lucian, Dea Syria 28.
43. MacMullen (1976) 27 nn.14f.
44. Plin., Paneg. 31£.; d. 80.4, Jupiter rules the world, barely condescending to
lower his eyes to it.

3. How the Divine World Was Envisioned

I. Plut., Moral. 165C, 168A-C, 170E, d. (in other essays) 26B and HOlD, offer·
ing (166Df.) the "true" view of the benevolent gods; and deisidaemonia can be
avoided by an approach through philosophy (355C, in the essay on Isiacism). For
modern assessment of these views of Plutarch, d. Wey (1957) 141 n.444 and Koets
(1929) 58£. and 68-80. For the connection between fear (of angry gods) and super·
stition in other authors, d. Lucian, Pro imag. 7; Clem. Alex., Strom. 2.8.61 (PC
8.976) and 7.1.4 (PG 9.408A) and Coh. ad gent. 3.78 (PG 8.129), the rationalizin~·
explanation (quoted in the text) lying in Strom. 7.4.20-25 (PG 9.429); similarly, in
Ep. ad Diognetum l.494B. Among Latin authors, clearly drawing on Greek, d.
Cic., De nat. deor. l.ll7; Varro, in Aug., Civ. dei 6.9; Sen., Ep. 123.6; Lucr. 5.1161£.
and other poets, in Heuten (1937) 3-5 and Dionigi ( 1976) 120. On superstitio in later
(second-fourth cent.) texts, cf. Martroye (1915) 287-290; idem (1916) 106£.; and
Momigliano (1977) 144-147.
2. References in order of citation: Moral. 369Af. and 417C, god is good; 23D;
24B; 20E-F; 996C and 355B (d. 358E); 414E and 418E (cf. 381E), god must maintain
remoteness. Terms like to 1tp60'1popov (374E) and td 1tp&1tovta (383A) are opposed
to to pappaptJCcimpov (418E, 358F), in describing divinity.
3. God is goodness, Dio Chrysos., Or. 32.15, cf. Paus. 8.36.5; beautiful, 12.59, cf.
Celsus in Orig., C. Cels. 6. 75, and Porph., De abst. 2.38, in Euseb., Praep. ev. 4.22
(172a); the allied reasoning, a form beautiful must be divine, a cliche, in, e.g.,
Lucian, Alex. 3 and Xen., Ephes. 1.2.7; gods delight in gifts from the good, Dio
Chrysos., Or. 1.16 and 3.52; but need no statues or sacrifices, 31.15.
4. De sacr. If. and 4 (the poets misrepresent the gods as bribable), Demonax 11
and ]up. conf. 7; remoteness of deity, Max. Tyr. in Merlan (1967) 81, opposed by
 Notes to Pages 75-76 181

Marc. Aurel. 2.11; for Lucian as hostile to anthropomorphizing in general, see Betz
( 1961) 27, and compare Hephaestus in ]up. conf. 8; ibid. on manacles, Prometheus,
wounds, shackles, and idols melted, cf. Prometheus 1-these various indignities
followed out in many dialogues, e.g., lcaromenippus or ]up. trag. Indignities to
idols, ]up. trag. 7£. and Gall. 24; Egyptian animal gods, in Concil. deor. 10£. and De
sacr. 14f.
5. Orig., C. Cels.l.5: it is~Tj eill.oyov ... , saysCelsus; thesamewordcommonin
this work, and others like it, e.g., 1.32, eiiA.oyov, euA.oyrotepov; ~A.t9trotatov and
t~~povrrjtov (5.39), dto1tOV (8.49); in Justin, dl.oyov (1 Apol. 9); in Theoph., Ad
Autol. 3.7, 3ewov JCai d9e~ttov; in Athenagoras, A.fjpov JCai yel.rota (Leg. 21) or
d7ti9avov (20); in Aristides' Apology, yeA.oia JCai ~ropa Kai dae~l'j (8. 4) or dvow (9.5 );
in Acta S. Apollonii (ca. A.D. 185) 17, ti~ 6 A.Tjpo~ tl'j~ d1tat3euaia~; and pagans
accuse Christians of ~ropia, d1tovoia, insania, Acta S. Cononi 4.6, Acta SS. Agapae,
Eirenae et Chionae 6.2, and Passio S.lrenaei 3.4. For Christian writers drawing on
pagan critics of paganism, for use against pagans, see chap. 2.1 n.56 and chap. 2.2
n.l2; and below, n.9. Koets (1929) 39, use of Menander and Bion by Clem. and
Theodoret (also Justin, d. I A pol. 20); A pol. 12.6, using Seneca; Lact., Div. inst.
4.28, using Cic., De nat. deor. 2.28; for Xenophanes apparently underlying various
writers, cf. Raabe (1893) in his notes passim, and Cardauns (1958) 53; for Plato used,
too, d. ibid. and Euseb., Praep. ev. 2.8 (76cf.); the debt to philosophers, e.g., to
Plato, is stressed by Geffcken (1907) xvii and Chadwick (1965) x£. For the date of
Heraclitus, see Buffiere (1962) ix.
6. Philo, De plant. 130; lren., C. haer. 3.25.3 (PG 7.968); Iambl., De myst. 1.5;
Porphyry in Prod., In Tim. p. 208 Diehl; Apul., De Plat. 12, nee ullius mali causa
deo poterit adscribi, the text accepted as genuine by Witt (1979) 384; Plot., Enn. 2.9.1;
and to 3at~6vtOv "most benevolent," in Dexip. frg. 23, FHG 3.680 (note the setting).
7. Ps.-Apul., De deo Socr. 3.123 and 7.137; De mundo 397b and 398a-b; Apul.,
De Plat. 1, ultramondanus, incorporeus; Iambl., De myst. 1.5 and 1.20, even above
rational knowledge, cf. Celsus in Merlan (1967) 81 and Andresen (1955) 159-162;
Sext. Emp., Adv. dogm. 3.33, "all men have thesameassumptionincommonabout
god, that he is a being blessed, indestructip1e and complete in happiness"; and,
broadly, Geffcken (1907) 171 with refs. For the view that god takes thought only
about himself, as the Stoics say, see Theoph., Ad Autol. 2.3, strongly opposed by
Just., I Apol. 28 and Marc. Aurel. 2.11.
8. Ps.-Apul., De deo Socr. 12.145£.; Porph., De abst. 2.60; Philostr., Ep. Apollon.
26, the gods need no sacrifices, d. other passages in Windisch (1920) 310f., Des
Places (1969) 343, and Andresen (1955) 63; Sext. Emp., Pyrr. hyp. 162, dna9t~ elvat
to 9etov; Clem. Alex., Excerpta Theodoti 30f., and Strom. 2.18 (PG 8.1020B), dvev-
3et~ ... to 9eiov JCai d1ta9e~; and Diog. Laert. 1.5 in the third cent., indignant at the
idea of attributing to the gods 1tdv to dv9pronetov mi9o~ (early third cent.).
9. Shocked by Homer's tales, Plut., Moral. 20E, and Dio Chrysos., Or. 11.19,
with more texts and discussion in Dreyer (1970) 45-47; the gods change, Arist.,
A pol. pp. 8f. Raabe, and Tat., Ad Graec. 21; but should not change, Ps.-Apul., De
deo Socr. 12.146; for the fleshly indignities in which the gods are involved, see
Heraclitus, Quaest. homericae 26.1; 30.1; and 69.1; Celsus in Orig., C. Cels. 5.14;
Philodemus, Piet. 18.13£., 47.14, with other passages from that author in Geffcken
(1907) xviiif., or, more thoroughly, pp. 11, 16-19,39£.,46-48, 52£., 62-64, and 132ed.
182 Notes to Pages 76-77

Gomperz, Philodemus being a writer of the first century B.c.; Varro, in Cardauns
(1958) 53f.=Aug., Civ. dei 6.5; and Sext. Emp., Adv. dogm. 3.147. Itis even seen as
beneath their dignity that the gods should marry, d. Seneca, above, chap. 1.1 at
n.67, and Plio., N.H. 2.5.17. All of these notions being in common circulation, the
Apologists make great capital of them, e.g. Athenagoras, Leg. 21£.; Theoph., Ad
Autol. 1.9 and 3.8; Arist., Apol. 8.4, 9.5, and 10£.; Iren., C. haer. 1.4.4 (PG 7.485A-B)
and 3.52.4; Clem. Alex., Coh. ad gent. 2.164 (PG 8.105); Tat., Ad Graec. 8; and
among later writers, Min. Fel. 23.5; Lact., Div. inst. 1.16; Firm. Matern., De errore
profan. relig. 2.1; Athanas., C. gent. 11£.; and Geffcken (1907) 204.
10. Isoc., Busiris 38, much pillaged by later writers; Sext. Emp., Pyrr. hyp. 159,
and Adv. dogm. 3.193 (drawing on Xenophanes); Calcidius, Comment. 128; and
Philostr., Vita Apollon. 5.14. Among Christian writers, see Clem. Alex., Coh. ad
gent. 4.58.3 and 4.60.1£.; Theoph., Ad Autol. 1.9 and 2.8; Athenag., Leg. 20 and 30;
Athanas., C. gent. 12 and 15; among Latin Apologists, Tert., Apol. 11.12, Min. Fel.
25.2£., and Lact., Div. inst. 1.9-11.
11. On divine grandeur: Sen., Ep. 41.4f., to be added to many similar texts in
Bieler (1935-1936) 1.11 and passim. On menial roles vel sim., Philodemus, Piet. 34;
Sext. Emp., Pyrr. hyp. 157; Cic., De nat. deor. 1.38; Porph., Philos. ex orac., in
Euseb., Praep. ev. 5.9 (196bf.) and Ep. ad Anebo. 28 in Euseb., Praep. ev. 5.10
(197df.); Ps.-Aristot., De mundo 6. Among Christians, note Just., I Apol. 9; Arist.,
Apol. 10 p. 270 Raabe; Clem. Alex., Coh. ad gent. 2.199 (PG 8.113); Lact., Div. inst.
1.9; and Tert., Apol. 14.4.
12. Gods misshapen or freaks, hermaphrodite: Sen., in Aug., Civ. dei 6.10; Min.
Fel. 23.5, d. Athanas., C. gent. 22; Athenag., Leg. 20f., on deities with many breasts,
one hundred hands, four eyes, etc.; as animals, in anti-Cleopatran passages of
Horace, Propertius and Vergil, e.g., Aen. 8.698; Stat., Silv. 3.2.113; Juv. 15.1-8; and
Apologists de rigueur.
13. Hdt. 2.172 underlies Acta S. Apollonii 17, perhaps Just. I Apol. 9, dttJ.1WV
OK&urov, and Min. Fel. 22.4. For pagan ideas about idols of materia vilissima, d.
Sen., in Aug., Civ. dei 6.10; Varro, ibid. 4.31; Oenomaus, in Euseb., Praep. ev. 5.36;
mice and birds at work on idols, Geffcken (1907) xx and Iren., C. haer. 3.52.4. For
many other passages, see Ep. ad Diognetum 2.3£. and n. ad loc., in the edition of
H. I. Marrou (Sources chretiennes 33, 1951) 106 n.2.
14. Marrou 110 gives relevant passages from the Apologists; d. Lucian, Desacr.
9; on human sacrifice, Tert., Apol. 9.2 and Euseb., Praep. ev. 4.17 (164c-d). A. Alfoldi
(1963) 33 n.7 cites many passages supporting his view that the practice was a fact in
early Rome. SHA Aurelian 20.7, cuiuslibet gentis captos, must indicate human
sacrifice in late Rome, in the view of the source.
15. The oracle, above, chap. 1.1 n.65; in Pack (1965) I count the Homer-papyri
(not commentaries) dated second cent., third, third/fourth (but not first/second), in
total over 330. In the Apologists, it is easier to find harsh scorn of Greek philoso·
phers ("Constantinian" by anticipation, e.g., Tat., Ad Graec. 2f., or Theoph., Ad
A utol. 2. 4£.) than respect for them; or the competence of the audience in philosophy
is self-consciously emphasized, by Just., I Apol. 14£., repeated address to ~paatai
1ta1l)&i~ and "philosophers," or Athenag., Leg. praef. This is not to call the Apolo-
gists philosophers themselves. Cf. Bardy (1948) 11 and 22£. on Theophilus's ignor-
ance and ineptitude.
 Notes to Pages 78-79 183

16. Menander, De hymn. 2 and 5, in Spengel, Rhet. graeci 3.333 and 337; hymns
in Porph., De imag., in Euseb., Praep. ev. 3.9 (lOOa-d)=Stob., Eel. 1.2.23, which I
quote in the text; Orig., C. Cels. 1.18; initiations reinterpreted, the Mithraic, accord-
ing to Pallas, in Porph., De ab.st. 4.16; initiations in general, Proclus, In Plat.
Rem pub l. ed. Kroll2.1 08, quoted by Pepin (1966) 262, and PI ut., De Daed. Plataeen-
.sibu.s, in Euseb., Praep. ev. 3.1 (83d.); Phrygian and others, in Hippolyt., Ref. 5.2-4,
interpreted by the Naassenes; statues, Plut., Moral. 381Df.; rites, ibid. 368D.
17. Menander 5 p. 336 Spengel. On Hera=air, d. Porph., De imag., in Euseb.,
Praep. ev. 3.11 (108b), with more of the same for Demeter or other gods, 109£., e.g.
(113 b), "Cerberus has three heads because the upper positions ofthe sun are three-
rising, noon and setting"; Heraclitus, Quae.st. Homericae 15.3, Hera=air; similar
equivalences in Philodemus, Piet. 11-22, in Biicheler (1865) 530f.
18. Sallust, De dii.s et mundo 4; Paus. 8.8.3.
19. I leave aside allegorical interpretation in Philo, Clement, Origen, and others;
a large subject. Within paganism, it had its roots deepest in Stoicism, d. Decharme
(1904) 305-353 passim; Tate(l929) 40-43 and(l930) 1-7; Buffierexxxviiif.; and some
applications of the methods are touched on above, chap. 2.2 n.31. On the division of
theology into three strands of development, d. Strabo 10.3.23 and Porph., in Euseb.,
Praep. ev. 4.1 (130a-d); and, in the Latin tradition, Tert., Ad nat. 2.1, going back to
Varro, d. Aug., Civ. dei4.27 (and 6.5), in turn pointing toScaevola, cos. 95 B.c., cf.
Cardauns (1958) 36-39. On rejection of the art, d. Sen., De benef. 1.3.9f., making
harsh fun of it; also Dio Chrys., Or. 11.17f., and Philo ofByblos, his contemporary,
who expressed suspicion of recent intemperate innovating through allegorizing in
initiatory mysteries. Cf. Troiani (1974) 47f., on frg. I, 26; ibid. 13 and Barr(1974) 16,
on Philo's dates; and compare Plut., Moral. 19Ef. Contra, in support of allegoriz-
ing, d. Cornutus, Theol. gr. compendium, passim, Dreyer (1970) 67, and Plut.,
Moral. 355Bf., 366E, 368D, 374E, 381, 996C; his commentary on Hesiod, Moral.,
Loeb ed. vol. 15, e.g., pp. 115 and 119; the art of allegorizing employed by Egyptian
hierophants or theologues, cf. Merkelbach (1962) 57£.; Orig., C. Cels. 1.12; Corp.
herm. 13.1; perhaps Ps.-Clem., Homil. 6.2.1, where the spokesman for Hellenism
shows familiarity with the uses of allegory to hide great truths "from the unworthy
and careless," cf. 4.24 (PG 2.173). Lucian, Cone. deor. 11, shows Zeus defending
Egyptian monster-gods as allegories. On the chief practitioner of the art, Porphyry,
d. Bidez (1913) 32 and Pepin (1966) 261£., where Porphyry's aside, concerning alle-
gorizing for the masses, is stressed (De antro 3 p. 56, d. 36 p. 81).
20. Heraclitus, Quae.st. Homericae 69.8.
21. Plato's demons: Arnim (1921) 62-67; Andresen (1955) 99; Wey (1957) 144;
Chadwick (1965) xviif.; Merlan (1967) 34£.; Chadwick (1967) 165; and Den Boeft
(1977) 3-15, 42, and 56.
22. Merlan (1967) 35 (Plotinus), 60 (Plutarch), and 72 (Apuleius); Brenk (1973)
2-8, on Plutarch, with such passages as Moral. 361Af.,415Af., and944Cf. (and Stoic
views in 277A); Apul., De Plat. 11; Celsus's views in Andresen (1955) 62, 99, and 103;
those of the Ps.-Plut., De fato=Moral. 572f., of the early second cent., d. DeLacy and
Einarson in the Loeb edition (1959) vol. 7 pp. 303f.; Artemidorus's views in Oneir.
2.34, distinguishing the ethereal powers above the planetary, who are above the
powers resident on earth, in oceans, etc., each of the three types divisible intovontoi
or ala9t'Jtoi; for Iamblichus's views, see De my.st. 1.3-9, closely similar to the fore-
184 Notes to Pages 79-82

going, and at 2.3 distinguishing angels, archangels, demons, rulers, and spirits; at
5.}4, also distinguishing the "material" from the "nonmaterial" gods. Less ordered
is Porph., De abst. 2.37, and still less (with casual indication of the God, and be-
neath, gods plural, and then demons) the author of Corp. herm. 16.17. Euseb.,
Praep. ev. 4.5 (14lb), summarizes paganism in general, ~ 'EA.A.TJVtiCl\ OeoA.oy{a:
topmost, Divinity; next, gods; next, demons; next, heroes; next, evil demons, with
Dio Chrys., Or. 3.54, closely approaching this schema of beings. For the seven
spheres of Mithraism, alleged, see Cumont (1915b) 163£. or Gordon (1972) 97, com-
paring St. Paul, 2 Cor. 12.2, and Valentinianism. On the latter, c£. Sagnard (1948) 12
and 25-28 (on the Demiurge in the Seventh Heaven, the Savior in the Eighth, ibid.
27); Clem. Alex., Excerpta Theodoti 14 and passim, and Strom. 5.6. (PG 9.61Bf.)
and 7.10 (PG 9.481A); Iren., Demonstration of Apostolic Prediction 9, trans. L. M.
Froidevaux (1959) 44, seven heavens, but ten according to Marcus, Valentinus's
disciple, or 365 according to Basilides, c£. idem, C. haer. 1.5.2 (PG 7.493B), 1.7 and
1.24.5 (PG 7.637A and 678B). For a parody, d. ibid. 1.11.4 (PG 7.567=Epiphanius,
Panarion 32.6. 7).
23. Homer: see the usage through any index verborum, but also Plut., Moral.
415A; usage in a specific literary genre, hymns, e.g., Orphei hymni 2 and passim, c£.
the Index in Quandt's edition ofthe hymns; Heitsch (1963) 1.166 and 172, third cent.
hymns to Sarapis and Tyche; Paus. 1.2.5; Celsus in Orig., C. Cels. 7.68; perhapsDio
Chrys., Or. 33.4, emended by Emperius; but pagans sometimes sharply distin-
guished between gods and demons, e.g., Plut., Moral. 418E, and Clem. Alex., Coh.
ad gent. 2.38 (PG 8.121). The distinction is normal in the Apologists and other
Christian writers, c£. the refs. in MacMullen (1968) 93 n.42, though concession is
made through a special rank of Great Demons (or one word, ,.u:yai..o~ai~OVS<;, not in
LSJ, but in Stephanus) in Clem. Alex., Coh. ad gent. 2.41.2 (PG8.124, Apollo, even
"Zeus himself"), and Euseb., De laud. Canst. 9.4 (PG 20.13648, Apollo, etc.).
24. Dio Chrysos., Or. 1.39-40 and 12.22, for illustration of Zeus basileus; of
queens, d., e.g., AE 1943, 243 (Isis) and IGLS 2964 (Juno); compare Hadrian's epi-
gram to accompany a dedication to Zeus Casius, ~eo{pavo~ dvOpro1trov Kotpcivcp dOa-
vcitrov, in Suidas s.v. Kciotov c5po~; Steinleitner (1913) 77, for Men (and Rhea Cybele
as queen); Plut., Moral. 353B and 354F, Osiris basileus; Attis, in Chadwick (1952)
90-92, second cent.; Helios, in Ramsay (1895-1897) 1.308, A.n.l74/5. I omit the king
of quite a separate realm, Pluto (e.g., Plutoni regi magna in ILS 4454).
25. Nock (1925) 97, in part drawing on Cumont (1920) 183£. (better, 185); but
Cumont erred in supposing the Romans are imitated. The practice begins much
earlier, and with other models. Cf. Seyrig (1970) 77£. and 83, offering the general
danger of real life as the reason for arming gods who had nothing military about
them-unarmed was weak. Further, idem (1971a) 88 and 95, cuirassed Palmyrene
gods. For the Doliche relief, d. Merlat (1951) Sf. Fora broad discussion, d. Kantoro-
wicz (1961) 369-390, e.g., 379 (Mithra), 382 (coin styles generally), and 384 (Christ).
26. Tudor (1969-1976) 2.81 (date, post-A.D. 150), Ill (reliefs show uniforms), and
126 (gods with dragon-banners); at Siliana, Saturn and the Dioscuri, in Charles-
Picard (1943-1945) 378, and idem (1954) 120f. and Fig. II.
27. CIL 13.7399 and 8523f., Sol invi~tus imp.; 13.7571a, [im]p. noster; c£. 7815,
deo invict. regi (all these from Germany); and so to Julian, Or. 4 to King Helios,
e.g., 132£.; CIL 8.18219, lovi . .. deorum principi gubernatori omnium terrarumque
 Notes to Pages 82-84 185

rectori, Lambaesis in Tetrarchic times, cf. Prosop. of the Later Rom. Empire s.v.
Flavius no. 21.
28. Iren., C. haer. 1.2.6 (PG 7.465) and 2.7.4 (PG 7.730A); c£. Tert., Apol. 24.4,on
the procurantes et praefecti et praesides of God, and the Orphic verses cited by
Clement, "around your flaming throne stand the angels, rendering much service,
who see to it that all is performed for mortals," the passage brought to notice by
Cumont (1915b) 173; Ael. Arist., Or. 43.18, quoted in Chadwick (1965) xix, and
Celsus, in Orig., C. Cels. 8.35, divine "satraps and ministers" (d. also 8.28); Euseb.,
De laud. Canst. 1.2 (PG 20.1320A) and 3.6 (PG20.1332A), God as King, so hailed by
Constantine's troops; cf. Melito, De bapt. frg. 4, ed. 0. Perler(l966)232; PGM no. 3
lines 555f.; by Origen (not the Christian) in Porph., Vita Plot. 3.33, the treatise titled
"The King Alone is Creator"; Plot., Enn. 3.56; and the Ps.-Arist., De mundo398af.,
with date A.D. 100, in Lesky (1966) 574, and discussion of the hierarchy of powers in
Maguire (1939) 150 and 155£.
29. Demons used for vengeance, in Plut., Moral. 277A, and Iambl., De myst. 2.7,
cf. Origen agreeing, C. Cels. 8.3; used for magic, PGM passim, or Hopfner (1922)
243; used for presages and oracles, Apul., De deo Socr. 4.128-and 6.132; Schermann
(1909) 12, magical papyri; and Plut., Moral. 418Df.
30. Ps.-Piut., De fato=Moral. 573A; Apul., De deo Socr. 6.13.2; cf. deities' over-
sight of specific functions in Plut., Moral. 377A and 758A ..
31. Manilius, Astr. 4.696-700, each gens assigned by god to its own star. Note also
Clem. Alex., Strom. 6.17 .82£. (PG 9.389) and 7.2.52f. (PG 9.409) and Coh. ad gent. 2
(PG 8.121Bf.), saying that each people has its own angel but also its own pagan
demon, the closeness of the pagan and Christian views emerging likewise in Orig.,
C. Cels. 1.24 and 5.25, where Origen and Celsus both believe a special supernatural
being (and both use the word demon) is in charge of each people; again, in Min. Fel.
6, dei municipes and ritus gentiles, according to the pagan spokesman; each god
hears us best in some certain place, according to Marcus Aurelius, and he addresses
"all the dei praesides o£ all populi," Ep. ad Front. p. 47 Naber; Iambl., De myst.
5.25; and Symm., Rel. 3.8.
32. Barb (1963) 104, speaking of belief in spirits and demons of the period from
Domitian to Ammianus (103), which in the fourth century "grew to mountainous
height." He uses more highly colored wording, p. 108. For'the abundant evidence of
magic and superstition (let us call it that, loosely) in the same period, cf. MacMullen
(1966a) chap. 3. As a specially relevant example of free borrowing by magic from
regular cults, notice "liturgical acclamations as magical formulae," like "One is
Zeus Sarapis," in E. Peterson (1926) 232.
33. Above, n.27; Maximin Daia's letter to Antioch, in Euseb., H.E. 9.7.7, Zeus
credited as tou~ 1tatpq\ou~ u~iilv 9£o~ ... pu6~£VO~. Cf. Celsus, in Orig., C. Cels. 8.66
(perhaps only for the sake of argument?), imagining a great god, ~ty~ 9£~, set over
Helios or Athena; or compare Aoth, "before whom every god does obeisance and
every demon shudders," a late text (A.D. 300-350), PGM 12 lines 115f.
34. Apul., Met. 9.14; for the inscription in which the erasure was detected by
Cumont, see Vidman (1965) 391; cf. Acta Pionii 19.11, Zeus maker of all creation.
35. CIL 8.12493; cf. IGR 1.144=/G 14.1084, A.D. 146, from Rome, the same triad,
and again in Robert (1958) 32 in the second cent. and at Stratonicea in the third
cent., Laumonier (1958) 289; in Latin, examples in Malaise (1972) 195 n.I3; and a
186 Notes to Pages 84-85

general collection in Vidman (1965) 390, the earliest known being of A.D. 142.
36. Arrian, speaking of broad practices of religion in the province, in FGr.H.
156F22; Ohlemutz (1940) 83, an inscription of the second cent. with far earlier ante-
cedents; E. Peterson (1926) 228-231, amulet-inscriptions, the formula itself going
back (p. 236) to Flavian times; MAMA 6.360 and Malaise (1972) 196.
37. Cumont (l923a) 67-69, citing Arrian on Bithynia, a Phrygian inscription,
and an incident in Aurel. Viet. 39.13; also two inscriptions from Rome, p. 66. Cf.
Helios Apollo in a Milesian inscription, Rehm (1958) no. 504, Tetrarchic, and the
identification Heliodorus draws at the end of the Aethiopica {late Severan?) between
Helios and Apollo.
38. For example, PGM 7lines 529f. (third cent.); 4lines 481£., Helios Mithras the
Great; lines 640f., "0 Sun, Lord of the sky and earth, god of gods" (but, lines 987f.,
Horos Harpocrates is addressed in closely similar terms); lines 118lf., "Helios,
father of the Cosmos," etc., d. 1997£.; 12lines 115£. (A.D. 300-350), "the fiery god" is
"master of the whole cosmos."
39. Cumont's long-popular interpretations are disposed of by Seyrig (197lb)
338-366. One of the latter's main points is the distinction between Sol simply being
worshiped (e.g., p. 365), and being worshiped as Zeus (or as some king-god). On
Aziz, see further Hanslik (1954) 180, the god identified with Apollo and put on coins
of Gallus, A.D. 252.
40. Dothan (1968) 99-103; the gesture is often discussed, as for example by
L'Orange ( 1935) 89. It appears on coins from Sept. Severus on, as characteristic of
Sol, taken up simultaneously by the emperors (p. 95) along with the globus in the
left hand (p. 97); but Brilliant (1963) 109 Fig. 3.13 and p. 110 shows the gesture
earlier (Trajan), d. also 149 and Fig. 3.111 (Commodus) and 209 (Sarapis in 216)-
so Sol has no monopoly.
41. Kantorowicz ( 1963) 117-126; Fears (1977) 241, Trajanic and Hadrianic aurei;
257 and 267, Gordianic; and the same symbolism with Aurelian, ibid. 285, and with
Licinius and Maximin Daia in A.D. 308, d. Seston ( 1956) 185. Against Kantorowicz,
I think Oriens means "the eastern regions" (sometimes personified as a woman, d.
RIC 5,1.290), not "on the rise" or "rising-ness." Against Fears, I think Sol giving
the globus even to Jupiter indicates friendliness, not superiority, and is therefore
not "a numismatic statement of solar monotheism" (p. 285£., cf. Saturn giving Jove
his thunderbolt, p. 287-not Saturnian monotheism).
42. Sol invictus on second cent. coins, Kantorowicz (1963) 128; the survey carried
into the third cent. by Storch (1968) 200-203; the epithet is applied also to Perse-
phone, below, n.53. It is attached to Commodus and Severus on coins and to Cara-
calla and emperors thereafter in coins and inscriptions, L'Orange (1935) 97; first
epigraphic appearance in A.D. 158, Seyrig (197lb) 366; Sol invictus on Gallienus's
coins, in De Blois (1976) 115, that emperor being the first to wear Sol's jeweled
diadem instead of the rayed crown, d. Arnheim (1972) 25; Aurelian also wore the
diadem, ibid. 26, d. Aur. Viet., Epit. 35.3; and his coins proclaimed Sol dominus
imperi Romani, c£. A. Alfoldi (1967) 376=CAH 12 (1939) 193, and Redo (1973) 68.
Notice, however, that (except for a period in the Serdica mint) Sol does not crowd
out Mars and Jupiter from Aurelian's coinage (RIC passim). On Sol as conservator
Augusti under Probus, see Fears (1977) 293. For Sol invictus on Tetrarchic coins, see
M. R. Alfoldi (1963) 197-199, there being one gold issue by Galerius, twelve by
 Notes to Pages 85-87 187

Maximin Daia, and two by Constantine in A.D. 305-309; thereafter, fifteen by Con-
stantine or his sons; cf. Fears (1977) 302, the lastSol-issuesdating to 324 or 325, with
A. Alfoldi (1948) 55£. and Lafaurie (1966) 803. Seston (1956) 185 notes Licinius's
coins honoring Sol in 308 and 317 (and see further, below n.53). The best-known
Constantinian Sol-Comes issue is closely derived from an issue of Probus, cf. M. R.
Alfoldi (1963) 51. The cult offered by Constantine is termed, by L'Orange (1935) l 0 I
and 107, a Staatsreligion, staatlichen Sonnenkult. Compare Merkelbach, below,
n.46, or Latte (1960) 365, where Aurelian and others from the mid-third century are
seen trying to establish a "Reichsreligion" through Solar cult, very much as
Swoboda (1964) 72 can say that "Sol as invictus from Aurelian on stood at the center
of the state cult." But no such descriptions seem justifiable.
43. Paneg. vet. 9( 12).2.5, on the dii minores, the orator ending, 9( 12).26.1, with an
invocation to "the supreme creator of things, summus rerum sator, whose names
are as many as the tongues of peoples."
44. SHA Elagabalus 7.4; for the coins, cf. CAH 12 (1939) 54 n.2; the main sources
are Herodian 5.5.6-5.6.8 and Dio 80.11-13, Dio expressing outrage at the subordi-
nation of Zeus himself. Ephemeral traces of Elagabalus's efforts are canvassed by
Dupont-Sommer and Robert (1964) 79£.
45. Zos. 1.61 and SHA Aurelian 25.4 and 31.7-9 seem in conflict. Homo (1904)
184-189 refers to the Roman temple attested in the Regionaries, etc., and the
pontifices dei Solis in CIL 6.2151, etc. For a short and more sober account, see
Mattingly in CAH 12 (1939) 309. On the buildingatPalmyra,cf. Seyrig(l971) 110£.:
Belus is actually Bel, nota solar deity at all. Note CIL 11.6308 (Pisaurum), Herculi
Aug. consorti d. n. Aureliani, at odds with any picture of solar monotheism.
46. Grant (1968) 173 speaks of the moment around A.D. 309. Cf. Merkelbach
(1962) 234: "The last stage of ancient paganism is the syncretistic Helios cult. In the
3rd cent. it was for long periods the official religion of the Roman state; all other
religions were to be subsumed in that of Helios." The statement is in accord with
the importance most scholars attribute to the cult.
47. Plut., Moral. 413C; Firm. Matern., Math. 1.10.14, writing in the 330s, quot-
ing almost verbatim from Cic., Rep. 6.17; and Porphyry, Isagoge p. 181 and in
Servius, ad Eel. 5.66, "according to Porphyry's book that he called The Sun, the
potestas of Apollo is triple, and it is the same being that is the Sun among the gods,
Liber [Dionysus] on earth, Apollo in the lower regions'' (quoted again, more or less,
in Macrob. 1.17.5£.). Some of these passages are adduced by Cumont (1913)452,453
n.l, and 461£. n.3. I am not clear about the meaning of Pantheus (CIL 8.12493, Zeus
Helios Pantheus Sarapis; 9.3145, Liber Pantheus; etc.). Cf. Latte (1960) 334.
48. Dio Chrysos., Or. 31.11, cf. the preceding note, for Porphyry, and Seneca, De
benef. 4.8.1-3 and Nat. quaest. 2.45, and Varro earlier still, in the synopsis of Aug.,
Civ. dei 4.11: Si omnes dii deaeque sit unus Jupiter, sive sint, ut quidam volunt,
omnia ista partes eius sive virtutes eius, sicut eis videtur quibus eum placet esse
mundi animum, etc. Compare the Ps.-Aristot., De mundo 40la, "Being One, you
[god] are of many names, being called by all the conditions, nci9&ow, which you
have created" -Zeus, with many epithets, etc. etc., in a long, hymnlike section.
Strohm (1952) 169 notices the literary connections.
49. Sen., De bene/. 4.8.3, concluding, omnia eiusdem dei nomina sunt varie
utentis sua potestate; and Plut., Moral. 366C, a~~oO 8uva11t~ ... citv{'t&t'tat, and
188 Notes to Pages 87-89

367C, on Dionysus, Hercules, Ammon, Demeter, etc. His use of dynamis as "god"
in the Stoic sense is common, cf. 375B or Nock (1926) Iii£. and n.62. On ainigmata as
a means of reading deeper truths into myths, see above, nn.16-19.
50. Macrob. 1.18.18-20; for the date of Labeo, quite uncertain, see Bardon (1956)
264. On the oracle's philosophizing, see above, chap. 1.1 at nn.53 and 62-65.
51. Alex. Rhet. in L. Spengel, Rhet. graeci 3 pp. 4£. =C. Walz 9 pp. 337f., the form
of praise- "worshipped in all lands," vel sim.-a cliche, e.g. in S/G3 867. a. Sext.
Emp., Pyrr. hyp. 3.219, "in actuality," i.e., among o{ Katd tov P!ov, "some people
say there is one god, some, many gods that differ also in shape."
52. Philos. 17(1).5=11.5. Eusebius, De laud. Const. 1.3 (PG 20.1321) says the
same, just as emphatically; but from him, the accuracy of the statement is very
suspect.
53. Above, n.41 ad fin.; on Sol's gesture, n.40; on invictus applied to Persephone,
see Flusser (1975) 13f., apparently Tetrarchic. Notice also Mars, Mercury, and Sara-
pis, all invictus, in RIC 5, 1, p. 305, GIL 9.425 and AE 1968 no. 232, and Mil dvdJC1']-
to~ (Cybele) in Pergamon, Ohlemutz (1940) 185. On aetemus, see Cumont (1888)
185, arguing that, because 10 of 21 known uses of the epithet are post-100, all must
be (I)-repeated, idem in RE s.v. Aeternus (1894) col. 696 and idem (1929) 268, cf.
further DE s.v. Aeternus (1895) 320: the adjective is restricted to sky deities, Diana
(GIL 3.6161), Caelus, etc. For dominus, see above, n.42-in Greek, lCUpto~. e.g.,
"Zeus, Lord" being "widespread" in the later Empire, Robert (1950) 67f., speaking
of inscriptions; likewise in papyri, e.g., POxy. 1678 (third cent.), prayers to "the
Lord God," the editors comparing 1680 and 1773 (both Tetrarchic), and 2984,
"Lord Sarapis" (Severan). For local supreme gods, see Ammon among Egyptians,
Plut., Moral. 354D, "who they think is the whole"; Men in Lydia, "the one god in
the heavens, great dynamis of the undying god," "the one and only," in Herrmann
and Polatkan (1969) 51£., second half of the third cent.; and a variety of hyperbolic
phrases in magical papyri or invocations in PGM passim; Schermann (1909) 6£., 9
and passim; Keil (1939) 119, "thou, Artemis, greatest name among the gods!"
(Ephesus); and Nilsson (1963) 112£. For deities calledGijftato~, "highest," cf. Poland
(1909) 179 on Zeus and his guild in Miletus and Sabazius in Thrace; Robert (1968)
594 on the Milesian guild (but Robert is cautious about a Judaizing element, here,
and notes the god's priest also worships Sarapis and Apollo); Simon (1972) 376
showing Zeus the usual recipient of the title "highest," which (384) is translated
summus in the prayer of Licinius's troops; but (376) Attis also is "highest," cf.
Cumont (1929) 59 and 227 n.57 (A.D. 370); other gods also, MAMA 5.211 =IGR 4.542
(Phrygia) and E. Peterson (1926) 211. And finally, among the Tetrarchs, Sol can be
"Guardian" of Galerius (above, n.42) without challenging the tenets of the Jovian
dynasty.
54. MacMullen (1962) 379f.; but one could add a great deal from Greek inscrip-
tions, as can be sensed from refs. in the preceding note. On acclamations, see chap.
2.2, n.28, or E. Peterson (1926) passim, e.g., 205,24l,and254(andamongtheseveral
hundred inscriptions he gathers, barely a handful pre-date A.D. 300, unless they are
also Christian).
55. Noticed by Rose (1924) 61, in Plutarch "as with so many Greek writers," the
indiscriminate use of "the divine" or "gods"; c£. also Beaujeu (1955) 1.335; and add
two significant texts: C/G 5041 (Nubia, second cent.), "venerate divinity, sacrifice to
 Notes to Pages 89-92 189

all the gods!" (line 2); and Max. Tyr., Philos. 4(34).6, "Zeus and the gods, fathers
and makers of land and sea ... "(note the plurals, compared to n.52, above). For
usage in Latin, see Perret (1949) 799 and Cardauns (1958) 60.
56. IGR 3.903, cf.IG 4, 12.499, Artemis Hecate (second cent.), or the triple Mother
Atemis Anaitis, taken as Ephesian Artemis and Persian Anahita by Anderson (1913)
272, but Buckler (1914-1916) 177 gives itasAtemis; forMa=ArtemisorMa=Cybe1e,
cf. RE s.v. Ma (A. Hartmann, 1928) col. 89; Cybe1e=Anaitis on coins in Robert
(1967) 74£.; Fish wick (1967) 145 on CIL 9.3146, showing "the cult image of Bellona
actually stood in the temple of Magna Mater" at Corfinium; and, at Baalbek, Cybele
portrayed as Atargatis, in Graillot (1912) 190. So Cybe1e is rightly called nol..uci>w-
1-lOc; Mt\tllP sometimes, cf. Orph. hymn. 27 line 4 or Vermaseren (1977) 81.
57. Apul., Met. 11.2, cf. 11.5 where the goddess calls herself by an endless number
of names, Proserpina, Cybele, Venus, etc.; Proserpina called "polyonymous" and
"the all-regal goddess," CIG 2415. Again, note the latedejixioin Wiinsch(1909)42,
dated ca. A.D. 400, p. 45; Cybele "polyonymous," see the preceding note, and Pro·
thyraia, too, Orph. hymn. 2; Isis polyonymous in fact or by name (or JJUptci>vuJJoc;)
in CIL 3.4017; Plut., Moral. 372E; Jerphanion and Jalabert (1908) 472; Vidman
(1969) 14, 166, 177, 235, and 280; Witt (1971) 112 and 121; POxy. 675; and Bernand
(1969) 2.150, 167 and elsewhere. Compare Selene "polyonymous" in IG 4, 12.422, or
the third cent. hymn to Tyche in Heitsch (1961-1964)2nded., vol.l, beginning, "0
goddess of many colors, of changing shapes, winged of foot ... , whether we call you
Clotho the Dark or Necessity," etc. On Zeus polyonymous, cf. Cook (1914-1940)
3,1.945; above, n.48; and Celsus in Orig., C. Cels. 5.41 and 45, "it makes no differ-
ence whether we call Zeus the Most High or Zen or Adonaios or Sabaoth or, as the
Egyptians, Ammon, or the Scythians, Papios."
58. POxy. 1380 lines 20£., 72f., 78f., 96f., 113, and Rome and Italy in lines 83 and
109; attribut~s and deeds, e.g., lines 181-184; similar long lists of titles or qualities in
Hermes hailed by Poimandres, cf. Schermann (1909) 6f.; Cybele's titles or qualities
in Orph. hymn. 14 and 27, or more loosely in Graillot (1912) 196-208; and above,
chap. 1.1 nn.63-75.
59. Soc., H.E. 3.23 (PG 67.204); Caputo (1977) 119, Lepcis, cf. Eros=Liber/
Apollo, IRT 299.
60. Imhoof-Blumer (1911) 18 (Alexandrian coins of Domitian's reign), 15 (of
Severus's reign}, 13 (the coins almost all belong to the second/third cent.); and ex·
amples, 13-23 passim.
61. Merlat (1951) 166.
62. Mocsy (1974) 254, speaking of several hundred examples known, peaking
late, "a pantheistic synthesis of the gods most worshiped toward the middle of the
3rd cent."; and Tudor (1969-1976) 2.81 and Ill, dated to after A.D. 150; random,
irrelevant symbols added by artists, p. 155. For other kinds of document, which I
know only casually, see Levi (1944) 271-313, esp. 306, and Nilsson (1945b) 1-6, on
wildly complicated Aion/Kronos symbolism; Charles-Picard (1954) 120-124 and
LeGlay (1966) 47 on Saturn reliefs in Africa, the latter speaking of "surcharge
decorative ... decoration somptueuse ... typique de l'epoque severienne"; and
Eitrem (1939) 59 on magical gem reliefs.
63. Toynbee (1944) 153, cited by Nock (1946) 150 n.42. He cites also Jahn, who
draws on undated sarcophagi but, before and after, speaks of Pompeian painting; so
190 Notes to Pages 92-94

clearly Jahn means sarcophagi of the Empire in general; also Kuster cited, Die
Schlange ... (1913) 83, who shows that he is speaking of "the Hellenistic-Roman
period" (whatever that means); also Bonner cited, where the artist being described is
said not to understand his own symbols, but belongs to the late fourth/early fifth
cent.; also Nock cited by Nock himself, the earlier (1929) article referring to a single
coin of 87 B.c. All five of Nock's references, spread over 600 years, or irrelevant, thus
prove nothing. He goes on to say, (1946) i51, "analogies from theMiddleAgesmay
be misleading. In the Middle Ages elaboration was individual. ... " All artfrom all
Europe over a millennium!!
64. SHA Alex. Sev. 29.2-highly doubtful. Dodds (1965) 107 cites it without
question (and as "Lampridius"l) and it is offered as illustration of syncretism by
older writers like Eitrem (1939) 58 or more recent like Fredouille (1972) 339.
65. Vermaseren (1956- 1960) 1.333 no. 276; Witt (1975) 486; Brommer (1973) Mf.;
Hatt (1978) 284, with instances of other Gallic deities; Forrer (1915) 48; Vuckovic-
Todorovic (1964-1965) 179 and 182, at Egeta, Severan; Lane (1971-1978) 4.62; C.
Picard (1922) 49; and Merlin (1908) 13f. and 23. Further examples in G. C. Picard
(1972) 48£.; Vermaseren (1977) 37; Collins-Clinton (1977) 38f.; Ohlemutz (1940) 83;
and LeGlay (I 976) 367f. Compare the altars to Hermes, Helios, Asclepius, Heracles,
and Zeus in the Demeter sanctuary at Pergamon, Ohlemutz (1940) 218£. And ex-
amples could be multiplied further.
66. Plut., Moral. 364E; CIL 3.686=Vermaseren (1956-1960) 2.382 no. 2343; Mer-
tat (1951) 155 and 157f.; CIL 14.123; similar instances of private individuals honor-
ing one deity through another, e.g. CIG 3159 or Fontrier (1883) 505. Compare the
instances of persons who are priests of more than one god: Laumonier (1958) 278
(late Severan, Caria), JGR 4.739 (post-A.D. 200, Phrygia); AE 1969/1970 no.116(A.D.
330s, Formiae); CIL 9.6099 (Brundisium); CIL 14.429 (second cent., Ostia); 9.1153
(late first cent., Aedanum); other examples in F. Dunand (1972- 1973) 3.185 n.2, all
in Greece and the isles. Or compare dedications to one god by priests of another,
e.g., Merlat (1951)200, to Sol byaDolichenus priest; CIL6.31181 (later second cent.,
Rome); 14.375f. (mid-second cent., Ostia); 12.2215 (Grenoble); 3.973 (Apulum) and
78681; 8.24522; and Robert (1955) 100 (third cent., Ephesus).
67. Vermaseren (1977) 36; IG 4, J2. p. xxxviii, A.D. 296; CIL 8.24519; above, the
preceding note, and chap. 1.1 n.73 and chap. 1.2 n.67.
68. AE 1975 no. 722, Liberi et Liberae secundum interpr[etati]onem s[acr?]orum
So[lis1] in tabula[m scriben]dum curavit; Robert (1955) 240; CIL 6.659; Schwert-
heim (1975) 357; ILS 3831, A.n.26l, Mauretania; CIG3538, A.n.l70;andabove, n.59,
Rhodes.
69. Vidman (1965) 393, on the miniature icons as guests, to which compare
parallels in Cybele cult, Graillot (1912) 204 n.l 0 (though his source, Mionnet, is not
very dear). On the feast calendar, see Teixidor (1977) 135.
70. MacMullen (1968) 81-96, esp. 86.
71. Nock (1958a) 294, rightly sees the presence ofvisitor gods in another's temple
as no sign of syncretism, but the reverse: the visitors retain their individuality; and
Robert (1969) 290 rightly characterizes W. M. Ramsay's views as "hautement fan-
taisiste," where Ramsay attempts to "syncretize" all gods into each other by endless
equivalences. Both Robert and Nock (and I) would not accept some of the views of
Toutain (1907-1920) 1, 2.230-236.
 Notes to Pages 96-97 191

4. Conversion
1. Iren., C. haer. 1.13.1 and 7 (PG 7.577 and 592), Marcus "possesses the greatest
dynamis"; Marc. Diac., Vita Porph. 21, A.D. 396, cf. 31 forasimilarcauseandeffect;
Euseb., Mart. Pal. 4.15 (long recension. ed. G. Bardy, Sources chretiennes no. 55
p. 136); Greg. Nyss., Vita S. Gregorii (PG 46.918); Tert., Apol. 23.18; and Acta
Phileae4.3, p. 349 Musurillo. Musurillo (1972) xlviiassigns the Greek text to a date
"not long after the actual martyrdom" of A.D. 306, i.e. (p. lxxi) 320-350, and Emmett
and Pickering (1975) 99£. defend its historicity. The Greek text, p. 335 Musurillo,
shows Phileas assuming the governor will accept the logic of belief based on
miracles. For passages showing similar reasoning-because miracles have been
wrought, therefore people did believe, or one should believe-see, e.g., Acts 8.16;
Acta Petri et Pauli 32-34 pp. 193£. Lipsius, dated in the second cent. by james(l924)
304f.; Orig., C. Cels. 1.46, converts would not have been made without miracles, cf.
1.2, 2.52, 3.3, and 8.45; lren., C. haer. 2.30.7 (PG 7.820A) and 2.31.2 (PG 7.824);
Hippol., Ref. 4.42; Theoph., AdAutol. 1.13; Just., IApol.19andDial. 7.3, 11.4, and
69.6; Hierodes in Euseb., Treatise on the Life of Apollonius 2; Lact., De mort.
persecut. 2.5; Euseb., H.E. 1.13.1; 2.1.7; 2.1.11; 2.2.2; 2.14.5; 5.15.17; 6.5.6; and9.1.9;
Demonstr. ev. 3.7 (PG 22.244), 3.5 (22.221) and 3.7 (22.244); Ps.·Ciem., Homil. 2.25
(PG 2.93); and on exorcism, above, pp. 50£.
2. lambI., Vita Pythag., 28.143; Marcus Aurelius, Med. 12.28; and Plut., Moral.
434D·F. Compare the inscription from the early Empire in Robert (1939) 201, to
Augustus "demi-god, manifest through his efficacy," t1tupavet ~pep d1to tfl~ tp-
yao{a~; the inscription from the late Empire, CIL3.13132 (A.D. 312, Arycanda), "the
gods revealed in deeds"; Philostr., Heroikos passim, as Eitrem(l929) lOpointsout;
or Eunap., Vit. soph. 458, where Iamblichus's disciples see "the manifest proofs ...
and believed everything."
3. Moral. 360C; d. 1126C·E; and among Christian writers, the Ep. ad Diog.
7.Sf.; Clem. Alex., Strom. 6.18.22£. (PG 9.400£.); Orig., C. Cels. 1.67; Acta S. Pionii
13.4f.; Euseb., H.E. 10.4.16f. and De laud. Const. 16.9 and 11 (PG 20.1427£.).
4. Apul., Met. 11.15, the irreligiosi will now be convinced they are wrong, see·
ing Lucius relieved of his sufferings; Celsus, in Orig., C. Cels. 2.20 (particularly
explicit); Maximin Daia, in Euseb., H.E. 9. 7.4, Tyre is happy because the gods
dwell there; pagans as reported ibid. 9.1.8; Symm., Rel. 3.8; and Christians like
Origen, C. Cels. 1.68, cf. 2.48, or Eusebius, Praep. ev. 4.16 (162c) and H.E. 9.8.14.
5. Philostr., Vita Apollon. 6.10; Porph., Ad Marc. 14, love of the flesh leads to
love of money, and so to injustice, but love of god leads away from the flesh; Tat.
21.1-6= Euseb., H.E. 4.16.8, the false philosopher is a glutton; lren., C. haer. 1.24.2
(PG 7.675), followers of Saturninus eat no meat, "by pretended abstinence of this
sort leading many astray"; and the debate in Euseb.; Treatise on the Life of Apol·
lonius 5 and 29, no real prophet takes money, cf. above, chap. 1.3 n.36 and Euseb.,
Demonstr. ev. 3.6 (PG 22.224A), with the allied and more common reasoning that a
true philosopher or prophet lives simply, e.g., in Clem. Alex., Strom. 3.5.35£. (PG
8.1148), Hippol., Ref. 9.7, and Irenaeus and Apollonius (not of Tyana) in Euseb.,
H.E. 5.7.5 and 5.18.4.
6. Ael. Arist., Or. 46, II pp. 399-406 Dindorf. As consensus settled on these
persons not being Christians, the passage dropped out of notice. It is last seen, so to
192 Notes to Pages 97-99

speak, in Boulanger (1923) 249-265, esp. 256£. On the type of itinerant and stleet·
corner philosopher, see MacMullen (1966a) 59£. and notes.
7. Reynaud et al. (1975) 57£., the text dated (p. 61) to the second half of the second
cent., the deceased a "professeur ou philosophe," possibly(?) crypto-Christian (pp.
72 and 74). But Christianity I think never presented itself as a regional faith.
8. Poimandres, Corp. Herm. 1.27£.
9. For Jewish missionaries, cf. Val. Max. 1.3.3, " ... who attempted to infect
Roman ways with the worship of Jupiter Sabazios" in 140 B.c.; and Just., Dial. 17 =
Euseb., H.E. 4.18.6.
10. Lucian, Alex. 10, 15, and esp. 24 and 36.
11. Euseb., Praep. ev. 4.2 (133b), speaking of that aspect of religion he calls
politikon, including both city centers and their territories (130b); cf. the votary of
Bes (who had a famous oracle), in a casual inscription at Abydos, claiming that the
god is "testified to in all the civilized world," in Cumont (1929) 233 n.9; for that
phrase, a cliche, chap. 2.3 n.51; and advertising on stelae, chap. 1.1 n.67, and chap.
1.2 nn.32, 64f., and 69-72.
12. Graillot (1912) 312£. gathers the refs. on Cybele beggars; Tert., Apol. 13.6,
says that (pagan) "religion goes the rounds of the cafes, begging"; Val. Max. 7.3.8
describes an Isiac beggar in Italy and Greece; and Fossey (1897) 60 reports an odd
inscription set up by a "servant of the Syrian Goddess," "sent by his Mistress Atar·
gatis" to Hierapolis and back twenty times, "filling his satchel"; but there is
nothing to show that he begged for her, i.e., that he undertook "quetes faites au
profit de Ia deesse" (p. 61 ). Other allegations of evangelizing are easy to find but
unsupported, e.g., in Cumont (1929) 81, Festugiere (1949) 231, Mocsy (1974) 255,or
Krill (1978) 40.
13. The multiple priest in Becatti (1954) 67 and Vermaseren (1956-1960) 1.127
and 131, nos. 255 (A.D. 162) and 269; the Hierocaesarean priest in Robert (1948b) 53;
a temple to Apollo Belenus built ex responso antistis in Prieur (1968) 174; and per-
haps ILS 3332, to Ops "by order of the priests" -of whose cult, however, is not
specified. So that last inscription perhaps belongs in the category illustrated in
chap. 2.2 at n.24.
14. I. Toth (1971) 24: Dolichenus priests in 60+ out of 400 texts, cf. forMithra, 40
out of 2000.
15. Merlat (1951) 72f. and others before him date the inscription to the Tetrarchy;
but I am better persuaded by Fitz (1959) 241, on CIL 3.3342f., and Mocsy (1974) 394
n.l87. The Dolichenus priests were esp. active in imperial cult and prayers for the
throne, cf. I. Toth (1971) 27 and Angyal (1971) 17.
16. CIL 3.14445 and 7760.
17. On temple funds, cf. above, chap. 1.2 n.34; also IG 4, 12.612, "the holy city of
Epidaurus" erects a statue of the emperor "from the god's monies." Eastern origin
of "the great majority" of priests: I. Toth (1971) 25; but arguments ibid. that they
came actually from around Doliche itself seem to me weak.
18. Malaise (1972) 73£.; Porph., De abst. 4.9-but c£. fake Egyptian (hieroglyphic)
inscriptions on statues in Rome, Italy and elsewhere in the West, in Tran Tam Tinh
(1971) 7.
19. Lucian, Cone. deor. 9, Mithra "not speaking Greek" (=Persian? Latin?
merely a comic touch?); Epiphanius, Panarion haer. 51.22.11; Fraser (1960) 65,
 Notes to Pages 99-101 193

Locrian "as a ritual language," cf. Merkelbach (1973) 49£. on IG 10, 2.255, a testi·
monium to the god's powers from the Sarapeum in Thessalonica in Locrian; and
Serv ., ad Georg. 2.394, Greeks and Romans may use their own language for hymns,
e.g., to Liber=Dionysus; but not to Cybele.
20. CIL 3. 7728, sacerdos creatus a Palmyrenis, domo Macedonia et adventor
huius templi.
21. At year's start: Tert., De corona 12. At Lambaesis, the two kinds of vow were
next door to each other, cf. the Capitolium there attested, in LeGlay (1971) 125.
Capitolia resembled that at Rome, cf. Wissowa (1912) 41. To the listofCapitolia in
Toutain (1907-1920) l,l.l84f. or DE s.v. Capitolium 93-95, other examples could
now be adduced from archaeology or epigraphy, e.g., Jobst (1976) 3!1-52 passim or
AE 1925 no. 30 and 1926 no. 26.
22. BGU !162 section V. Wilcken (1885) 462 suggested that "ancestral" described
Jupiter for Egyptians post-212; further, jouguet (1911) 402£. For Caracalla's edict,
see PGiess. 40 lines 309f. as restored (not with certainty) by Oliver (1978) 405.
23. Laumonier (1958) 278.
24. CIL 13.7281, A.D. 236, restoration ofthe Mons Vatican us by the local hastiferi;
13.1751 and 7281, Lyon.
25. Ambros., Ep. primae classis 18.30 (PL 16.980), simulato Almonis in flumine
lavat Cybele; C. Picard (1922) 58, Castalia at Antioch and Claros, as at Delphi; c£.
the possible imitating at Lyon of the triangular temenos of the Ostian Phrygianum,
Vermaseren (1977) 61. On canopusas a generic term, cf. Malaise (1972) 107, on CIL
9.1685; on a Sarapis-statue at Rome "like that in theCanopus-temple," IG 14.10!10.
26. Robert (1965) 124 shows copies of old, famous icons planted in Marseilles,
etc. (see Strabo 4.1.4); and Malaise (1972) llO notes Isiac icons in Portus, on true
Egyptian models.
27. CIL 13.7317; Graillot (1912) 12!1 n.5 on CIL 14.324; cf. IG 14.1084, decree of
hymners of Sara pis dated by both Roman and Egyptian calendar, "no doubt exactly
as if in an Egyptian temple," says Weber (1911) 16.
28. Robert (1960b) 342 points out that, at one time, 32 of 36 hymnodPs are Roman
citizens.
29. Toutain (1907-1920) I, l.249f.; on the Syrian shrine on the Janiculum, c£.
Graeve (1972) 322-331 on the building and 334 on the image, and Goodhue (1975)
65f. pointing out eccentricities in the plan, Roman mixed with Syrian.
!10. Malaise (1972) 133, 1!16, and 1!19 on Isiacism; Bruhl (1953) 276,285, and 288f.
(somewhat confused discussion, and basing too much on IGR 1.637); on Saturn,
Merlin (1910) 38, noting the assimilation of a local ba'al to Saturn by the second
cent.; further, Charles-Picard (1954) 163; Benabou (1976)373;andLeGlay(l966)70,
73, 86, and 89. These scholars also note regional variations, not only chronological,
in the susceptibility of Saturn worship to external influences. For other variations
regionally, see Edelstein and Edelstein (1945) 2.189 on cult rules of Asclepius, and
75£. on variations in Asclepius's story. For another instance of cult practices chang·
ing over time, see Duthoy (1969) 88, 93 n.6, 96 n.7, and 102£. n.7, on the clearly
marked phases of Cybele worship.
31. The variations are most compendiously surveyed by Beskow (1978) 9-12;
further, Francis, in Cumont (1975) 154f.; Cumont (1975) 199-203, giving forced
arguments on behalf of the seven grades; Porph., De abst. 4.16, on women Mith-
194 Notes to Pages 101-103

raists, with G1L 14.69; 10.1591, if it is Mithraic itself, as assumed by Vermaseren

(1956-1960) 1.104 no. 177; CIL 13.7958=Vermaseren (1956-1960) 2.54 no. 1034; fur·
ther examples of women votaries Procrusteanly rejected, idem ( 1956-1960) 1.254 no.
696 and 1.256 no. 705, with nos. 1463 and 1952 (perhaps not Mithraic): idem (1956-
1960) 2.310 no. 2065; Toutain (1907-1920) 1,2.122£. andSchwertheim (1979) 35, on
cult-relief variation, and 26, on Mithra Mercurius; chap. 2.3 n.65 on Mithra's
synnaoi.
32. In Aug., Giv. dei 6.11; d. Firm. Matern., De errore profan. relig. 16.1, report·
ing confusions in Minerva or Liber cult.
33. Plut., Moral. 352A, 355F, and 356Af.; d. Lucian, Dea Syria 7, and Firm.
Matern., De errore profan. relig. 1.2, on Osiris's various meanings; on Cybele, the
contradictions reviewed by Vermaseren (1977) 88-92 and 111£.
34. Rehm (1958) 301.
35. Menander Rhet., Epideictica 8 p. 342 Spengel, Rhet. graeci. Compare the
wording in Firm. Matern., Math. 1.1.4, speaking evidently of writers on religion
generally, irresponsibly inventive; and note the sense of uncontrolled profusion of
beliefs in Alex. Rhet., in Walz, Rhet. vol. 9 p. 337, on Heracles legends, etc.
36. Imperial monies to temples or cults, e.g., at Teos by Hadrian, in Robert
(1946b) 89; at Didyma by Trajan, in Haussoullier (1902) 154f. and 281; at Aezani by
Hadrian, in GIL 3.355 =OGIS2 502; at Pergamon by the same emperor, but to
hymnodes of theimperialcult, GIG3148lines33f., in Keil (1908) 104n.9;atBaalbek
by Trajan and Caracalla, in Rey-Coquais (1967) 38; at Magliano, Tuscany, by
Hadrian, AE 1946 no. 222; at Puteoli by Pius, in Franciscis (1954) 285, and at Lam·
baesis by the same, ILS 3282; at Kourion, Cyprus, byTrajan, Mitford(l971)208no.
108; at Carnuntum by the Tetrarchs, in A.D. 308, GIL 3.44 U; and dedications or gifts
of statues by emperors, at Pergamon, in Ohlemutz ( 1940) 150 and elsewhere, above,
chap. 1.1 n.20. I omit temples built by emperors in Rome itself, e.g., by Caligula to
Isis (rebuilt by Domitian after fire) or by Severus to Liber, ILS 3361 =Palmer (1978)
1088f. On the Pompeian Isis temple, see Tran Tam Tinh (1964) 24f.-though
Henzen and Huelsen, whom he cites (at GIL l, 2 p. 334), do not support him. The
two categories, Roman civic cults and extra-Roman, overlap (as with Jupiter, the
most obvious example); or they seem to overlap (as with Vesta-d. Nock (1930]
257-260, surely exaggerating the reach of her cult very greatly).
37. The whole of Antinous's cult is called "artificial and fragile" by Beaujeu
(1955) 256, in summing up pp. 249-252; further traces of the cult in IRT 279,
Antinoo deo frugifero Lepcitani pub lice, and chap. 2.3 n.59. Hadrian's encourage·
ment of the cult of Disciplina is traced by Beaujeu (1955) 164, and of Urbs Roma, by
Fink et al. (1940) 103 and 110f.
38. Graiilot (1912) 153 on Frg. Vat. 148 (=FIRA 2 2.496); Beaujeu (1955) 314f.,
who rejects a Claudian date for the new law; and Lambrechts (1952) 15lf. and 157f.
(noting changes within the cult's calendar, p. 144), which Duthoy (.1969) 116f.
gingerly accepts.
39. On Domitian's Isiacism, d. the priests at his dinner table, Plin., Paneg. 49.8;
on the emperor's favorites, d. IRT 295, lar Severi patrius. I withhold dozens of
examples of all sorts of male and female deities, eastern, western, or whatever,
labeled "August" in inscriptions from Augustus on. Nock (1925) 92 offers some
 Notes to Pages 101-106 195

speculation on the usage and Fish wick (1978) carries it further, but with results still,
I think, a little blurry and imprecise.
40. On Elagabalus, see above, pp. 85f.; against the Severi as "Orientalizers," d.
Mundie (1961) 228-237, methodical and persuasive.
41. The inscriptions are given by Graillot (1912) 228 and Tran Tam Tinh (1972)
119. The supporting guilds of cannophori etc. had also to be licensed, d. Graillot
(1912) 143-but that was true of guilds of other sorts, too, secular included.
42. CIL 3.7751, of the A.D. 170s; IGR 4.1431, d. BGV 176 (Hadrian) showing the
governor interfering in the staffing of temples (but the text is very fragmentary).
43. Ulpian, Regulae 22.6 (FIRA 2 2.285-d. above, chap. 1.1 n.25); AE 1933 no.
123; SIG 3 867; CIL 10.7946; and Ivanov (1974) 63.
44. Plin., Ep. 10.49; on Republican practices controlling temple construction,
see Cic., De domo 49.127; Livy 9.46.7, in RE s.v. Consecratio (Wissowa, 1901) col.
898; CIL 6.360= ILS 366, showing that the curatores aedium sacra rum must author-
ize use of space in Rome for worship, d. Dig. 1.8.9.1 and CIL 6.31128, the latter cited
by Stambaugh (1978) 559, with Dig. 1.8.6.3 and Plin., Ep. 10.8.2 (A.D. 99); and see
also above, chap. 1.1 n.66.
45. A samplingofmuchevidence: CIL8.1100(Cybele); 5.484and8225(1sis), 6351
and 6358 (Hercules); 11.696 (Oolichenus); 18.1788 (Isis); and Fishwick (1967) 146,
locations for Bellona; Toutain (1907-1920) 1,2.59 points out the lack of any such
texts for Syrian deities, which is interesting. In Africa, the abbreviation seems to be
simply dec. dec. (the curia approved the dedicant's act), d. ILS 3181 and AE 1938 no.
288; but cf. the same bare d. d. (no locus mentioned) in AE 1975 no. 267 (Paestum) or
CIL 9.5177, decreta ordinis (Asculum).
46. In Africa: CIL 8.26241 (temple for Saturn); AE 1899 no. 111 (dedication to
Caelestis); AE 1957 no. 68(templeforCybele); ILAfr.551 (temple to Saturn); Merlin
(1908) 21, temple decorations (now=CIL 8.25520); Merlin (1910) 23, to local ba'al
and Tanit; and LeGlay (1966) 405. In other provinces and Italy, see Plin., Ep. 8.6;
IGR. 1.420; CIL 11.1545; 9.981; Tran Tam Tinh (1964) 83 n.l; CIL 18.8701 (Mars in
Ger. lnf.); ILS 8191 =AE 1969/1970 no. 441 (Mercury in Ger. Inf.); and AE 1960 no.
855=Parvan (1918) 386 (temple for Nemesis).
47. On Scaptopara, d. above, chap. 1.2 n.36; on religious fairs, above, pp. 25£.
48. CIL 10.846.
49. Duthoy (1969) 86 and ILS 419, etc.; CIL 12.4883 (A.D. 11-13).
50. Robert (1937) 30 improves the text.
51. "Senate and people" active in religious oversight at Attaleia, in Dupont-
Sommer and Robert (1964) 80; at Laodicea, CIG 3936; at Stratonicea, above, chap.
1.4 n.l8; the gerousia at Ephesus, in C. Picard (1922) 95, though note the role of the
demos in SIG 3 867; Lydian villages honor Men in A.D.l61/2andSabaziusin 101, in
Schwertheim (1975) 357 and Robert (1948b) 112f.; on other secular/sacred connec-
tions, compare the sanctification as "hero," e.g., CIG 3936 or above, chap. 2.1 n.42,
with the gods applauding magistrates, allegedly, above, chap. 2.1 n.37; or on non-
professional priests, the sharing of facilities, etc., above, chap. 1.4 nn.l-3.
52. On the Syrian triad, c£. Prentice (1922) 184; d. also the confusions of cult in
Lucian, Dea Syria 4 (Sidon), 6£. (Byblos), llf., 28 and 88. In Greece, notice, e.g.,
Paus. 2.23.6; 7.6.5; and 7.17.10.
196 Notes to Pages 106-109

53. Above, chap. 1.4 n.l4 and chap. 2.4 nn.30f.

54. On Artemis, cf. Forsch. in Ephesos 3 (1923) 144, the priestess of the early third
cent. who "renewed all the mysteries of the goddess and established them in their
ancient form" -though notice the collapse of other Artemis mysteries in the same
city a generation or so earlier, in C. Picard (1922) 290£.; on Demeter in Pergamon,
see Ohlemutz (1940) 221£.; on Lagina, Laumonier (1958) 255 and 257, first mention
under Marcus Aurelius, and note the invention of the Komyria in the first cent.,
ibid. 247; on Abonuteichus, see Lucian, A lex. 38£.; and there are the otherwise un-
known Asclepius mysteries in Athens, in Philostr., Vita Apollon. 4.18, evidently a
variant on the Eleusinian. Festugiere (1935) 202 draws out a few more signs from
inscriptions of Magnesia, etc., suggesting mutual imitation and stereotyping among
mystery ceremonies.
55. On the panegyris at Cos, cf. Themanakis (1893) 208, second cent.; c£. the
priest praised as "having renewed the panegyris of Zeus Protomysius" (near Prusa,
second/third cent.), in Robert (1949) 35; and above, chap. 1.4 n.l9.
56. In inscriptions, the phrase is common, vetustate conlapsum, e.g., in GIL
3.1790,3342,4540, and 11676; 8.25520; and 13.7281; Inscr.1tal. 10, 1.585; /LS374l;
Milik (1972) 197; AE 1974 no. 574; esp. GIL 3.4796, a Mithraic "temple in ruins,
deserted for over 50 years." In Greece, cf. Paus. 1.44.3; 2.15.2; 2.24.3; 2.34.10; 5.27.11;
6.24. 10; 8.14.4; 8.15.5; 8.30.6; 8.32.3; 8.42.12£. (roof fallen in, three generations ago,
but worship continues); and 10.8.6. In Africa, note the unexplained decline of the
Tanit shrine at Hadrumetum from the end of the first cent., in Cintas (194 7) 79£.-
worshipers few and poor. On other shrines, see Kunze et al. (1958) 59 (Olympia); C.
Picard (1922) xxviii and 86£. (Ephesus) and 122£. (Claros); Robert (1953a) 20, the
second century was "Ia grande epoque" for Claros; further on C1aros, above, chap.
2.1 nn.54f.; Laue (1931) 129 and 134 (Epidaurus); and Kotting (1950) 32(Epidaurus).
57. Matz (1940) 502f.; above, chap. 2.1 n.l3.
58. Beaujeu (1966) 4.
59. Crema (1959) 472; Tran Tam Tinh (1971) 7£., speaking of "l'amourdu 'mys-
terieux' ... , le gout exotique ... , egyptomanie ou amour d'Isis?"; idem(I964)50£.;
Vidrnan (1966) 116, "a temporary fashion which did not affect hearts and souls very
deeply."
60. Strabo 14.1.5, compared by C. Picard (1922) 46 to the never-finished shrine of
Apollo at Claros.
61. Dussaud (1922) 232f., the Damascus temple possibly begun in Severan times;
Golfeto (1961) 42 and 48. For other shrines constructed as family projects, see Merlin
and Poinssot (1908) 113£., at Uchi Maius; G. C. Picard (1977) 101; at Yzeures=Icio-
durum, in Severan times; Inscr. /tal. 10, 1.640 (Ruginium); GIL 5.308f. (Histria)
with PIR s.v. Vibius 404, late Republic; Prentice (1908) 104-123, esp. 122£., an
interesting story in a Syrian shrine stretching through the first and second centuries;
and above all, in Palmyra, the activities of the four main clans, in Gawlikowski
(1973) 38£. and 48f. In Pompei, note the special connection of the very distinguished
Popidii with Isiacism, in Tran Tam Tinh (1964) 41£., modified by Castren (1975)
207.
62. Merely as illustrations, note the individual slaves as builders in GIL 5.8237
and /G 3.74=SJG5 1042, dated by Oliver (1963) 318 and Sokolowski (1969) 106f. For
building by cult groups, see GIL 3.1547; 6.647; 7.11; and 8.18810 and 8457=20343,
 Notes to Pages 109-111 197

the religiosi of Cybele in A.D. 288; AE 1948 nos. 26f., late second cent. in Ostia to
Bellona, with discussion by Fishwick (1967) 146.
63. Builders of shrines on their own estates: Siiss(l972) 169-173, A.D. 180 through
the third century in Friedberg; Hatt (1978) 285, Mithra and a local god in Mack-
wilier housed by an eques; a second-cent. imperial bailiff in Galatia builds a
temple, in Anderson (1937) 19f. and Macpherson (1972) 219; a late Severan eques
and decurion of Aquincum establishes a cult on family lands at the request of the
villagers, in Balla (1971) 55; and an imperial slave builds to Liber, probably on
praedia Augustorum, Inscr. Ital. 10, 1.585 (Pola). On the Torre Nova Bacchic
group, see Vogliano (1933) 216f., where more than three-fourths ofthe members are
Greek (pp. 227-231) but by now probably libertine. For the Samothrace compari-
son, d. above, chap. 2.1 n.l9.
64. Casual attention, no special search, produces AE 1919 no. 26, M. Aur. Deci-
mus of the later third cent. in Lambaesis, to a minor ba'al; 1973 no. 631 =CIL
8.2678a add.; Marcillet-Jaubert (1974) 250£.; and Birley (1978) 1523 n.9, altars by the
same man to I.O.M., Juno, Minerva, Sol Mithras, etc.; a Severan centurion restoring
a Mithraeum and setting up a Dolichenus altar, in Vermaseren (1956-1960) 1.69;
ibid. 102 no. 171, the individual offers a dedication to Mithra and to Mercury in
Italy; the Severan tribune in Pannonia who restored/built temples to Diana Tifa-
tina and Sol Elagabalus, in Fiilep (1954) 247 and 322, Fitz (1959) 240, and idem
(1972) 90; Tib. Cl. Thermadon setting up buildings/statues to Sol Mithra and
Fortuna Primigenia in Italy, CIL 6.3723= 11.2684 (and p. 424: Bolsena) and 14.2853
(Praeneste); the aedile of Vulcan and Ostian magnate who restores the temple of
Castor and Pollux, Fortuna and Venus and Ceres, mid-second cent., CIL 14.373-
376; the senator Antoninus building temples and amenities at several shrines at and
near Epidaurus, in Paus. 2.27 .6f.; the wealthy Cornelia Baebia building to Poseidon
and another deity under Commodus, in Scranton (1944) 344-348; the senator Ruti-
lianus serving several deities in Rome and Tibur, in Stein (1925) 260, in the A.D.
160s; a praef. co h. from Africa honoring I.O.M. and Liber in two Dacian cities, CIL
3.6257 and AE 1975 no. 722; and P. Ael. Euphorus building for Mithra in one city
(AE 1971 no. 384) and honoring Silvanus in another (CIL 3.1363) in Dacia.
65. Leschi (1957) 177f.
66. Above, pp. 66£.; Malaise (1972) 111, Isiac priests' gifts; d. Mart. S. Pionii 3.1
and 7.1, in Smyrna (Euseb., H.E. 4.15.46), the neocoros is leader, and Mart. S.
Cononis 2, at Magydos the neocoros leads accusations.
67. LeGlay (1954) 273-276 and Daniels (1975) 254; Habicht (1969) 63.
68. The Feriale Duranum meticulously edited by Fink et al. (1940). The editors
assume extreme centralization: the calendar was not only drafted in Rome but
Ulpian "doubtless had a hand in the composition of the Feriale" (p. 37). No evi-
dence supports this; and Ulpian is not so good a candidate for the conjecture as
Comazon, cf. Syme (1972) 407f. The holidays are all old-Roman or old-Italian, e.g.,
the Neptunalia (Fink et al. [1940] 148).
69. Seston (1956) 184-186 on ILS 8940, set in Salsovia in Scythia. For Licinius's
loyalty to gods, plural, though by Christian writers called "demons,'' cf. Euseb.,
H.E. 10.8.4; Vita Const. 1.54 and 2.5, with emphasis on the reliability of the latter
text by A.H.M. Jones (1962) 113. For the prayer to summus deus of A.D. 324, d.
Lact., De mort. persecut. 46.6; on Sol worship earlier, see above, pp. 84£.
198 Notes to_ Pages 111-111

70. Corporate, troop-unit dedications or offerings or construction: for Diana

Tifatina, in Fiilep (1954) 322 and Fitz (1972) 179; to Sol Elagabalus, ibid. 178 (AE
1910, 141); to Jupiter Heliopolitanus, CIL 6.421; to Silvanus in Lambaesis, LeG lay
(1971) 131; to Silvanus in Aquincum, LOrincz (1978) 307; to Jupiter Dolichenus,
Vuckovic-Todorovic ( 1964-1965) 176; to Jupiter Optimus Maximus, a number of
examples, e.g., ILS 3061 or Birley (1978) 1510; to Liber, ILS 3381 =CIL 3.1790; to
Mars, ILS 3154; to Hercules and Silvanus, ILS 3740; to Isis, CIL5.404l and A£ 1948
no. 145 (with Sarapis); to an unknown deity in A.D. 190, Leonhard (1914) 46; and
many from outside our region of study, e.g., Birrens (RIB 2092-3-7, 2100-4-7) or
Maryport (see Wenham's study cited in RIB p. 271, and nos. 815-7,830-1-7-8 and
842-3). For dedications inside a camp, d. Schwertheim (1979) 26, for Mithra. And
note the Severan s(acerdos) l(ovis) D(olicheni) ad leg(ionem) in Angyal (1971) 17 =
AE 1965 no. 30. There are governors directing temple construction or dedications,
e.g., ILS 3957, to Tellus; to Fortuna in Oescus, with the town bearing the cost,
Ivanov (1974) 63; to Diana Plestrensis, Velkov (1974) 152; and n.65, above. Legion-
ary commanders offer dedications often (I withhold many inscriptions by lower
officers, tribunes, centurions, and prefects), sometimes clearly not ex officio-e.g.,
with their wives, in ILS 3085-3087 or 3092 (or wife and daughter, CIL 8.2630), or
from a vision, ILS 3229; cf. CIL 3.5862, dated to late third cent. by Schwertheim
(1974) 221. Other texts do not indicate whether or not the governor appears ex officio
(presumably not): CIL 3.1111, 1118,4796, and 5862; 13.6754; in Africa, a third of all
named dedications to Mithra are by consular governors or legionary commanders,
in Vermaseren (1956-1960) 2.92-98. Citing several of these texts, LeG lay (1954) 277
infers "the official character of this cult in the military capital of Roman Africa."
The inference is gratuitous.
71. Above, chap. l.l n.37; Lucian, Alex. 35.

5. The Dynamic Cults

I. In the east, religious movements include Manichaeism-too late for this
book. Also, new but not major cultS for Antinous and Glykon, d. above, chap. 2.4
nn.IO and 37 and MacMullen (1966a) 115-119.
2. In IGR, Jupiter is the most often cited western god-with a mere seven men-
tions. Mithra is unknown, Dolichenus almost so.
3. Toutain (1907-1920) l,l.l88£., 191, and 198, accepted and a little expanded by
Frend (1952) 78-81; Charles-Picard (1954) 162£.; LeGlay (1966) 71, 73, and 86on the
long-delayed Romanization of Saturn cult, which is also little seen in the interior of
Proconsularis; 402f., worshipers are of the lower classes; Benabou (1976) 373 on the
success of Saturn in taking on a broad spectrum of needs, on the model of Jupiter;
and Fentress (1978) 508£. Compare I.O.M. apparently quite accepted by the local
aristocracy around Aquincum, but not by other classes, in Alfoldy (1961) 106. On
Jupiter opt. max. Apennininus conservator, see ILS 3074 (Rusicade).
4. For the Dacian inscription, see ILS 3086=CIL 3.1075 (Apulum). On M.
Campanius Marcellus who built the Diana temple at Intercisa, see Fitz (1968) 318
and 323 and idem ( 1972) 90£. and 200, where I cannot accept the idea of Diana being
"Orientalized." On her Campanian roots, d. ILS 1398 (Capua). That the "official
Diana-cult experienced a new upswing in the 3rd century" is the view of Fitz (1972)
261, who refers to RA C s. v. Diana (A. Michels, 1957) 966£. Michels rather singles out
 Notes to Pages llJ-116 199

the age of the Antonines for this Aufschwung, in tum relying on Aymard (1951)
155f.; and Aymard simply speaks of the Antonine age as "the Golden Age of hunt·
ing" (throughout the whole Empire, one must suppose), basing himself entirely on
a small number of coins. Other local Italian Dianas are easy to find, e.g., Nemoren-
sis, CIL 3.1773 in Dalmatia; d. Fortuna of Praeneste invoked in Sarmizegetusa, CIL
3.1421.
5. On Silvanus in Dacia, though not solely the Italian deity, seeS. Toth (1967)
78.
6. Negotiatores may identify themselves as such when they make dedications,
e.g., in CIL 3.7761 (Apulum, Suri to Dolichenus) or 11812 (Mainz, to Dolichenus);
but the practice is far too rare to produce usable statistics. The clustering of Mercury
inscriptions on commercial points and routes produces a striking pattern in the
Dijon-Langres area, d. Renardet (1973) 122, patterns which could no doubt be
found elsewhere, too. L. Toth (1974) 346, 352, and map on p. 357, finds it in Pan·
nonia, for Isis inscriptions; Malaise (1972) 350, too, for Isiacism in Italy. He infers
traders spread the cult.
7. Vidman (1966) 113 points that, aside from a single example in Lambaesis and
another in Britain, there are no Isiac temples in military contexts anywhere in the
west (in the east, the picture changes slightly, e.g., CIL 3.1342, to the goddess by an
ala I Hi11p. Campagonum, and see following note).
8. The Isis temple (VIII 7, 28 on the map) was never large, though eventually
ornate. On the slave population in Italy (or a fair guess), see Hopkins (1978) 9; on
the class of Isis's congregation, d. Castren (1975) 115n.8and207. Note also Malaise
(1972) 76f. for Italy broadly, with ingenui totaling three in the first cent., seventeen
in the first/second, and twenty-four in the second/third (p. 77 n.2). Compare Zoto-
vic (1968) 63, finding Isiacs in Dalmatia and Pannonia (esp. Pann. Sup.) to be
largely slaves, and of Greek descent. ·
9. Malaise (1972) 71-73. Malaise, esp. in pp. 26-37, 70-74, and 103, offers very
valuable remarks on method.
10. Malaise (1972) 77 n.2; Vidman (1965) 389, first dating the high point of
Isiacism in the west merely to the third century, and citing only Cumont (1929) 11-
who himself, without specifying Isiacism or offering any substantiation, speaks of
the apogee of the "Oriental" cults in that period. But Vidman then turns to "base
himself above all on inscriptions," finding their apogee under Caracalla and (p.
394) thereafter a sharp decline. On their chronological distribution, see Mrozek
(1973) 114-116, in an article of exceptional value. Mrozek 115 n.2 gently points out
how his results invalidate certain findings in Roman religious history. His findings
received confirmation in the same year he offered them. Lassere (1973) 133-151
offered approximate dates for pagan epitaphs from a number of major sites in
Africa Proconsularis and Numidia. The great bulk of his examples, roughly 4,000
texts, can be arranged into centuries (his own categories, such as "end of the Repub·
lie to Trajan," "second/third century," do not lend themselves to a table or graph):
they are in the proportions 1:3:2 for the periods A.D. 1-100, 101-200, and201-"mid-
third cent." ("milieu du Ille siecle," meaning evidently not exactly A.D. 250 but, let
us say, 240-260). The high point is Severan (texts dated "end of second or third
century," "end of second/beginning of third century," and "Severan" total ca.
1250). After "mid-third century" there are only seven out of the whole 4,0001
200 Notes to Pages 116-117

11. Toutain (1907-1920) 1,2.34, colliding with a statement handiest in Cumont

(1929) 81. Toutain's work seems to me admirable. In all my reading, however, I
recall seeing it cited only by Frend (1952)-himself the best of judges. Cumont's
work, on the other hand, entirely dominates the study of religion in Roman
antiquity. I have offered elsewhere (above, chap. 1.3 nn.I9 and 28; in chap. 2.1
nn. 15, 18, 36, and 55; in chap. 2.2 n.9; in chap. 2.3 n.39; below, nn.l9, 24, and 26) a
sampling of points which suggest how little trust should, in my opinion, be placed
in Cumont's statements. The collision between him and Toutain was renewed:
Toutain (1907-1920) 1,3.5 and 10 against Cumont (1912) 127, regarding the accept·
ance of Isiacism in North Africa. Charles-Picard (1954) 225 attempted to correct
Toutain: there is evidence for the cult in Africa. True-but nothing to invalidate
Toutain's method, heavily relying on inscriptions as opposed to literary texts; and
even in 1954, there was not enough new material seriously to modify conclusions
based on the data available in 1920. The indices of GIL 8 yield only two inscriptions
(excluding, as I do from this book, Mauretania).
12. Hatt (1978) 277 defends the importance of "Oriental" cults in Gaul, but
means mostly Cybele and Mithra. On Lyon, see LeG lay (1972) 13: out of I, 150 texts,
306 are of Greek-speaking origin, all texts being almost exclusively of the second
and third centuries. In GIL 13, in Gaul, I count seven Isiac texts, of which one comes
from Lyon, the rest from Belgica and Germany.
13. For distribution of inscriptions according to sex, notice those of the Mainz
area which give the age of the deceased: 186 men to 15 women, in Szilagyi (1961)
128f. In GIL 13, in compiling the figures offered in the text, I count under Mithra the
dedications to Cautes, Sol, and Invictus; under Isis, also Sarapis; under Cybele, also
Mater Deum, Magna Mater, and Attis. In the same volume, nos. 1-5192 and 11000-
11600 are from the interior; 5193-8860 and ll601-12085 are from the limes, some-
what arbitrarily defined (the omitted nos. are milestones, graffiti, and on ceramic
material). Of all, 40 percent or more are from the frontier. To show that Cybele
worshipers were "old-Italian," observe that in the western and southern peninsula,
where there was the least moving in and out of population, this cult shows its
largest share among the several "Oriental"-Malaise (1972) 460 lists 37 inscriptions
for Isis, 8 for Dolichenus, 17 for Mithra, but 80 for Cybele. Quite different in
Venetia: the same four cults produce figures of 44, 5, 25, and 20. The same contrast
appears in Dalmatia and Pannonia: Cybele-worshipers are much more likely to be
Roman citizens than worshipers of other "Oriental" cults. See Zotovic (1968) 62.
14. In GIL 5, where I add Jupiter to the deities listed in the table (above, p. 6),
and where I count Dolichenus, Cybele, Isis, and Mithra as "Oriental," the numbers
(out of 500 all told) are 394 men honoring Greco-Roman names as against 55
women; 50 men honoring "Oriental" gods as against 1 woman (her memorial
coming from Milan, where the men's ratio is 15:8, Greco-Roman and "Oriental,"
i.e. 60 percent of the men favor "exotic" deities, not I0 percent as in the area overall).
15. For example, in Merlat (1951) passim, 123 inscriptions attest Jupiter Doli·
chen us: 113 male (35 being soldiers or veterans) with 10 women (6 ofthem as wives).
In Dacia, GIL 3. 787-1627 and 7624-8060 yield 69 dedications in which the sex of the
dedicant is dear and in which the deities are "Oriental" (Mithra, Isis, Cybele,
Dolichen us, Dea Syria, and Malagbel ). Of these 69, 64 are men (42 for Mithra alone),
The women worship Isis and Cybele. The percentages arrived at by L. W. Jones
 Notes to Pages 117-119 201

(1929) 298, comparing men's and women's attachment to "Oriental" or "Italian"

deities, are perhaps credible for men, but I suspect that the sampling for women
skews the truth through being too small. It shows a positive preference for the
"Oriental," 36.6 percent to 31.7 percent.
16. On Mainz, see above, n.l3, and Clauss ( 1973) 399, who compares distribution
among epitaphs also in other towns (ibid.): 73.4 percent of all that give age at death
are of soldiers and kin in Carnuntum, 60 percent in Viminacium, etc. At Singi·
dunum, city and territory, half of some 62 inscriptions are of soldiers and veterans,
cf. Mirkovic and Dusanic (1976) 39. Mocsy (1970) 167 estimates troops in Moesia
Superior, including veterans, at less than 5 percent of the total population but they
account for 25 percent of the inscriptions ( 13-60 percent in different districts). How
such maldistribution might affect our picture of a cult may be guessed from com·
paring these figures with others, for Dolichenus, in chap. 1.1 n.29, above. Note, too
(Mocsy [1970] p. 208) that 92 percent of all inscriptions come from 20percentof the
area of the province (or 80 percent come from 8 percent of the areal).
17. Mocsy (1970) 193; Toutain (1907-1920) 1,3.445, on 24 out of 520 votive texts
from Gaul; cf. Drexel ( 1922) 25 describing the Roman and Celtic element in Rhaetia
as "the very thinnest crust," covering native deities not one of whose names we
know from inscriptions; and (p. 24) we are similarly in the dark about religion in
the Rhine provinces. Etienne et al. ( 1976) 98 offer a revealing figure: at Conimbriga,
in Latin inscriptions on stone, one-third of the cognomina are native; but once one
turns to those inscribed on clay, the figure rises to three-fourths.
18. For the dates outlining the Do1ichenus cult, cf. Merlat (1951) passim; I. Toth
(1971) 23; Malaise (1972) 461; and Speidel (1978) 4, 9f., and 73. Fitz (1959) 259 more
than once speaks of the cult as being "offizielle," thanks to Sept. Severus. The view
and terminology, though conventional, are mistaken. See above pp. 103£. and
chap. 2.4 n.70.
19. Cumont (1939)69=Vermaseren (1956-1960) 1.51 no. 23, which Gordon (1975)
229 and 231 n.64 oddly dates to 78 B.c.; Gough (1952) 131, an inscription which I
judge to be no earlier than A.D. 175; ibid. for the Tarsian coins of A.D. 238, surely
indicating the special celebrity of Mithraism in the city; and Plut., Pomp. 24. Cf.
also the Hadrianic essayists on the cult referred to by Porph., De abst. 2.56. Gordon
(1975) 237f. persuades me that Dio Chrys., Or. 36.39£., shows no knowledge of
Mithraism. But the cult is better known ca. A.D. 160, when Justin assumes that an
eastern audience will have heard of its cult meals, pial. 70and1Apol. 66. Cumont's
views on the connections between Mithraism and Persia, and hence the meaning of
"Persian" applied to the cult, are in process of dissolution. See, e.g., Wikander
(1950) 6-9, 14, 19, and 36; Hinnells (1975) 303£.; or Beskow (1978) 7£.
·20. Stat., Thebaid 1.717; Swoboda (1964) 193; Vorbeck and Beckel (1973) 14 and
29.
21. Vermaseren (1956-1960) 2.356 no. 2269 (Moesia), dated in Beskow (1978) 12;
Schwertheim (1979) 27 (Heddernheim);and CIL 6.718=Cumont (1896-1899) 1.245,
cf. PIR 2 s.v. Tib. lui. Cl. Livianus (Rome).
22. Becatti (1954) 35£.; Vermaseren (1956-1960) vol. 1 nos. 216, 285£. and 362.
23. For ratios of inscriptions, the Mithraic compared to the whole corpus, I
exclude CIL 4 (graffiti), producing Italy minus Rome, about 38,000; Rome, about
40,000; the rest, about 70,000. Among Mithraic inscriptions, I give numbers where
202 Notes to Pages 119-121

Gordon (1972) 103 gives percentages (except on p. 109, where he discusses Greek
names at Ostia and Rome). We differ inconsequentially. I count, in Vermaseren
(1956-1960) vols. 1 and 2, every person connected with the cult, whether or not
doubtful, except where the names occ.ur in inscribed lists of dedicants (of which lists
there are two in Italy and one to the north), or where the text has lacunae that might
have contained indications of military status, active or retired. I accept questionable
texts in the collection, including "Sol" texts, some of which are possibly, some cer-
tainly, not Mithraic. In Africa, Mithra is as unknown as in the east, save for the
Lambaesis area, as Charles-Picard (1954) 222 pointed out. On the high proportion
of soldiers in the epigraphy of the frontier, see above, nn.l3 and 16. Bivar(l975) 281
is right to say that Mithraism is conventionally viewed as specially appealing to the
military; but the view itself is unfounded.
24. SHA Commod. 9-a wild paragraph in an unreliable author. Cumont(l929)
139 offers no substantiation for saying "the conversion[!] caused an immense stir."
The phrase was repeated verbatim from Cumont (1896-1899) 1.277 and a third time
in Cumont (1902) 70. So truth becomes established, d., e.g., Ferguson (1970) 48,
Commodus's initiation "led to the first great period of the cult." For the prefect
mentioned, L. Flavius Aper in CIL 3.15156, see Hoffiller and Saria (1938) 1.146£. and
149; and proportions of high officers, etc., are noted by Gordon (1972) 103f.
25. Packer (1967) 86 offers ca. 27,000 as the max. pop. for Ostia (including unex·
cavated parts; earlier estimates are double that); but density estimates in Packer 86
n.55 are to be modified upward by Frier (1977) 28£. The Mithraea were generally
small, mean structures, holding less than twenty up to a maximum of forty, d.
Meiggs (1960) 372. Not all fifteen were certainly in use at any one time. Schwertheim
(1979) 57 mentions an average capacity of 20-30, counting larger examples in
Rome. For crowds at temple gates, no more than a suggestive picture, d. Herondas,
Mimiambi 4.54, cited in Edelstein and Edelstein (1945) 1.273.
26. Cumont (1929) 130 calls Mithraism a more and more pressing presence in the
second century, "remaining until the end of the 4th century the most important cult
in paganism." His view might still receive general support. Discouraging.
27. M. P. Nilsson's suggestion is quoted as a fact by Nock (1958a) 292-with
which Wikander (1950) 44 more or less agrees. Cumont (1975) 205 pictures Mith·
raism "finally coming under the control of the State and the emperors," so that he
can explain the wide dispersal of the cult, e.g., to the Dura chapel, without having
to admit a native eastern branch; but d. above, on "state" cults. I withhold my own
inexpert ideas about pre-100 Mithraism, not to add to inconclusive debate.
28. Above, at chap. 1.4 n.l6; d. theuj.lVl']tpia in the Demeter mysteries, Ohlemutz
(1940) 221.
29. Paus. 8.2.5, complaining of evil days as the reason why no new gods come
into being from men. The passage happens to coincide with, and contradict, CIG
5980=1G 14.966=SIG2 807, which reports a miracle in Rome, bystanders rejoicing
"because the powers were living then in the reign of our emperor Antoninus."
Porphyry in Euseb., Praep. ev. 5.1 (179df.) speaks of the declining presence of the
gods, but in a polemical setting.
30. Cumont ( 1929) 38-a sort ofrephrasing ofthe "failure-of-nerve" interpreta·
tion. Gordon (1972) 93£. rightly puts the latter interpretation aside, speaking of
Dodds (but there are others behind Dodds, e.g., Gilbert Murray). For similargener·
 Notes to Pages 12J-127 20J

alizing about spiritual sturdiness, compare A.H.M. Jones (1963) 19: "the peasant
masses, who were made of tougher stuff than the townsmen," clung to Christianity
under the Tetrarchs.
31. Cumont (1929) 39~above, p. 64, and similarly in Laurin (1954) 8.
32. In works demonstrating the most admirable scholarship and powers in phi·
lology, one may find statements that no historian would dream of making. Drawing
from major authorities, throughout this chapter's notes, I have scattered examples
which it would be invidious to recall at this point.
33. Gordon (1972) 94.
34. Gordon (1972) assumes Mithraism's exclusion of women, special connection
with soldiers, moral dualism and notions of sin, derivation from Iran and many
other features contrary to my understanding of the facts or wholly unsubstantiated;
but, in addition, I cannot accept his main point, that "the departmental boss," slave
owners, "various superiors," "senior officers" (pp. 107f. and 112)-in short, the
ruling class-encouraged the cult as a "self-conscious piece of opportunism"
through "conscious realization" of its usefulness (pp. 110 and 112). That implies a
manipulative conspiracy among upper-class worshipers supported by no evidence
and, I would say, by no likelihood.
35. Vermaseren and van Essen (1965) 232.
36. Vermaseren (1971) 3, starry vault; Cumont (1975) 167f., gilt and jewels, and
curtain; Vermaseren (1956-1960) 2.24 no. 230, ocular in vault; rayed hole, in Bull
(1978) 78; pivoted relief, Turcan(l978) 153andSchwertheim(l979)27 Abb. 32f.; use
of paint or gilt on idols and reliefs, Vermaseren (1965-1960) 1.128 no. 258; 239no.
641; etc.
37. Altars, in Becatti (1954) 13; Vermaseren (1956-1960) 1.126no. 253and 170no.
392; Wortmann (1969) 411-421, full discussion of lighting tricks; and Schwertheim
(1979) 34 Abb. 49.
38. Schwertheim (1979) 29 Abb. 38f. and p. 73; on apparent victims in initiatory
shows, combined with literary texts, cf. Vermaseren (1971) 26-29, and above, chap.
2.1 n.24. E. D. Francis, adding a note toCumont(1975)206n.!H2, supports the view
that the rites were dumb show, not recitations of doctrine.
39. Tert., De corona 15; SHA Commod. 9.6; and compare the use in Dionysiac
initiations of "spooks and horrors," Orig., C. Cels. 4.10.
40. Note, for example, the elaborate underground facilities for miming and
initiation at Eleusis, or at Claros, Robert (1953a) 16-20. As L. Deubner (1932) 86
puts it, "Zauberritus and Gotterritus steht im Kulte sehr hliufig nebeneinander."
On Hippolytus's passage on theatrical effects in cult rites, see Refut. 4.28-41 and
6.34-36, supposed by Ganschinietz (1913) 22f. to be drawn from some magician's
handbook that Hippolytus had before him. For a Gallic equivalent of Alexander of
Abonuteichus, or, perhaps better, of Simon Magus, see Polemo in R. Foerster's
Latin translation of the Arabic, Scriptores physiognomonici graeci et latini (1893)
1.162.
41. Liebeschuetz (1979) 244 and n.3 on the small Tetrarchic pantheon, compared
with the large one, Vesp.·Hadrian. But continue the series: Ant. Pius dropped
Bacchus, Castor, the Dioscuri, Mercury, Neptune, Saturn, Sarapis, Sol, Tell us, and
Vulcan-from lethargy?
42. Bosch (1935) 170; also above, chap. 1.2 n.71. ForthegodMenalone,oncoins,
204 Notes to Pages 127-132

Lane (1975-1978) vol. 2 passim has collected the data: 64 cities minting, over 90
percent of their issues datable in the periods A.D. 1-50 (six issues that show Men);
50-100 (12); 100-150 (28); 150-200 (67); 200-250 (184, a great rise); 250-300 (55,
largely in the 250s only). But the datable coins of Brit. Mus. Lycia, Pamphylia, and
Pisidia in the same six half-centuries run 17, 13, 30, 67, 231, 177. Hence the two
curves, for Men and for all coins (of one region), differ (i.e., they tell us something
about Men) only in the third century, esp. 235 on, since in both curves the Severan
rulers account for the identical 28 percent.
43. Geffcken (1920) 15-needing no significant updating; on Vesta, see Laue
(1960) 360, the acta full until A.D. 241, the last frg. dated to 304, and taurobolia are
frequent until241, thereafter only 2 in the century; Vestal records (ibid.) are full up
to ca. 255. On Saturn, see Frend (1952) 76, noting the abrupt falling-off in ex-votos
and temple building after 235, and modifications of this picture by LeG lay ( 1966),
accepting the decline, pp. 96f., but insisting (without really very much data, pp.
101£.) on the cult's continued survival; for Jupiter Dolichenus, see Zotovic (1968) 73.
44. Jalabert (1908) 719: pagan to A.D. 250; Christian after 324. On coin distribu-
tion across time, d. n.42 above; on various other categories of silence or lack of
evidence-on various elite groups, but also on their fine pottery-see MacMullen
(1976) 215 n.l; also 240 n.44, where I report on the dating of 5,000 papyri and
ostraca. Their numbers first diminish sharply at the accession of Sept. Severus, fall
off a second big step (say, by a half, from the second-cent. average) in A.D. 225, and a
third, in 300.
45. Above, chap. 2.4 n.56, on Greece and other impoverished points; on bloom·
ing areas even in bad times, see MacMullen (1976) 119 and notes.
46. Above, n.44; below, Epilogue, n.8; Frend (1952) 76; Geffcken (1920) 20f.,
appealing to nonepigraphic testimonies.
47. Mrozek (1973) 118.
48. The angle of attack chosen by the later, heavier persecutions is obvious. Gage
(1964) 327 draws out some of its chief features.
49. Ox prices, approximate, are based on Frank (1933-1940) l.l89, wages, and
200, oxen, in the second cent. B.c.; Sperber (1966) 184 and 188 in later second-cent.
A.D. Egypt and Palestine.
50. MacMullen (l966a) 355f., on eastern native cults, to which add the evidence
on Men, above, n.42, and the interesting statistics in Keil (1923) 253-266, esp. 266,
on pre-Greek Lydian cults; and MacMullen (l966a) 36If., on northern native cults,
to which add Balla (1967) 73, on reviving Illyric-Celtic cults from the later second
century.

EPILOGUE
1. Euseb., Praep. ev. 4.2 ( l35c-136a) and H ist. eccl. 9.11.6 on the pagan leader-
ship, hierophants, magistrates, and intellectuals of Antioch and Miletus.
2. To A.H.M. Jones (1964) 938-943 and refs. in 1391 nn.7-l3, there is little to
add from literary sources; but, on Shenute' s campaigns, d. passages in Kaegi (1966)
255£., and, on compulsion of peasants, MacMullen (1980) 28. Among inscriptions,
Geffcken notices Forschungen in Ephesos i.l03.
3. As to the army, its constituents before A.D. 312 of course were pagan, and "its
 Notes to Pages 1J2-JJ5 205

intake must have remained predominantly pagan" throughout the fourth cent.-
so says A.H.M. Jones (1968) 24. CT 7.20.2 receives more weight than it deserves
(ibid.), since it gives the shouts, dei te nobis seroent, of only one army unit at one
date-both unknown. Jones intends to follow Seeck (who surmises A.D. 826, on the
basis of an emendation); but his confused sentence actually indicates 822. The
fourth-cent. burial practices of soldiers in the West are abundantly attested, but only
show that some were pagan, not how many. The only indications of religion in the
ranks of the competing forces in 894 are Christian-in Theodosius' s army folded in
with Eugenius's, after the Frigidus. Cf. Hoffmann (1968) 24f. and his nos. 11, 14,
and 85.
4. Old inscriptions, like CIL 6.102, 987 and 50=IG 14.1018; new, like AE 1969-
1970 no. 116 and 1975 no. 54; and neglected details, e.g., Macrob. 1.28.18, Antium,
and Maximus of Turin, Sermo 72.2.42 (for the date, cf. Mutzenbecher in Corp.
Christ. Lat. p. XXXV n.J ),
5. E.g., CIL 8.6975; 12272; and 24521.
6. Cf. texts like Aug., Ep. 16£. and Civ. dei 7.26. On the strength of fourth cent.
African paganism in general, see Warmington (1954) 85£. and above, chap. 2.5 n.lO.
On Saturn cult, against Frend (1952) 88f., I would set LeG1ay (1966) 101£. and
Frend' s own picture (pp. 61-64) of the sharp decline of the classes and centers which
were most likely to produce records and buildings of any sort. The existence of a
church in a town or village, here in Africa or anywhere else, of course does not prove
Christian predominance. Christians and pagans everywhere coexisted. That is
easily shown to be the rule-with noted exceptions.
7. ILS 4197 (A.D. 860s) and Paneg. vet. 12(2).87.4 (A.D. 888).
8. Pautalia (in Moesia, later Dacia Mediterranea), in Hoddinott (1975) 182.
Mocsy (1974) 824 offers more (but undatable) examples of vandalized shrines and
statues; also, p. 828, some interesting speculation on the epigraphic evidence, with·
in which, in the earlier 300s, pagan epitaphs vanish and Christian begin to appear.
For the burning and smashing of pagan temples and images in the upper Danube
areas, see Alfoldy (1974) 210f., A.D. 380 and later, and Stlihelin (1948) 580, A.D. 840s
and later.
9. Frantz (1965) 191-197; Edelstein and Edelstein (1945) l.323f. and 423f.; 2.257;
and IG 42, 1.438. On the rest of the east, cf. A. H. M. Jones, cit. above, n.2, to which I
can add only Zos. 1.38 and 4.18; Expositio tot. mundi 22, 30, and 34£.; and the in·
teresting circle of pagan speculation revealed in PHermopolis 2-4. On pagaq cults
vigorous specifically in Egypt, see Hanslik (1954) 180£. and Remondon (1952)
63-78.
10. Petit (1955) 200£.; on state functionaries, p. 202: in A.D. 360, in the east,
8 Christians, 15 pagans (and after Julian, far more paganized).
11. Bell et al. (1962) 20 and papyri nos. 4-8 etc. (Christian) and 36 (pagan oath
formula-e£. POslo 113 of A.D. 346).
12. Marc. Diac., Vita Porph. 41 (Arcadius's speech), and miracles at 19, 81, 61,
and 66f., with conversions, 21; for other mass rollversions produced. by miracles see
Theod., Hist. relig. 17 (PG 82.1421-1423) and Soz. 5.15.14, comparing above, chap.
2.4 n.l.
U. Brown (1964) 109.
14. For example, Scroggs (1980) 170-175 and 177-179 provides recent handy
206 Notes to Pages 1Jj-JJ7

access to a long tradition of writing, "bourgeois" and Marxist, on the socioeco-

nomic orientation of the Church. But discussion of what life and views the convert
would find in the Church does not seem to me very useful in explaining the growth
of the Church, unless one supposes that many converts (in the sense of persons now
believing that the Christian God really existed and really was supreme) would
promptly have fallen away if they had not found other essential things that held
them unshakably in the faith.
15. A.H.M. Jones (1964) 938-943 collects the legal texts.
16. Adolf von Harnack in his Mission und Ausbreitung des Christentums(l902
and later edns.) rightly claimed that his was the first work devoted to the subject of
his title; and so far as I know it is the last-certainly still standard. It is justly ad-
mired for its scholarship. Among its thousands of references to sources, however, I
can find not one to a pagan source and hardly~ line indicating the least attempt to
find out what non-Christians thought and believed. Thus to ignore the prior views
of converts or to depict the Mission as operating on a clean slate is bound to strike an
historian as very odd indeed.
 Bibliography of
Secondary Works Cited

Agache (1973)-Ag ache, R., "Quelques fana reperees par avion en Picardie et en
Artois," Caesarodunum 8 (1973) 50-55.
Albertini (1931)-Alb ertini, E., "Inscriptio ns d'El-Kantara et de Ia region," Rev.
afr. 72 (1931) 193-261.
a
Albertini (1943)-Alb ertini, E., "Une nouvelle basilique civile Cuicul (Djemila),"
CRAII943, 376-395.
A. Alfoldi (1948)-Alf oldi, A., The Conversion of Constantine and Pagan Rome,
Oxford 1948.
A. Alfoldi (1963)-Alf oldi, A., Early Rome and the Latins, Ann Arbor 1963.
A. Alfoldi (1967)-Alf oldi, A., Studien zur Geschichte der Weltkrise des 3. ]ahr-
hunderts nach Christus, Darmstadt 1967.
M. R. Alfoldi (1963)-Alf oldi, M. R., Die Constantinische Goldpriigung. Unter·
suchungen zu ihrer Bedeutung fur Kaiserpolitik und Hof-Kunst, Mainz 1963.
Alfoldy (1961)-Alf oldy, G., "Geschichte des religiosen Lebens in Aquincum ,"
Acta arch. acad. sci. Hung. 13 (1961) 103-124.
Alfoldy (1974)-Alfo1dy, G., Noricum, London and Boston 1974.
Amelung (1905)-Am e1ung, W., "Ex-voto an Ask1epios," Archiv fur Religions-
wissenschaft 8 (1905) 157-160.
Anderson (1913)-An derson, J.G.C., "Festivals of Men Askaenos in the Roman
colonia at Antioch of Pisidia," ]RS 3 (1913) 267-300.
Anderson (1937)-An derson, J.G.C., "An imperial estate in Galatia," ]RS 27
(1937) 18-21.
Andreae (1963)-An dreae, B., Studien zur romischen Grabkunst (Mitt. des Deutsch·
en Arch. lost., Rom. Abt., Erganzungsheft 9), Heidelberg 1963.
Andresen (1955)-An dresen, C., Logos und Nomos. Die Polemik des Kelsos
wider das Christentum, Berlin 1955.
Angyal (1971)-Angya1, K. B., "Epigraphi ca. Contributio n al'etude historiqued es
religions orientales de Dade," Studium 2 (1971) 17-21.
Archambault (1909)-Arc hambault, G., ed., Justin. Dialogue avec Tryphon, Paris
1909.
Arnheim (1972)-Arn heim, M. T. W., The Senatorial Aristocracy in the Later Ro-
man Empire, Oxford 1972.
Arnim (1921)-Ar nim, H. von, Plutarch uber Diimonen und Mantik (Verhandel·
ingen der Koninklijke Akademie van Wetenschappen te Amsterdam, Afdeeling
Letterkunde 22, 2), Amsterdam 1921.

207
208 Bibliography

Attridge (1978)-Attridge, H. W., "The philosophical critique of religion under

the early Empire," ANRW II, 16, 1 (Berlin 1978) 45-78.
Audollent (1890)-Audollent, A., "Mission epigraphique en Algerie," MEFR lO
(1890) 397-588.
Audollent (1901)-Audollent, A., Carthage romaine, 146 avant ]esus-Christ-698
apres jesus-Christ (Bibl. des Ecoles Fr. d'Athenes et de Rome 84), Paris 1901.
Audollent (1904)-Audollent, A., Dejixionum tabellae quotquot innotuerunt tam
in Graecis Orientis quam in totius Occidentis partibus praeter Atticas in 'Cor-
pus Inscriptionum Atticarum' editas, Paris 1904.
a
Aymard (1951)-Aymard, J., Essai sur les chasses romaines, des origines lajindu
siecle des Antonins, Paris 1951.
Balla (1967)-Balla, L., "Zur Geschichte des religiosen Lebens von Savaria," Acta
class. univ. sci. Debrecen. 3 (1967) 67-76.
Balla (1971)-Balla, L., "Possessores et vicani vici Vindoniani," Debreceni Veri
Muzeum Evkonyve 1971, 55-63 (Fr. summary, p. 64).
Barb (1963)-Barb, A. A., "The survival of magic arts," in The Conflict Between
Paganism and Christianity in the Fourth Century, ed. A. Momigliano, Oxford
1963, 100-125.
Bardenhewer (1912-1932)-Bardenhewer, 0., Geschichte der altkirchlichen Li-
teratur'l, 5 vo1s., Freiburg im Breisgau 1912-1932.
Bardon (1956)- Bardon, H., La litterature latine inconnue, 2: L 'epoque imperiale,
Paris 1956.
Bardy (1948)-Bardy, G., Theophile d'Antioche. Trois livres ti Autolycus, Paris
1948.
Bark6czy et al. (1954-1957)-Barkoczy, L., et al., Intercisa (Dunapentele-Szttilin-
vtiros). Geschichte der Stadt in der Romerzeit, 2 vols., Darmstadt 1954-1957.
Barr (1974)-Barr, J., "Philo of Byblos and his 'Phoenician History,'" Bull. fohn
Rylands Lib. 57 (1974) 17-68.
Bataille (1952)-Bataille, A., Les Memnonia. Recherches de papyrologie et d'epi-
graphie grecques sur la necropole de la Thebes d'Egypte aux epoques helle-
nistique et romaine, Le Caire 1952.
Baur (1902)-Baur, P. V.C., Eileithyia, Columbia (Missouri) 1902.
Baynes (1931)-Baynes, N. H., Constantine the Great and the Christian Church,
Oxford 1931.
a
Beaujeu (1955)-Beaujeu, J., La religion romaine Npoque de l'empire, 1: La
politique religieuse des Antonins (96-192), Paris 1955.
Beaujeu (1966)-Beaujeu, J., "Religion in the second century A.D.," Bucknell Re-
view 14, 2 (1966) 1-18.
Becatti (1954)-Becatti, C., I Mitrei (Scavi di Ostia 2), Rome 1954.
Behr (1968)-Behr, C. A., Aelius Aristides and the Sacred Tales, Amsterdam 1968.
Bell et al. (1962)-Bell, H. 1., et al., The Abinnaeus Archive, Oxford 1962.
Benabou (1976)-Benabou, M., "Resistance et romanisation en Afrique du Nord
sous le Haut-Empire,'' in Assimilations et resistance ala culture greco-romaine
dans le monde ancien. Travaux du VI Congres international de la Federation
Internationale des Associations d'etudes classiques . .. I974, Paris 1976, 367-
375.
 Bibliography 209

Bernand (1969)- Bernand, A., Les inscriptions grecques de Philae .. . , 2 vols., Paris
1969.
Bernand (1972)- Bernand, A., Le Paneion d' El Kana is. Les inscriptions grecques,
Leiden 1972.
Beschaouch (1975)-Beschaouch, A., "A propos de recentes decouvertes epigra·
phiques dans le pays de Carthage," CRAI1975, 101-118.
Beskow (1978)-Beskow, P., "The routes of early Mithraism," in Etudes Mithri-
aques. Actes du 2e Congres international . .. 1975, Leiden 1978, 7-18.
Besset (1901)-Besset, A., "Inscription d'Asie Mineure," BCH 25 (1901) 325-336.
a
Bessou (1978)-Bessou, M., "Le £anum de Camp-Ferrus Loubers (Tarn)," Gallia
36 (1978) 187-212.
Betz (1961)-Betz, H. D., Lukian von Samosata unddasNeue Testament. Religions·
geschichtliche und pariinetische Parallelen ... (Texte und Untersuchungen
76), Berlin 1961.
Bickerman (1973)-Bickerman, E., "Consecratio," in Le culte des souverains dans
/'empire romain (Entretiens sur l'antiquire classique 19, Fondation Hardt),
Geneva 1973, 3-25.
Bidez (1913)-Bidez, J., Vie de Porphyre, le philosophe neo-platonicien, avec les
fragments des traites flep\ dyaA.J.Uitcov et De regressu animae, Gand and Leip·
zig 1913.
Bieber (1910)-Bieber, M., "Attische Reliefs in Cassel," Ath. Mitt. 35 (1910) 1-16.
Bieler (1935-1936)-Bieler, L., 0EIOt ANHR. Das Bild des 'gattlichen Menschen'
in Spiitantike und Friihchristentum, 2 vo1s., Vienna 1935-1936.
Binsfeld (1969)-Binsfeld, W., "Das Quellenheiligtum Wallenborn bei Hecken-
miinster (Kr. Wittlich)," Trier. Zeitschr. 32 (1969) 239-268.
Birley (1974)-Birley, E., "Cohors I Tungrorum and the oracle of the Clarian
Apollo," Chiron 4 (1974) 511-513.
Birley (1978)-Birley, E., "The religion of the Roman army: 1895-1977," ANRW
II 16, 2 (Berlin 1978) 1506-1541.
Bivar (1975)-Bivar, A.D. H., "Mithra and Mesopotamia," in Mithraic Studies.
Proceedings of the First International Congress of Mithraic Studies, ed. J. R.
Hinnells, Manchester 1975, 275-289.
Blinkenberg (1941)-Blinkenberg, C., Lindos. Fouilleset recherches 1902-1914, II:
Inscriptions, 1-11, Copenhagen 1941.
Boehringer (1959)-Boehringer, E., "Pergamon," in Neue deutsche Ausgrabungen
im Mittelmeergebiet und im vorderen Orient, Berlin 1959, 121-171.
Bogaers (1955)-Bogaers, J.E.A. T., De gallo-romeinse tempels te Elst in de Over-
Betuwe, The Hague 1955.
Bookidis (1973)-Bookidis, N., "The sanctuary of Demeter and Kore in Corinth.
Excavations 1970-1972," A]A 77 (1973) 206-207.
Boulanger (1923)-Boulanger, A., Aelius Aristide et la sophistique dans la province
d'Asie au ll1 siecle de notre ere, Paris 1923.
Boyce (1937)-Boyce, G. K., "A corpus of the lararia of Pompeii," MAAR 14 (1937)
5-112.
Bosch (1935 )-Bosch, C., Die kleinasiatischen Miinzen der romischen Kaiserzeit (II,
Einzeluntersuchungen; 1, Bithynien), Stuttgart 1935.
210 Bibliography

Brenk (1973)-Brenk, F. E.," 'A most strange doctrine': Daimon in Plutarch," CI

69 (1973) 1-11.
Brilliant (1963)-Brilliant, R., Gesture and Rank in Roman Art. The Use of Ges-
tures to Denote Status in Roman Sculpture and Coinage, Copenhagen 1963.
Brommer (1973)-Brommer, F., Der Gott Vulkan auf provinz.ialromischen Reliefs,
Cologne and Vienna 1973.
Brown (1964)-Brown, P.R.L., "St. Augustine's attitude to religious coercion,"
IRS 54 (1964) 107-116.
Brown (1971)-Brown, P., "The rise and function of the holy man in Late An-
tiquity," IRS 61 (1971) 80-101.
Brown (1972)-Brown, P., Religion and Society in the Age of Saint Augustine,
London 1972.
Bruhl (1953 )-Bruhl, A., Liber Pater. Origines et expansion du culte dionysiaque a
Rome et dans le monde romain, Paris 1953.
Buckler (1914-1916)-Buckler, W. H., "Some Lydian propitiatory inscriptions,"
BSA 21 (1914-1916) 169-183.
Buckler (1933)-Buckler, W. H., "A pagan recantation," CR 47 (1933) 7-8.
Buckler et al. (1926)-Buckler, W. H., et al., "Monuments from central Phrygia,"
IRS 16 (1926) 53-94.
Buckler and Robinson (1932)-Buckler, W. H., and D. M. Robinson, Sardis VII:
Greek and Latin Inscriptions, Part I, Leiden 1932.
Biicheler ( 1865)-Biicheler, F., "Philodemos 1ttpi eliaepeia~," N Ibb. fiir Philologie
9 (1865) 513-541.
Buecheler (1895-1897)-Buecheler, F., Carmina latina epigraphica, 2 vols., Leip-
zig 1895-1897.
Buffiere (1962)-Buffiere, F., Heraclite, Allegories d'Homere, texte et traduction,
Paris 1962.
Bull (1978)-Bull, R. 1., "The Mithraeum at Caesarea Maritima," in Etudes M ith-
riaques. Actes du 2e Congres international . .. 1975, Leiden 1978, 75-90.
Butler (1915)-Butler, H. C., Ancient Architecture in Syria (Publications of the
Princeton University Archaeological Expeditions to Syria in 1904-05 and 1909,
II A, Part 5), Hauran Plain and Djebel Hauran, Leiden 1915.
Cagnat (1898)-Cagnat, R., Cours d'epigraphie latine!, Paris 1898.
Calder (1935)-Calder, W. M., "Silius Italicus in Asia," CR 49 (1935) 216-217.
Cameron (1939)-Cameron, A., "Inscriptions relating to sacral manumission and
confession," HThR 32 (1939) 143-179.
Canarache et al. (1963)-Canarache, V., et al., "Le depc)t des monuments sculptu-
a
raux recemment decouverts Constanta," in Acta antiqua philippopolitana.
Studia archaeologica, Sophia 1963, 133-152.
Capello (1941 )-Capello, C. F., "Una stipe votiva d'etarornana sui Monte Genevris
(Alpi Cozie)," Rivista Ingauna e lntemelia 7 (1941) 96-137.
Caputo (1977)-Caputo, G., "Sincretismo religioso ed espressione figurativa in
Tripolitania," Quademi di archeologia della Libia 9 (1977) 119-124.
Caquot (1954-1957)-Caquot, A., "Quelques nouvelles donnees palmyreniennes,"
Comptes rendus du groupe linguistique d' etudes chamito-semitiques 7 (1954-
1957) 77-78.
Cardauns (1958)-Cardauns, B., Varros Logistoricus iiber die Gotterverehrung (Curio
 Bibliography 211

de cultu deorum). Ausgabe und Erkliirung der Fragmente, Wiirzburg 1958.

Cardauns (1978)-Cardauns, B., "Varro und die romischen Religion. Zur Theo-
logie, Wirkungsgeschichte und Leistung der 'Antiquitates Rerum Divina-
rum,"' ANRW II 16, 1 (Berlin 1978) 80-103.
Carton (1908)-Carton, L., "Le sanctuaire de Tanit aEl-Kenissia," Memoires de
l'Acad. des Inscriptions et Belles-Lettres 12, 1 (1908) 1-160.
Casson (1974)-Casson, L., Travel in the Ancient World, London 1974.
Castren (1975 )-Castren, P ., Ordo populusque Pompeianus. Polity and Society in
Roman Pompeii, Rome 1975.
Chadwick (1952)-Chadwick, H., "An Attis from a domestic shrine," ]ThStudies2
3 (1952) 90-92.
Chadwick (1965)-Chadwick, H., Origen: Contra Celsum, translated with an In-
troduction and Notes, Cambridge 1965.
Chadwick (1967)-Chadwick, H., "Philo and the beginnings of Christian thought,"
in The Cambridge History of Later Greek and Early Medieval Philosophy, ed.
A. H. Armstrong, Cambridge 1967, 137-194.
Charles-Picard (1943-1945)-Char1es-Picard, G., "La collection Boglio aSiliana,"
BCTH 1943-1945, 375-381.
Char1es-Picard (1954)-Charles-Picard, G., Les religions de l'Afrique antique,
Paris 1954.
Charpin (1930-1931)-Charpin, L., "Testimonianze cristiane sul Teatro Romano
dell'eta imperiale," Atti, Reale Istituto Veneto 90 (1930-1931) 571-591.
Cherniss (1976)-Cherniss, H., Plutarch's Moralia (Loeb Classical Library), vols.
13, 1-2, Cambridge and London 1976.
Chilton (1967)-Chilton, C. W., Diogenis Oenoandensis fragmenta, Leipzig 1967.
Chilton (1971)-Chilton, C. W., Diogenes of Oenoanda. The Fragments. A Trans-
lation and Commentary, London 1971.
Cintas (1947)-Cintas, P., "Le sanctuaire punique de Sousse," Rev. afr. 91 (1947)
1-80.
Clauss (1973)-Clauss, M., "Probleme der Lebensalterstatistiken auf Grund rO-
mischer Grabinschriften," Chiron 3 (1973) 395-417.
Clerc (1885)-Clerc, M., "Inscription de Nysa," BCH 9 (1885) 124-Ul.
Clermont·Ganneau (1920)-Clermont-Ganneau, C., "Decouverte aJerusalem d'une
synagogue de l'epoque herodienne," Syria 1 (1920) 190-197.
Coarelli (1974)-Coarelli, F., Guida archeologica di Roma, Milan 1974.
Collart (1937)-Collart, P., Philippes, ville de Macedoine, depuis ses origines jus-
qu'a la fin de l'epoque romaine, Paris 1937.
Collart (1944)-Collart, P., "Rejouissances, divertissements et artistes de province
dans l'Egypte romaine," Rev. de philol. 5 18 (1944) 134-152.
Collart and Ducrey (1975)-Collart, P., and P. Ducrey, Philippes I: Les reliefs
rupestres (BCH Suppl. II), Paris 1975.
Collins-Clinton (1977)-Collins-Clinton, J ., A Late Antique Shrine of Liber Pater
at Cosa, Leiden 1977.
Conze and Schuchhardt (1899)-Conze, A., and C. Schuchhardt, "Die Arbeiten zu
Pergamon 1886-1898," Ath. Mitt. 24 (1899) 97-240.
Cook (1903)-Cook, A. B., "Zeus, Jupiter and the Oak," CR 17 (1903) 174-186,268-
278, 408-421.
212 Bibliography

Cook (1914-1940)-Cook, A. B., Zeus. A Study in Ancient Religion, 3 vols., Cam-

bridge 1914-1940.
Corbett (1970)-Corbett, P. E., "Greek temples and Greek worshippers: the literary
and archaeological evidence," Bull. Inst. Class. Studies (Univ. London) 17
(1970) 149-158.
Cousin (1904)-Cousin, G., "Inscriptions du sanctuaire de Zeus Panamaros," BCH
28 (1904) 20-53, 238-262, 345-352.
Crema (1959)-Crema, L., L'architettura romana (Enciclopedia classica VII, xii,
1), Turin 1959.
Cumont (1888)-Cumont, F., "Les dieux etemels des inscriptions latines," Rev.
arch. 11 (1888) 184-192.
Cumont (1896-1899)-Cumont, F., Textes et monuments figures relatifs aux mys-
teres de Mithra 1 (Brussels 1899) and 2 (Brussels 1896).
Cumont (1902)-Cumont, F., Les mysteres de Mithra, 2 Paris 1902.
Cumont (1912)-Cumont, F., "J. Toutain-Les cultes paiens ... .,"Rev. de l'hist.
des religions 66 (1912) 125-129.
Cumont (1913)-Cumont, F., "La theologie solaire du paganisme romain," Me-
moires, Acad. des Inscriptions et Belles-Lettres 12 (1913) 447-479.
Cumont (1915a)-Cumont, F., "Decouvertes nouvelles au mithreum de Saint·
Clement aRome," CRAII915, 203-211.
Cumont (1915b)-Cumont, F., "Les anges du paganisme," Rev. de l'hist. des
religions 72 (1915) 159-182.
Cumont (1917)-Cumont, F., "Gainas le 8&11tVOKpit'1~,'' CRAI1917, 275-284.
Cumont (1920)-Cumont, F., "Groupe de marbre du Zeus Dolichenos," Syria 1
(1920) 183-189.
Cumont (1923a)-Cumont, F., "II sole vindice dei delitti ed il simbolo delle mani
alzate," Memorie I, I (1923), Atti della Pont. Accad. Romana di Archeologia
(Serle III) 65-80.
Cumont (1923b)-Cumont, F., After Life in Roman Paganism. Lectures Delivered
at Yale University ... , New Haven 1923.
Cumont (1926)-Cumont, F., Fouilles de Doura-Europos (1922-1923), Paris
1926.
Cumont (1929)-Cumont, F., Les religions orientales dans le paganisme romain 4,
Paris 1929.
Cumont (1939)-Cumont, F., "Mithra en Asie Mineure," in Anatolian Studies Pre-
sented to William Hepburn Buckler, eds. W. M. Calder and J. Keil, Manchester
1939, 67-76.
Cumont (1942)-Cumont, F., Recherches sur le symbolisme funeraire des Ro-
mains, Paris 1942.
Cumont (1949)-Cumont, F., Lux perpetua, Paris 1949.
Cumont (1975)-Cumont, F., "The Dura Mithraeum," in Mithraic Studies. Pro-
ceedings of the First International Congress of Mithraic Studies, ed. J. R.
Hinnells, Manchester 1975, 155-207.
Dain (1933)-Dain, A., Inscriptions grecques du Musee du Louvre. Les textes in-
edits, Paris 1933.
Daniels (1975)-Daniels, C. M., "The role of the Roman army in the spread and
practice of Mithraism," in Mithraic Studies. Proceedings of the First Interna·
 Bibliography 21J

tional Congress of Mithraic Studies, ed. J. R. Hinnells, Manchester 1975,249-

274.
D'Arms (1972)-D'Arms, J. H., "A new inscribed basefromfourthcenturyPuteoli,"
La parola del passato 21 (1972) 255-270.
De Blois (1976)-De Blois, L., The Policy of the Emperor Gallienus (Studies of the
Dutch Archaeological and Historical Soc. 7), Leiden 1976.
Decharme (1904)-Decharme, P., La critique des traditions religieuses chez les
Grecs, des origines au temps de Plutarque, Paris 1904.
Degrassi (1962)-Degrassi, A., "I culti della Venezia Tridentina," Scritti vari di
antichita II (Rome 1962) 998-1009.
Dell (1893)-Dell, J., "Ausgrabungen in Carnuntum im Jahre 1891," AEM 16
(1893) 156-204.
DenBoeft (1977)-DenBoeft, J., Calcidius on Demons (Commentarius Ch./27-136),
Leiden 1977.
Deonna (1924)-Deonna, W., "Les collections archeologiques et historiques ... ,"
Genava 2 (1924) 29-57.
DeRicci (1934)-DeRicci, S., "Deux nouvelles inscriptions grecques d'Egypte,"
CRAI1934, 256-261.
Deschamps and Cousin (1887)-Deschamps, G., and G. Cousin, "Inscriptions du
temple de Zeus Panamaros," BCH 11 (1887) 873-891.
Deschamps and Cousin (1891)-Deschamps, G., and G. Cousin, "Inscriptions du
temple de Zeus Panamaros," BCH 15 (1891) 169-209.
DesPlaces (1969)-DesPlaces, E., La religion grecque. Dieux, cultes, rites, et senti-
ment religieux dans la Grece antique, Paris 1969.
L. Deubner (1932)-Deubner, L., Attische Feste, Berlin 1982.
0. Deubner (1977-1978)-Deubner, 0., "Das Heiligtum der alexandrinischen
Gottheiten in Pergamon gennant 'Kizil Avli' ('Rote Halle')," Mitt. des Deut-
schen Arch. lnst.,lstanbul21128 (1977-1978) 227-250.
Diehl and Cousin (1887)-Diehl, C., and G. Cousin, "Inscriptions de Lagina,"
BCH 11 (1887) 5-89, 145-163.
Dionigi (1976)-Dionigi, 1., "Lucr. 5, 1198-1203 e P. Oxy. 215 col. I 7-24," Studi
italiani di filol. class. 2 48 (1976) 118-139.
Dodds (1965)-Dodds, E. R., Pagan and Christian in an Age of Anxiety: Some As-
pects of Religious Experience from Marcus Aurelius to Constantine, Cam-
bridge 1965.
Dothan (1968)-Dothan, M., "The representation of Helios in the mosaic of
Hammath-Tiberias," in Tardo antico e alto medioevo. La forma artistica nel
passagio dall'antichita al medioevo ... 1967 (Accad. Naz. dei Lincei. Problemi
attuali ... 105), Rome 1968, 99-104.
Drachmann (1922)-Drachmann, A. B., Atheism in Pagan Antiquity, London
1922.
Drew-Bear (1978)-Drew-Bear, T., Nouvelles inscriptions de Phrygie (Studia Am-
stelodamensia 16), Zutphen (Netherlands) 1978.
Drexel (1922)-Drexel, F., "DieGOtterverell.rungimromischenGermanien," BRGK
14 (1922) 1-68.
Dreyer (1970)-Dreyer, 0., Untersuchungen zum Begriff des Gottgeziemenden in
der Antike, Hildesheim 1970.
214 Bibliography

Dubois (1902)-Dubois, C.-A., "Culte et dieux a Pouzzoles," MEFR 22 (1902)

23-68.
Dufournet (1974)-Dufournet, J., "Le dieu Vintius ... ," in Melanges offerts a
Roger Dion, ed. R. Chevallier, Paris 1974, 379-405.
Du Mesnil du Buisson (1943-1944)-Du Mesnil du Buisson, R., "L'offrande de
poisson a Ishtar," BSAF 1943-1944, 317-327.
F. Dunand (1972-1973)-Dunand, F., Le culte d'lsis dans le bassin orientale de La
Mediterranee, 3 vols., Leiden 1972-1973.
M. Dunand (1939)-Dunand, M., "Nouvelles inscriptions du Djebel Druze et du
a
Hauran," in Melanges syriens offerts Monsieur Rene Dussaud . .. , 2 (Paris
1939) 559-576.
Dupont-Sommer and Robert (1964)-Dupont-Sommer, A., and L. Robert, La
deesse de Hierapolis Castabala (Cilicie), Paris 1964.
Dussaud (1922)-Dussaud, R., "Le temple de Jupiter Damascenien et ses transfor·
mations aux epoques chretienne et musulmane," Syria 13 (1922) 219-250.
Duthoy (1969)-Duthoy, R., The Taurobolium.lts Evolution and Terminology,
Leiden 1969.
Edelstein and Edelstein (1945 )-Edelstein, E. J. and L., Asclepius. A Collection and
Interpretation of the Testimonies, 2 vols., Baltimore 1945.
Eitrem (1929)-Eitrem, S., "Philostrats Heroikos," Symb. Oslo. 8 (1929) 1-56.
Eitrem (1939)-Eitrem, S., "Die magischen Gemmen und ihre Weihe," Symb.
Oslo. 19 (1939) 57-85.
Elderkin (1937)-Elderkin, G. W., "Two curse inscriptions," Hesperia 6 (1937)
382-395.
Emmett and Pickering (1975)-Emmett, A.M., and S. R. Pickering, "The impor·
tance of P. Bodmer XX, the Apology of Phileas," Prudentia 7 (1975) 95-103.
Engelmann (1976)-Engelmann, H., ed., Die Inschriften von Kyme, Bonn 1976.
Erbse (1952)-Erbse, H., "Plutarchs Schrift 1tEpi 5sLaL5aL~ovia~," Hermes 80 (1952)
296-314.
Etienne et al. (1976)-Etienne, R., et al., "Les dimensions sociales de Ia romanisa·
tion dans Ia Peninsule iberique des origines aIa fin de l'Empire," in Assimila-
a
tion et resistance la culture greco-romaine dans le monde ancien. Travaux du
VJe Congres international de la Federation Internationale des Associations
d'etudes classiques . .. 1974, Bucharest 1976, 95-109.
Evans (1939)-Evans, E. C., The Cults of the Sabine Territory, Rome 1939.
Fears (1977)-Fears, J. R., Princeps a diis electus. The Divine Election of the Em-
peror, as a Political Concept at Rome, Rome 1977.
Fentress (1978)-Fentress, E., "Dii Mauri and dii patrii," Latomus 37 (1978) 507-
516.
Ferguson (1970)-Ferguson, J., The Religions of the Roman Empire, London
1970.
Festugiere (1935)-Festugiere, A.-J., "Les mysteres de Dionysos," Rev. bibl. 44
(1935) 192-2ll, 366-396.
Festugiere (1945)-Festugiere, A.-J., Traites I-XII (Corpus Hermeticum I, texte
etabli par A. D. Nock et traduit par A.·J. Festugiere), Paris 1945.
Festugiere (1949)-Festugiere, A.·J., "A propos des areta1ogies d'Isis," HThR 42
(1949) 209-234.
 Bibliography 215

Festugiere (1959)-Festugiere, A.-J., Antioche paienne et chretienne. Libanius,

Chrysostome, et les moines de Syrie, Paris 1959.
Fevrier (1977)-Fevrier, P.-A., "A propos du repas funeraire: culte et sociabilite,"
Cahiers archeologiques 26 (1977) 29-45.
Fink et al. (1940)-Fink, R. 0., et al., "The Feriale Duranum," Yale Class. Studies
7 (1940) 1-222.
Finke (1927)-Finke, H., "Neue Inschriften," BRGK 17 (1927) 1-107.
Fishwick (1967)-Fishwick, D., "Hastiferi," ]RS 57 (1967) 142-160.
Fish wick (1978)-Fishwick, D., "Augustus deus and deus Augustus," in Hommages
aM.]. Vermaseren, ed. M. B. de Boer and T. A. Edridge, I, Leiden 1978, !175-
!180.
Fitz (1959)-Fitz, J., "Der Besuch des Septimius Severus in Pannonien im Jahre 202
u.Z.," Acta arch. acad. sci. Hung. 11 (1959) 2!17-26!1.
Fitz (1968)-Fitz, J., "M. Campanius Marcellus," Acta ant. acad. sci. Hung. 16
(1968) !11!1-!12!1.
Fitz (1972)-Fitz, J., Les Syriens a Intercisa, Brussels 1972.
Flusser (1975)-Flusser, D., "The great goddess of Samaria," Israel Exploration
]ourn. 25 (1975) 1!1-20.
Fontrier (1883)-Fontrier, A., "Inscriptions d'Asie Mineure," BCH 7 (1883) 501-
506.
Formige (1944)-Formige, J., "Le sanctuaire de Sammy ... ," Gallia 3 (1944) 4!1-97.
Forrer (1915)-Forrer, R., Das Mithra-Heiligtum von Konigshofen bei Strassburg,
Stuttgart 1915.
Fossey (1897)-Fossey, C., "Inscriptions de Syrie, II. Djo1an et Hauran," BCH 21
(1897) 39-65.
Foucher (1964)-Foucher, L., Hadrumetum, Tunis 1964.
Franchi de' Cavalieri (1901)-Franchi de' Cavalieri, P., I martiri diS. Teodoto e di
S. Ariadne, con un appendice sul testa originale del martirio (Studi e testi 6),
Rome 1901.
Francis (1975)-Francis, E. D., "Mithraic graffiti from Dura-Europus," in Mithraic
Studies. Proceedings of the First International Congress of Mithraic Studies,
ed. J. R. Hinnells, Manchester 1975, 2.424-445.
Franciscis (1954)-Franciscis, A. de, "Regione I (Latium et Campania) (Cam-
pania)," Not. Scavi 79 (1954) 269-290.
Frank (193!1-1940)-Frank, T., An Economic Survey of Ancient Rome, 6 vols.,
Baltimore 19!1!1-1940.
Franke (1968)-Franke, P. R., Kleinasien zur Romerzeit. Griechisches Leben im
Spiegel der Miinzen, Munich 1968.
Frantz (1965)-Frantz, A., "From paganism to Christianity in the temples of
Athens," Dumbarton Oaks Papers 19 (1965) 187-205.
Fraser (1960)-Fraser, P.M., The Inscriptions on Stone (Samothrace. Excavations
Conducted by the Institute of Fine Arts of New York University, ed. K. Leh·
mann, 2, 1), New York 1960.
Fredouille (1972)-Fredouille, J.·C., Tertullien et la conversion de la culture an·
tique, Paris 1972.
Frend (1952)-Frend, W.H.C., The Donatist Church. A Movement of Protest in
Roman North Africa, Oxford 1952.
216 Bibliography

Frezouls (1952)-Frezou1s, E., "Les theatres romains de Syrie," Annales arch. de

Syrie 2 (1952) 46-100.
Friedlaender (1921-1923)-Friedlaender, L., Darstellungen aus der Sittengeschichte
Roms in der Zeit von Augustus bis zum Ausgang der Antonine, 10 ed. G. Wis-
sowa, vols. I (1922), 2 (1922), 3 (1923) and 4 (1921), Leipzig.
Frier (1977)-Frier, B. W., "The rental market in early imperial Rome," ]RS 67
(1977) 27-37.
Fiilep (1954)-Fiilep, F., "Epigraphie," in Intercisa I. (Dunapentele-Sztdlinvdros).
Geschichte der Stadt in der Romerzeit, eds. L. Barkbczi et al., Budapest 1954,
232-273.
Gaertringen (1935)-Gaertringen, H. von, "Ein Asklepiosorakel aus Athen," Ar-
chiv fiir Religionswiss. 32 (1935) 367-370.
Gage (1955)-Gage, J., Apollon romain. Essai sur le culte d'Apollon et le deve-
a a
loppement du 'ritus Graecus' Rome des origines Auguste, Paris 1955.
Gage (1964)-Gage, J., Les classes sociales dans l'empire romain, Paris 1964.
Gallavotti (1972)-Gallavotti, C., "Esegesi di due epigrafi Iarine," Giornale italia-
na di filol. 24 (1972) 354-362.
Ganschinietz (1913)-Ganschinietz, R., Hippolytos' Capitel gegen die Magier.
Refut. Haer. IV 28-32 (Texte und Untersuchungen 39), Leipzig 1913.
Gauckler (1899)-Gauckler, P., "Les fouilles aCarthage," BCTH 1899, clix-clxiv.
Gaur (1963)-Gaur, A., "Les danses sacrees en Inde," in Les danses sacrees ...
(Sources orientales VI), Paris 1963, 315-342.
Gawlikowski (1973)-Gawlikowski, M., Palmyre VI. Le temple palmyrenien.
Etude d'epigraphie et de topographie historique, Warsaw 1973.
Geffcken (1907)-Geffcken, J., lwei griechische Apologeten, Leipzig-Berlin 1907.
Geffcken (1920)-Geffcken, J., Der Ausgang des griechisch-romischen Heiden-
turns, Heidelberg 1920.
Gilliam (1976)-Gilliam, J. F., "Invitations to the klineofSarapis," in Collectanea
papyrologica. Texts Published in Honor of H. C. Youtie, Bonn 1976, UH5-
324.
a
Girard (1878)-Girard, P., "Ex-votos Esculape trouves sur Ia pente meridionale
de l'Acropole," BCH 2 (1878) 65-94.
Girard (1881)-Girard, P., L'Asclepieion d'Athenes d'apres de recentes decouvertes,
Paris 1881.
Girard and Martha (1878)-Girard, P., and J. Martha, "Inventaires de l'Ascle-
pieion," BCH 2 (1878) 419-445.
Glueck (1937)-Glueck, N., "A newly discovered Nabataean temple of Atargatis
and Hadad at Khirbet et-Tanniir, Transjordania," A]A 41 (1937) 361-376.
Golfetto (1961)-Golfetto, A., Dougga. Die Geschichte einer Stadt im Schatten
Karthagos, Basel 1961.
Goodhue (1975)-Goodhue, N., The Lucus Furrinae and the Syrian Sanctuary on
the ]aniculum, Amsterdam 1975.
Gordon (1972)-Gordon, R. L., "Mithraism and Roman society: social factors in
the explanation of religious change in the Roman Empire," Religion 2 (1972)
92-121.
Gordon (1975)-Gordon, R. L., "Franz Cumont and the doctrines of Mithraism,"
in Mithraic Studies. Proceedings of the First International Congress of Mith-
raic Studies, ed. J. R. Hinnells, Manchester 1975, 215-248.
 Bibliography 217

Gough (1952)-Gough, M., "Anazarbus," Anatolian Studies 2 (1952) 85-150.

Graeve (1972)-Graeve, V. von, "Tempel und Kult der syrischen GOtter am jani-
culum," /b. des Deutschen Arch. lnst. 87 (1972) 314-347.
Graillot (1912)-Graillot, H., Le culte de Cybele, mere des dieux, ti Rome et dans
l'empire, Paris 1912.
Grandjean (1975)-Grandjean, Y., Une nouvelle aretalogie d'lsis ti Maronee, Lei-
den 1975.
Grant (1968)-Grant, M., The Climax of Rome. The Final Achievements of the
Ancient World AD 161-337, London 1968.
Gregoire (1939)-Gregoire, H., "Les pierres qui crient. Les chretiens et I'oracle de
Didymes," Byz.antion 14 (1939) 318-321.
Gsell (1931)-Gsell, S., "Autel romain de lana (Algerie)," CRAI1931, 251-269.
Guarducci (1942-1943)-Guarducci, M., "Le impronte del Quo Vadis e monu-
menti affini, figurati ed epigrafici," Rend. Pont. Accad. Romana di Archeologia
19 (1942-1943) 305-344.
Guey (1953)-Guey, J., "Epigraphica Tripolitana," REA 55 (1953) 334-358.
Habicht (1959-1960)-Habicht, C., "Zwei neue lnschriften aus Pergamon," Mitt.
des Deutschen Arch. Inst., Istanbul9110 (1959-1960) 109-127.
Habicht (1969)-Habicht, C., Die lnschriften des Asklepieions (A1tertiimer von
Pergamon, ed. E. Boehringer, VIII 3), Berlin 1969.
Hall (1977)-Hall, A., "A sanctuary of Leta at Oenoanda," Anatolian Studies 27
(1977) 193-197.
Hanfmann and Waldbaum (1975)-Hanfmann, G. M.A., and j. C. Waldbaum, A
Survey of Sardis and the Major Monuments Outside the City Walls, Cambridge
1975.
Hanslik (1954)-Hanslik, R., "Apollo Pythius und sein Kult," Vigiliae Christianae
8 (1954) 176-181.
Hanson (1959)-Hanson, j. A., Roman Theater-Temples, Princeton 1959.
Harder (1944)-Harder, R., Karpokrates von Chalkis und die memphitische Isis-
propaganda, Berlin 1944.
Harnack (1905)-Hamack, A. von, Der Vorwurf des Atheismus in den drei ersten
/ahrhunderten (Texte und Untersuchungen 28, 4), Leipzig 1905.
Harper (1972)-Harper, R. P., "Tituli Comanorum Cappadociae iterum suppleti,"
Anatolian Studies 22 (1972) 225-239.
Hasluck (1904)-Hasluck, F. W., "Unpublished inscriptions from the Cyzicus
neighborhood," JHS 24 (1904) 20-40.
Hatt (1965)-Hatt, J.·J., "Essai sur l'evolution de Ia religion gauloise," REA 67
(1965) 80-125.
Hatt (1978)-Hatt, J.-j., "Divinites orientales et dieux gaulois," in Paganisme,
judaisme, christianisme ... Melanges oflerts ti Marcel Simon, Paris 1978, 277-
286.
Hatt (1979)-Hatt, j.-j., "Leculte de Mars indigene dans le Nord-Est de laGaule,"
Rev. arch. de l'Est 30 (1979) 121-138.
Haufeld (1920)-Haufeld, j., "Inscriptions de Lagina en Carie," BCH 44 (1920)
70-100.
Haufeld (1927)-Haufeld, j., "Inscriptions de Panamara," BCH 51 (1927) 57-122.
Haussoullier (1899)-Haussoullier, B., "Notes d'epigraphie milesienne. Ooopia,
9£(J)p{a, 9eopia," Rev. de philol. 23 (1899) 313-320.
218 Bibliography

Haussoullier (1902)-Haussoullier, B., Etudes sur l'histoire de Milet et du Didy-

meion, Paris 1902.
Haussoullier and Ingholt (1924)-Haussoullier, B., and H. Ingholt, "Inscriptions
grecques de Syrie," Syria 5 (1924) 316-341.
Heitsch (1961-1964)-Heitsch, E., Die griechischen Dichterfragmente der romi-
schen Kaiserzeit, 1st ed., vol. I, Gottingen 1961; ed. 2, 2 vols., Gottingen
1963-1964.
Hepding (1903)-Hepding, H., Attis, seine Mythen und sein Kult, Giessen
1903.
Herbillon (1929)-Herbillon, J., Les cultes de Patras . .. , Baltimore 1929.
Herrmann and Polatkan (1969)-Herrmann, P., and K. Z. Polatkan, Das Testa-
ment des Epikrates und andere neue lnschriften aus dem Museum von Manisa,
Vienna 1969.
Herzog (1931)-Herzog, R., Die Wunderheilungen von Epidauros. Ein Beitrag zur
Geschichte der Medizin und der Religion, Leipzig 1931.
Heuten (1937)-Heuten, G., "Primus in orbe deos fecit timor," Latomus 1 (1937)
3-8.
Heyder and Mallwitz (1978)-Heyder, W., and A. Mallwitz, Die Bauten im Kabiren-
heiligtum bei Theben (Das Kabirenheiligtum bei Theben II), Berlin 1978.
Hicks (1891)-Hicks, E. L., "Inscriptions from western Cilicia," ]HS 12 (1891)
225-273.
Hill (1897)-Hill, G. F., "Notes on additions to the Greek coins in the British
Museum, 1887-1896," ]HS 17 (1897) 78-91.
Hill (1911)-Hill, G. F., "Some Graeco-Phoenician shrines," ]HS 31 (1911)56-64.
Hinnells (1975)-Hinnells, J. R., "Reflections on the bull-slaying scene," in Mith-
raic Studies. Proceedings of the First International Congress of Mithraic Stud-
ies, ed. J. R. Hinnells, Manchester 1975, 290-312.
Hinrichs (1970)-Hinrichs, A., "Pagan ritual and the alleged crimes of the early
Christians. A reconsideration," in Kyriakon. Festschrzft johannes Quasten,
Munster 1970, 18-35.
Hoddinott (1975)-Hoddinott, R. F., Bulgaria in Antiquity ... , London 1975.
Hoffiller and Saria (1938)-Hoffiller, V., and B. Saria, Ant ike Inschriften aus ]ugo-
slavien, Heft 1: Noricum und Pannonia Superior, Zagreb 1938.
Hoffmann (1963)-Hoffmann, D., "Die spatromischen Soldatengrabschriften von
Concordia," Mus. Helv. 20 (1963) 22-57.
Hohlwein (1940)-Hohlwein, N., "Deplacement et tourisme dans l'Egypte ro-
maine," Chron. d'Egypte 30 (1940) 253-278.
Homo (1904)-Homo, L., Essai sur le regne de l'empereur Aurilien, Paris 1904.
Homolle (1882)-Homolle, T., "Comptes des hieropes du temple d'Apollon de-
lien," BCH 6 (1882) 1-167.
Hopfner (1922)-Hopfner, T., Ueber die Geheimlehren von ]amblichus, Leipzig
1922.
Hopkins (1978)-Hopkins, K., Conquerors and Slaves ... , Cambridge and New
York 1978.
Ianovitz (1972)-Ianovitz, 0., ll culto solare nella 'X Regio,' Milan 1972.
Imhoof-Blumer (1911)- Imhoof-Blumer, F., "Beitriige zur Erkllirung griechischen
Miinztypen," Nomisma 6 (19ll) 1-23.
 Bibliography 219

Ivanov (1974)-Ivanov, T., "Die 1etzten Ausgtabungen des romischen und friih-
byzantinschen Donaulimes in der VR Bu1garien," in Actes du IX' Congres
international d'etudes sur les frontieres romaines ... 1972, Bucharest 1974,
55-69.
Jalabert (1908)-Jalabert, L., "Deux missionsarcheologiquesamericainesenSyrie,"
Mel. de l'Univ. St. Joseph 3 (1908) 713-752.
James (1924)-James, M. R., The Apocryphal New Testament, being the Apocry-
phal Gospels, Acts, Epistles, and Apocalypses, Oxford 1924.
Jerphanion and Jalabert (1908)-Jerphanion, G. de, and L. Ja1abert, "Inscriptions
d'Asie Mineure (Pont, Cappadoce, Cilicie)," Mel. de l'Univ. St. Joseph 3 (1908)
437-479.
Jobst (1976)-Jobst, W., "Der Pfaffenberg-ein Zentrum romischer Staatsreligion
in Camuntum," Neue Forschungen in Carnuntum (Kultur und Natur im
Niederosterreich-Studien und Forschungen 1), Vienna 1976, 33-52.
A.H.M. Jones (1940)-Jones, A.H.M., The Greek City, from Alexander to Jus·
tinian, Oxford 1940.
A. H. M. Jones (1962)-Jones, A. H. M., Constantine and the Conversion of Europe,
New York 1962.
A.H.M. Jones (1963)-Jones, A.H.M., "The social background of the struggle
between pagans and Christians," in The Conflict Between Paganism and
Christianity in the Fourth Century, ed. A. Momig1iano, Oxford 1963, 17-37.
A.H.M. Jones (1964)-Jones, A.H.M., The Later Roman Empire 284-602. A
Social Economic and Administrative Survey, 2 vols., Norman 1964.
L. W. Jones (1929)-Jones, L. W., The Cults of Dacia, Berkeley 1929.
Jouguet (1911)-Jouguet, P., La vie municipaledans l'Egypteromaine, Paris 1911.
Jucker (1960)-Jucker, H., "Gekiisste GOtterbilder," Antike Kunst 3 (1960) 91-92.
Jullian (1908-1926)-Jullian, C., Histoire de la Gaule, 8 vols., Paris 1908-1926.
Kabbadia (1885)-Kabbadia, P., '"Emypa<pai b: tciiv 'ev 'Ent8aup{a dvao~ea<pciiv,"
Ephem. arch. 1885, 189-199.
Kaegi (1966)-Kaegi, W. E., "The fifth-century twilight of Byzantine paganism,"
Classica et mediaevalia 27 (1966) 243-275.
Kahrstedt (1954)- Kahrstedt, U ., Das wirtschaftsliche Gesicht Griechenlands in der
Kaiserz.eit. Kleinstadt, Villa und Domiine, Bern 1954.
Kaibel (1878)-Kaibel, G., Epigrammata graeca ex lapidibus conlecta, Berlin 1878.
Kalinka (1933)-Kalinka, E., "Aus Bithynien und Umgegend," JOAI 28 (1933)
Beibl. cols. 45-112.
Kan (1943)-Kan, A. H., Juppiter Dolichenus. Sammlung der lnschriften und
Bildwerke, Leiden 1943.
Kane (1975)-Kane, J.P., "The Mithraic cult meal in its Greek and Roman en-
vironment," in Mithraic Studies. Proceedings of the First International Con-
gress of Mithraic Studies, ed. J. R. Hinnells, Manchester 1975, 2.313-351.
Kantorowicz (1961)-Kantorowicz, E. H., "Gods in uniform," Proc. Am. Philos.
Soc. 105 (1961) 368-396.
Kantorowicz (1963)-Kantorowicz, E. H., "Oriens Augusti-Lever du roi," Dum-
barton Oaks Papers 17 (1963) 117-177.
Keil (1908)-Kei1, J., "Zur Geschichte der Hymnoden in der Provinz Asia," JOAI
11 (1908) 101-110.
220 Bibliography

Keil (192S)-Keil, J., "Die Kulte Lydiens," in Anatolian Studies Presented to Wil-
liam M. Ramsay, Manchester 192S, 2S9-266.
Keil (1939)-Keil, J., "Kulte im Prytaneion von Ephesos," in Anatolian Studies
Presented to W. H. Buckler, Manchester 19S9, 119-128.
Keil and Premerstein (1908)-Keil, J ., and A. von Premerstein, Bericht iiber eine
Reise in Lydien und der siidlichen Aiolis ausgefiihrt 1906 . .. (Denkschriften
der kaiserlichen Akad. der Wiss. in Wien, Phil.-hist. Kl. 5S, 2), Vienna
1908.
Kerenyi (1955)-Kerenyi, C., "The mysteries of the Kabeiroi," in The Mysteries
(Papers from the Eranos Yearbooks, 2), New York 1955, S2-6S.
Kern (1892)-Kern, 0., "ATJJl11tTJP Is)..'f\vTJ," Ephem. arch. 1892, 1JS-ll8.
Kern (1900)-Kern, 0., Die lnschriften von Magnesia am Maeander, Berlin 1900.
Koethe (19SS)-Koethe, H., "Die keltischen Rund- und Vielecktempel der Kaiser-
zeit," BRGK 2S (19SS) 10-108.
Koets (1929)-Koets, P. J., Astat3atJlov{a. A Contribution to the Knowledge of the
Religious Terminology in Greek, Purmerend (Netherlands) 1929.
Kotting (1950)-Kotting, B., Peregrinatio Religiosa. Wahllfahrten in der Antike
und das Pilgerwesen in der alten Kirche, Munster 1950.
Krill (1978)-Krill, R. M., "Roman paganism under the Antonines and Severans,"
ANRW II 16, 1 (Berlin 1978) 27-44.
Kroll (1924)-Kroll, W., Studien zum Verstiindnis der romischen Literatur, Stutt-
gart 1924.
Kunze et al. (1958)-Kunze, E., et al., VI Bericht iiber die Ausgrabungen in Olym-
pia . . . , Berlin 1958.
Labhardt (1960)-Labhardt, A., "Curiositas. Notes sur 1'histoire d'un motetd'une
notion," Mus. Helv. 17 (1960) 206-224.
Lafaurie (1966)-Lafaurie, J ., "Dies imperii Constantini Augusti: 25 decembre S07.
Essai sur quelques problemes de chronologie constantinienne," in Melanges
a
d'archeologie et d'histoire offerts A. Piganiol, Paris 1966, 795-806.
Lambrechts (1952)-Lambrechts, P., "Les fetes 'phrygiennes' deCybeleetd'Attis,"
Bull. Inst. Hist. Beige de Rome 27 (1952) 141-170.
Lambrechts and Noyen (1954)-Lambrechts, P., and P. Noyen, "Recherches sur 1e
culte d'Atargatis dans le monde grec," Nouvelle Clio 6 (1954) 258-277.
Lanckoronski (1890-1892)-Lanckoronski, K., Stiidte Pamphyliens und Pisidiens,
unter Mitwirkung von G. Niemann und E. Peterson, 2 vols., Vienna 1890-
1892.
Lane (1971-1978)-Lane, E. N., Corpus monumentorum religionis dei Menis, 4
vols., Leiden 1971-1978.
Lane (1978a)-Lane, E. N., "The temple-type of Prostanna: a query," in Studien
zur Religion und Kultur Kleinasiens. Festschrift fiir Friedrich Karl Domer ... ,
Leiden 1978, 2.540-545.
Lassere (197S)-Lassere, J.-M., "Recherches sur Ia chronologie des epitaphes pai-
ennes de I'Afrique," Antiquites africaines 7 (197S) 7-151.
Laue (19Sl)-Laue, K., "lnscriptiones Graecae ... ,"Gnomon 7 (19S1) 1U-U5.
Laue (1960)-Latte, K., Romische Religionsgeschichte (Handbuch der kl. Alter-
tumswiss. V, 4), Munich 1960.
 Bibliography 221

Lattimore (1942)-Lattimore, R., Themes in Greek and Latin Epitaphs, Urbana

1942.
Laumonier (1934)-Laumonier, A., "Notes sur une inscription de Stratonicee,"
REA 36 (1934) 85-87.
Laumonier (1937)-Laumonier, A., "Recherches sur Ia chronologie des pretres de
Panamara," BCH 61 (1937) 286-298.
Laumonier (1958)-Laumonier, A., Les cultes indigenes en Carie, Paris 1958.
a
Launey (1937)-Launey, M., "Le verger d'Heracles Thasos," BCH 61 (1937)
380-409.
Laurin (1954)-Laurin, J. R., Orientations mattresses des apologistes chretiens de
270 4 J61, Rome 1954.
Lebreton (1923-1924)-Lebreton, J ., "Le desaccord de la theologie savante et de Ia
foi populaire dans l'Eglise chretienne du mesiecle," Rev. d' hist. ecclesiastique
19 (1923) 481-506 and 20 (1924) 5-37.
LeGlay (1954)-LeG1ay, M., "Le mithraeum de Lambese," CRAI1954, 269-278.
LeG1ay (1966)-LeG1ay, M., Satumt! ~ricain, histoire, Paris 1966.
LeGlay (1971)-LeG1ay, M., "La v1e religieuse a Lambese d'apres de nouveaux
documents," Ant. afr. 5 (1971) 125-153.
LeG1ay (1972)-LeG1ay, M., "Unevillea vocationdecapita1e:Lugdunum," Arch-
eologia 50 (1972) 10-14.
LeGlay (1976)-LeGlay, M., "Hadrien et l'Asktepieion de Pergame," BCH 100
(1976) 347-372.
Lehmann (1989)-Lehmann (-Hartleben), K., "Excavations in Samothrace," A]A
43 (1989) 133-145.
Lehmann (1954)-Lehmann, K., "The mystery cult of Samothrace," Archaeology
7 (1954) 91-95.
Leonhard (1914)-Leonhard, F., Das Kastell Altstadt bei Miltenberg (Oberger-
manisch-Raetische Limes des Romerreiches Bill 38), Heidelberg 1914.
Leschi (1957)-Leschi, L., Etudes d'epigraphie, d'archeologie et d'histoire afri-
caines, Paris 1957.
Lesky (1966)-Lesky, A., A History of Greek Literature, trans. J. Willis and C. de
Heer, London 1966.
Levi (1944)-Levi, D., "Aion," Hesperia 13 (1944) 269-814.
a
E. Uvy (1968)-Uvy, E., "Nouveaux bijoux Delos," BCH 92 (1968) 523-539.
I. Uvy (1895)-Uvy, I., "Etudes sur Ia vie municipa1e en Asie Mineure au temps
des Antonins," REG 8 (1895) 208-250.
Liebeschuetz (1979)-Liebeschuetz, J. H. W. G., Continuity and Change in Roman
Religion, Oxford 1979.
Lorincz (1978)-LOrincz, B., "Die Besatzungstruppen des Legionslager von Aquin-
cumam Ende des 1.-Anfang des 2. Jahrhunderts," Acta arch. acad. sci. Hung.
30 (1978) 299-312.
Longo (1969)-Longo, V., Aretalogie nel mondo greco, 2 vo1s., Genoa 1969.
L'Orange (1935)-L'Orange, H. P., "Sol invictus imperator. Ein Beitragzur Apo-
theose," Symb. Oslo. 14 (1935) 86-114.
Maas (1933)-Maas, P., Epidaurische Hymnen (Schriften der KOnigsberger Ge-
lehrten Gesellschaft 9, 5), Halle 1933, 125-162.
222 Bibliography

MacMullen (1962)-MacMullen, R., "Roman bureaucratese," Traditio 18 (1962)

364-378.
MacMullen (l966a)-MacMullen, R., Enemies of the Roman Order ... , Cam-
bridge 1966.
MacMullen (l966b)-MacMullen, R., "A note on sermo humilis," ]ThStudies 17
(1966) 108-112.
MacMullen (1968)-MacMullen, R., "Constantine and the miraculous," GRBS 9
(1968) 81-96.
MacMullen (1970)-MacMullen, R., "Roman market days," Phoenix 24 (1970)
333-341.
MacMullen (1972)-MacMullen, R., "Sfiducia nell'intelletto nel quarto secolo,"
Rivista storica italiana 84 (1972) 5-16.
MacMullen (1974)-MacMullen, R., Roman Social Relations ... , New Haven
1974.
MacMullen (1976)-MacMullen, R., Roman Government's Response to Crisis ... ,
New Haven 1976.
MacMullen (1980)-MacMullen, R., "The power of bishops outside the Church,"
in The Role of the Christian Bishop in Ancient Society. Thirty-fifth Colloquy
... 1979 (Center for Hermeneutical Studies, Grad. Theol. Union, Berkeley
1980) 25-29.
Macpherson (1972)-Macpherson, I. W., "Six inscriptions from Galatia," Ana-
tolian Studies 22 (1972) 217-223.
Magie (1950)-Magie, D., Roman Rule in Asia Minor to the End of the Third Cen-
tury after Christ, 2 vols., Princeton 1950.
Maguire (1939)-Maguire, J. P., "The sources of Pseudo-Aristotle De Mundo,"
Yale Class. Studies 6 (1939) ll1-167.
Malaise (1972)-Malaise, M., Les conditions de penetration et de diffusion des
cultes egyptiens en ltalie, Leiden 1972.
Marchetti Longhi (1943-1945)-Marchetti Longhi, G., "Gli scavi del largo Argen-
tina," Bull. comm. 71 (1943-1945) 57-95.
Marcillet-Jaubert (1974)-Marcillet-Jaubert, J., "A propos de M. Aurelius Deci-
mus," ZPE 14 (1974) 249-251.
Martroye (1915)-Martroye, F., "Mesures prises par Constantin contre la super-
stitio," BSAF 1915, 280-292.
Martroye (1916)-Martroye, F., "Du sens du mot superstitio," BSAF 1916, 106-107,
lll-ll3.
Mattingly (1928)-Mattingly, H., Roman Coins from the Earliest Times to the Fall
of the Western Empire, London 1928.
Matz (1940)-Matz, F., "Die griechischen Sarkophage mit bakchischen Darstel-
lungen," in Bericht uber den VI. internationalen Kongress fur Archaologie . ..
1939, Berlin 1940, 502-503.
McCown (1938)-McCown, C. C., "The festival theater at the Birketein," in C. H.
Kraeling, Gerasa, City of the Decapolis . .. , New Haven 1938, 159-167.
McPhee (1966)-McPhee, C., Music in Bali, New Haven 1966.
Meiggs (1960)-Meiggs, R., Roman Ostia, Oxford 1960.
Mellor (1975 )-Mellor, R., ®EA POMH. The Worship of the Goddess Romain the
Greek World, GOttingen 1975.
 Bibliography 221

Merkelbach (1962)-Merkelbach, R., Roman und Mysterium in der Antike, Munich

1962.
Merkelbach (1973)-Merkelbach, R., "Zwei Texte aus dem Sarapeum zu Thessa·
lonike," ZPE 10 (1973) 45-54.
Merlan (1967)-Merlan, P., "Greek philosophy from Plato to Plotinus," in The
Cambridge History of Later Greek and Early Medieval Philosophy, ed. A. H.
Armstrong, Cambridge 1967, 14-136.
Merlat (1951)-Merlat, P., Repertoire des inscriptions et monuments figures du
culte de jupiter Dolichenus, Paris and Rennes 1951.
a
Merlin (1908)-Merlin, A., Le temple d'Apollon Bulla Regia, Paris 1908.
Merlin (1909)-Merlin, A., "Les dernieres decouvertes d'antiquites en Tunisie,"
BCTH 1909, ccxxxi-ccxxxviii.
Merlin (1910)-Merlin, A., "Le sanctuaire de Baal et deTanitpresdeSiagu,'' Notes
et documents 4 (1910) 3-58.
Merlin (1916)-Merlin, A., "Une nouvelle inscription aThuburbo Majus," CRAI
1916, 262-267.
Merlin and Poinssot (1908)-Merlin, A., and L. Poinssot, "Les inscriptions d'Uchi
Majus d'apres les recherches du Capitaine Gondouin," Notes et documents 2
(1908) 3-128.
Meslin (1974)-Meslin, M., "Realites psychiques et valeurs religieuses dans les
cultes orientaux (Jeqve siecles)," Rev. hist. 252 (1974) 289-314.
Meslin (1978)-Meslin, M., "Convivialite ou communion sacramentelle? Repas
mithriaque et Euchariste chretienne," in Paganisme, judaisme, christianisme
a
... Melanges o/ferts Marcel Simon, Paris 1978, 295-306.
Milik (1972)-Milik, J. T., Recherches d'epigraphie proche-orientale, I: Dedicaces
faites par des dieux (Palmyre, Hatra, Tyr) et des thiases semitiques al'epoque
romaine, Paris 1972.
Milne (1925)-Mi1ne, J. G., "The kline of Sarapis," ]EA 11 (1925) 6-9.
Mirkovic and Dusanic (1976)-Mirkovic, M., and S. Dusanic, Inscriptions de Ia
Mesie Superieure, I: Singidunum et le Nord-Ouest de la province, Belgrade
1976.
Mitford (1971)-Mitford, T. B., The Inscriptions of Kourion, Philadelphia 1971.
Mocsy (1970)-Mocsy, A., Gesellschaft und Romanisation in der romischen Provinz
Moesia Superior,.Amsterdam 1970.
Mocsy (1974)-Mocsy, A., Pannonia and Upper Moesia. A History of the Middle
Danube Provinces of the Roman Empire, London 1974.
Momigliano (1977)-Momigliano, A., Essays in Ancient and Modern Historiog·
raphy, Oxford 1977.
Mommsen (1868)-Mommsen, T., "Schauspielerinschrift von Philippi," Hermes
3 (1868) 461-465.
Mordtmann (1881a)-Mordtmann, J. H., "Zur Epigraphik von Kyzikos," Ath.
Mitt. 6 (1881) 121-131.
Mordtmann (1881b)-Mordtmann, J. H., "Inschriften aus Kallipolis," Ath. Mitt. 6
(1881) 256-265.
Mouterde (1930-1931)-Mouterde, R., "Le glaive de Dardanos. Objets et inscrip·
tions magiques de Syrie," Mel. de l'Univ. St. joseph 15 (1930-1931) 53-
137.
224 Bibliography

Mrozek (1973 )-Mrozek, S., "A propos de Ia repartition chronologique des inscrip-
tions latines dans le Haut-Empire," Epigraphica 35 (1973) 113-118.
Muller (1904)-Miiller, A., "Die ParasiteApollinis," Philologus63 (1904)342-361.
Mundie (1961)-Mundle, I., "Dea Caelestis in der Religionspolitik des Septimius
Severns und der Julia Domna," Historia 10 (1961) 228-237.
Musurillo (1972)-Musurillo, H., The Acts of the Christian Martyrs .. . , Oxford
1972.
Negev (1971)-Negev, A., "The Nabataean necropolis of Mampsis (Kurnub),"
Israel Exploration ]ournal21 (1971) 110-129.
Nestle (1948)-Nestle, W., Griechische Studien. Untersuchungen zur Religion,
Dichtung und Philosophic der Griechen, Stuttgart 1948.
Neubecker (1977)-Neubecker, A. J., Altgriechische Musik. Eine Einfilhrung,
Darmstadt 1977.
Nilsson (1941-1955)-Nilsson, M. P., Geschichte der griechischen Religion (Hand-
buch der Altertumswiss. 5, 2), 2 vols., Munich 1941-1955.
Nilsson (1945a)-Nilsson, M.P., "Pagan divine service in late paganism," HThR
38 (1945) 63-69.
Nilsson (1945b)-Nilsson, M. P., "The syncretistic relief at Modena," Symb. Oslo.
24 (1945) 3-7.
Nilsson (1947-1948)-Nilsson, M. P., "Die Religion in den griechischen Zauber-
papyri," Arsberiittelse, Kungl. Humanistiska Vetenskapssamfundet i Lund
(Bull. de la Soc. des Lettres de Lund) 1947-1948, 59-93.
Nilsson (1949)-Nilsson, M.P., "A propos du tombeau de Vincentius," Rev. arch.
31-32 (1949) 764-769.
Nilsson (1950)-Nilsson, M.P., "Lampen und Kerzen im Kultder Antike," Opus-
cula archaeologica 6 (1950) 96-111.
Nilsson (1963)-Nilsson, M.P., "The high god and the mediator," HThR56(1963)
101-120.
Nock (1925)-Nock, A. D., "Studies in the Graeco-Roman beliefs of the empire,"
]HS 45 (1925) 84-101.
Nock (1926)-Nock, A. D., Sallustius, Concerning the Gods and the Universe,
Cambridge 1926.
Nock (1930)-Nock, A. D., "A diis eJecta. A chapter in the religious history of the
third century," HThR 23 (1930) 251-274.
Nock (1933)-Nock, A. D., Conversion: The Old and the New in Religion from
Alexander the Great to Augustine of Hippo, Oxford 1933.
Nock (1934)-Nock, A. D., "A vision of Mandulis Aion," HThR 27 (1934) 53-104.
Nock (1937)-Nock, A. D., "The genius of Mithraism," ]RS 27 (1937) 108-113.
Nock (1946)-Nock, A. D., "Sarcophagi and symbolism," A]A 50 (1946) 140-170.
Nock (1958a)-Nock, A. D., "M. J. Vermaseren, Corpus inscriptionum et monu-
mentorum religionis Mithriacae, vol. 1," Gnomon 30 (1958) 291-295.
Nock (1958b)-Nock, A. D., "A cult ordinance in verse," HSCP63 (1958) 415-421.
Nock (1964)-Nock, A. D., Early Gentile Christianity and its Hellenistic Back-
ground, London 1964.
Ohlemutz (1940)-0hlemutz, E., Die Kulte und Heiligtilmer der Cotter in Perga-
mon, Wiirzburg 1940.
 Bibliography 225

Oliver (1936)-0liver, J. H., "The Sarapion monument and the Paean of Sop}lo-
cles," Hesperia 5 (1936) 91-122.
Oliver (1963)-0liver, J. H., "The Athenian archon Thisbianus," Hesperia 32
(1963) 318.
Oliver (1978)-0liver, J. H., "Hartmut Wolff. Die Constitutio Antoniniana ... ,"
A]P 99 (1978) 403-408.
Onurkan (1969-1970)-0nurkan,S., "Artemis Pergaia," Mitt. des DeutschenArch.
lnst., Istanbul19120 (1969-1970) 289-298.
Otto (1909)-0tto, W. F., "Religio und Superstitio," Archiv fiir Religionswiss. 12
(1909) 533-554.
Otto (1955)-0tto, W. F., "The meaning of the Eleusinian Mysteries," in The
Mysteries (Papers from the Eranos Yearbooks 2), New York 1955, 14-31.
Pace (1923)-Pace, B., "Artemis Pergaea," in Anatolian Studies Presented to Wil-
liam M. Ramsay, London 1923, 296-314.
Pack (1965)-Pack, R. A., The Greek and Latin Literary Texts from Greco-Roman
Egypt2, Ann Arbor 1965.
Packer (1967)-Packer, J. E., "Housing and population in imperial Ostia and
Rome," ]RS 57 (1967) 80-95.
Page (1962)-Page, D. L., Literary Papyri. Poetry (Select Papyri, ed. A. S. Hunt,3),
Cambridge and London 1962.
Palmer (1978)-Palmer, R. E. A., "Severan ruler-cult and the moon in the city of
Rome," ANRW II 16, 2 (Berlin 1978) 1085-1120.
Parassoglou (1976)-Parassoglou, G. M., "Circular from a prefect: sileat omnibus
perpetuo divinandi curiositas," in Collectanea papyrologica. Texts Published
in Honor of H. C. Youtie, I, Bonn 1976, 261-274.
Paribeni (1919)-Paribeni, R., "Rilievoconsceneegizie," Not. scavi 1919,106-112.
Parvan (1913)-Parvan, V., "Rumlinien," Arch. Am.. 1913, 364-392.
Pekary (1965)-Pekary, T., "Inschriftenfundeaus Milet 1959," Mitt. des Deutschen
Arch. Inst., lstanbul15 (1965) 118-134.
Pekary (1978)-Pekary, T., "Statuen in kleinasiatischen Inschriften," in Studien
zur Religion und Kultur Kleinasiens. Festschrift fiir Friedrich Karl Dorner ... ,
Leiden 1978, 727-744.
Pepe (1964)-Pepe, L., ,;Una 1ectio difficilior in Petronio (71, 10 triclia)," Giornale
italiana di filol. 17 (1964) 321-328.
Pepin (1966)-Pepin, J., "Porphyre exegete d'Homere," in Porphyre (Entretiens
sur l'Antiquire classique 12, Fondation Hardt), Geneva 1966, 231-266.
Perdrizet (1900)-Perdrizet, P., "Lettre au R. P. sejoume sur des inscriptions de
Syrie publiees dans Ia Revue biblique," Rev. bibl. 9 (1900) 429-443.
Perret (1949)-Perret, J., "Le polytheisme de Virgile," Rev. arch. 32 (1949) (Mil.
a
d'arch. et d'hist. o/ferts Ch. Picard 4) 793-802.
E. Peterson (1926)- Peterson, E., Etc; 9e6c;. Epigraphische, formgeschichtliche und
religionsgeschichtliche Untersuchungen, Gottingen 1926.
R. M. Peterson (1919)-Peterson, R. M., The Cults of Campania, Rome 1919.
a
Petit (1955)-Petit, P., Libanius et la vie municipale Antioche au IV' siecle apres
].-C., Paris 1955.
Pettazzoni (1939)-Pettazzoni, R., "La confession des peches en Syrie aux epoques
226 Bibliography

prechretiennes," in Melanges syriens offerts aM. Rene Dussaud, Paris 1939,

197-202.
Pflaum (1960-1961 )-Pflaum, H. G., Les carrieres procuratoriennes equestres sous
le Haut Empire romain, 4 vo1s., Paris 1960-1961.
C. Picard (1922)-Picard, C., Ephese et Glaros, Paris 1922.
C. Picard (1955)-Picard, C., "Le thefnre des mysteres de Cybe1e-Attis a Vienne
(lsere) et 1es theatres pour representations sacrees atravers le monde mediter-
raneen," CRAI 1955, 229-247.
G. Picard(1954)-Picard, G., "Un banquet costume surunemosaiqued'E1-Djem,"
CRAI 1954, 418-424.
G. Picard (1970)-Picard, G., "Les theatres ruraux de Gaule," Rev. arch. 1970,
185-192.
G. C. Picard (1972)-Picard, G. C., "Nouvelles images de dieux gallo-romains
accroupis," BSAF 1972, 48-50.
G. C. Picard (1977)-Picard, G. C., "Imperator cae1estium," Gallia 35 (1977) 89-
113.
Piganiol (1946)-Piganiol, A., "Observations sur le rituelle plus recent des freres
Arvales," CRAI 1946, 241-251.
Pohlmann (1960)-Pohlmann, E., Griechische Musikfragmente. Ein Weg zuralt·
griechischen Musik, Nuremberg 1960.
Poenaru-Bordea (1964)-Poenaru-Bordea, G., "Citeva inscriptii recentdescoperite
in Dobrogea," in Noi monumente epigrafice din Scythia Minor, ed. C. Daico-
viciu (Muzeul de Arheologie Constanta), Constanta 1964, 97-138.
Poland (1909)-Poland, F., Geschichte des griechischen Vereinswesens, Leipzig
1909.
Powell (1925)-Powell, J. U., Collectanea Alexandrina. Reliquiae minores poeta-
rum graecorum aetatis Ptolemaicae (323-146 a. C.), Oxford 1925.
Prentice (1908)-Prentice, W. K., Greek and Latin lnscriptions(Publicationso£an
American Archaeological Expedition to Syria in 1899-1900, Part III), New
York 1908.
Prentice (1922)-Prentice, W. K., Greek and Latin Inscriptions, Section B. Northern
Syria (Syria. Publications of the Princeton University Archaeological Expedi-
tions to Syria in 1904-5 and 1909, Division III), Leiden 1909-1922.
Price and Trell (1977)-Price, M. J., and B. L. Trell, Coins and their Cities: Archi-
tecture on the Ancient Coins of Greece, Rome and Palestine, Detroit 1977.
Prieur (1968)-Prieur, J., La province romaine des Alpes cottiennes, Villeurbanne
1968. .
Quandt (1962)-Quandt, W., Orphei hymni5, Berlin 1962.
Quasten (1930)-Quasten, J., Musik und Gesang in den Kulten der heidnischen
Antike und christlicher Friihzeit, Miinster 1930.
Quasten (1950-1960)-Quasten, J., Patrology, 3 vols., Westminster 1950-1960.
Raabe (1892)-Raabe, R., Die Apologie des Aristides. Aus dem syrischen iibersetzt,
Leipzig 1892.
Radet (1890)-Radet, G., "Inscriptions de Ia region du Meandre," BCH 14 (1890)
224-239.
Radnoti (1961)-Radnoti, A., "C. lulius Avitus Alexianus," Germania 39 (1961)
383-412.
 Bibliography 227

Ramsay (1889)-Ramsay, W. M., "Artemis-Leta and Apollo-Lairbenos," ]HS 10

(1889) 216-230.
Ramsay (1895-1897)-Ramsay, W. M., The Cities and Bishoprics of Phrygia ... ,
2 vols., Oxford 1895-1897.
Ramsay (1941)-Ramsay, W. M., The Social Basis of RomanPowerinAsiaMinor,
Aberdeen 1941.
Rebuffat (1975)-Rebuffat, R., "Trois nouvelles campagnes dans le sud de la Tri-
politaine," CRAI 1975, 495-505.
Redo (1973)-Redo, F., Numismatical Sources of the Illyr Soldier Emperors' Reli-
gious Policy, Budapest 1973.
Rehm (1938)-Rehm, A., "Kaiser Diokletian und das Heiligtum von Didyma,"
Philologus 93 (1938) 74-84.
Rehm (1958)-Rehm, A., Die Inschriften (Didyma, ed. T. Wiegand, II), Berlin
1958.
Reinach (1906)-Reinach, T., "Inscriptions d'Aphrodisias," REG 19 (1906) 79-
150, 205-298.
Reinach (1908)-Reinach, T., "llapeevrov," BCH 32 (1908) 499-513.
Remondon (1952)-Remondon, R., "L'Egypte et la supreme resistance au chris-
tianisme (Ve_VII• siecles)," Bull. Inst. Fr. d'arch. orient. 51 (1952) 63-78.
Renardet (1973)-Renardet, E., "Lieux-saints de la cite des Lingons," Caesarodu-
num 8 (1973) 120-127.
Reutersward (1958-1960)-Reutersward, P., Studien zur Polychromie der Plastik,
2 vols., Stockholm 1958-1960.
Rey-Coquais (1967)-Rey-Coquais, J.-P., Inscriptions grecques et latines de Syrie,
VI: Baalbek et Beqa', Paris 1967.
Rey-Coquais (1972)-Rey-Coquais, J.-P., "Une inscription du Liban nord," Mel.
de l'Univ. St. joseph 47 (1972) 87-105.
Reynaud et al. (1975)-Reynaud, J. F., et al., "Une nouvelle inscription grecque a
Lyon," Journal des Savants 1975, 47-75.
Riese (1894)-Riese, A., Carmina in codicibus scripta 2 (Anthologia Latina sive
poesis latinae supplementum, eds. F. Buecheler and A. Riese, I, 1-2), 2 vols.,
Leipzig 1894.
Robert and Robert (1946-1947)-Robert, J. and L., "Bulletinepigraphique," REG
59/60 (1946-1947) 298-372.
Robert and Robert (1958)-Robert, J. and L., "Bulletin epigraphique," REG 71
(1958) 169-363.
Robert (1937)-Robert, L., Etudes anatoliennes. Recherches sur les inscriptions
grecques de l'Asie Mineure, Paris 1937.
Robert (1939)-Robert, L., "Hellenica," Rev. de philol. 13 (1939) 97-217.
Robert (1940a)-Robert, L., Les gladiateurs dans !'Orient grec, Paris 1940.
Robert (1940b)-Robert, L., Hellenica 1, Limoges 1940.
Robert (1943)-Robert, L., "Voyages epigraphiques en Asie Mineure," Rev. de
philol. 17 (1943) 170-201.
Robert (1945)-Robert, L., Le sanctuaire de Sinuri pres de Mylasa, 1: Les inscrip-
tions grecques, Paris 1945.
Robert (1946a)-Robert, L., Hellenica 2, Paris 1946.
Robert (1946b)-Robert, L., Hellenica 3, Paris 1946.
228 Bibliography

Robert (1948a)-Robert, L., Hellenica 5, Paris 1948.

Robert (1948b)-Robert, L., Hellenica 6, Paris 1948.
Robert (1949)-Robert, L., Hellenica 7, Paris 1949.
Robert (1950)-Robert, L., Hellenica 9, Paris 1950.
Robert (1953a)-Robert, L., Les fouilles de Claros. Conference donnee al'univer·
site d'Anlr.ara ... 1953 . .. , Limoges 1958.
Robert (19511b)-Robert, L., "Le sanctuaire d'Artemis a Amyzon," CRAI 1958,
4011-415.
Robert (1955)-Robert, L., Hellenica 10, Paris 1955.
Robert (1958a)-Robert, L., "Inscriptions grecques de Side en Pamphylie (epoque
imperiale et Bas-Empire)," Rev. de philol. 112 (1958) 15-511.
Robert ( 1958b)-Robert, L., "Reliefs votifs et cultes d' Anatolie," Anatalia (=Ana-
dolu) 3 (1958) 103-136.
Robert (1960a)-Robert, L., Hellenica 11/12, Paris 1960.
Robert (1960b)-Robert, L., "Recherches epigraphiques, VII: Deaet de Ia Con-
federation lycienne a Corinthe," REA 62 (1960) 824-1142.
Robert (1964)-Robert, L., in N. Firatli, Les steles funeraires de Byzance greco-
romaine avec I'edition et l'index commente des epitaphes par L. Robert, Paris
1964.
Robert (1965)-Robert, L., Hellenica 18, Paris 1965.
Robert (1966)-Robert, L., Monnaies antiques en Troade, Geneva 1966.
Robert (1967)-Robert, L., Monnaies grecques. Types, Legends. Magistratsmone·
taires et Geographie, Geneva and Paris 1967.
Robert (1968)-Robert, L., "Trois oracles de Ia Theosophie et un prophete d' Apol-
lon," CRA/1968, 568-599.
Robert (1969)-Robert, L., Laodicee du Lyco.s. Les inscriptions (Laodicee du
Lycos. Campagnes 1961-1968: Le nymphee, by J. DesGagniers et al.), Quebec
and Paris 1969, 247-1189.
Robert (197la)-Robert, L., "Un oracle grave a Oinoanda," CRA/1971, 597-619.
Robert (1971 b)- Robert, L., "Les colombes d' Anastase et autres volatiles," ] oumal
des Savants 1971, 81-97.
Robert (19711)-Robert, L., "De Cilicie a Messine eta Plymouth avec deux inscrip-
tions grecques errantes," Journal des Savants 1973, 161-211.
Robert (1975)-Robert, L., "Une nouvelle inscription grecque de Sardes; Regle-
ment de l'autorite perse relatif aun culte de Zeus," CRA/1975, 1106-1130.
Robert and Robert (1954)-Robert, L. and J., La Carie. Histoire et geographie his-
torique avec le receuil des inscriptions antiques, II: Le plateau de Tabai et ses
environs, Paris 1954.
Robinson (19511)-Robinson, D. M., "A magical inscription from Pisidian Anti-
och," Hesperia 22 (1958) 172-174.
Roebuck (1951 )-Roebuck, C. A., The Aslr.lepieion and Lema ... (Corinth: Results
of Excavations Conducted by the Am. School of Class. Studies at Athens 14),
Princeton 1951.
Rose (1924)-Rose, H. J., The Roman Questions of Plutarch ... , Oxford 1924.
Rostovtzeff (19117)-Rostovtzeff, M., "Report sur les fouilles de Doura-Europos,
campagne de 1986-1987," CRA11987, 195-204.
 Bibliography 229

Roussel (1916)-Roussel, P., Delos, colonie athenienne, Paris 1916.

Roussel (1927)-Roussel, P., "Les mysteres de Panamara," BCH 51 (1927) 123-137.
Roussel (1931)-Roussel, P., "Le miracle de Zeus Panamaros," BCH 55 (1931)
70-ll6.
Roussel and de Visscher (1942-1943)-Roussel, P.,andF. de Visscher, "Lesinscrip-
tions du temple de Dmeir," Syria 23 (1942-1943) 173-200.
Safrai et al. (1974)-Safrai, S., et al., eds., The Jewish People in the First Century ... ,
I, Philadelphia 1974.
Sagnard (1948)-Sagnard, F., Clement d'Alexandrie. Extraits de Thiodote(Sources
chretiennes 23), Paris 1948.
Sabin (1978)-Sahin, S., "Zeus Bennios," in Studien zur Religion und Kultur
Kleinasiens. Festschrift fiir Friedrich Karl Dorner . .. , Leiden 1978, 2.771-790.
Salac (1956)-Salac, A., "Zwei vorislamische Gottheiten auf einer romischen In-
schrift," in Carnuntina. Ergebnisse der Forschung iiber die Grenzprovinzen
des romischen Reiches. Vortriige beim internationalen Kongress der Alter-
tumsforscher Carnuntum 1955, ed. E. Swoboda, Graz 1956, 168-170.
Salviat (1963)-Salviat, F., "Dedicace d'un tpuq>aJCto~ par les Hermaistes de liens,"
BCH 87 (1963) 252-264.
Scherer (1960)-Scherer, J., Entretien d'Origene avec Heraclide (Sources chreti-
ennes 67), Paris 1960.
Schermann (1909)-Schermann, T., Griechische Zauberpapyri und das Gemeinde
und Dankgebet im II. Klemensbriefe (Texte und Untersuchungen 34, 2b),
Leipzig 1909.
Schleiermacher (1966)-Schleiermacher, W., "Zu den sogenannten Kulttheatern in
Gallien," in Corolla memoriae E. Swoboda dedicata, Graz and Cologne 1966,
205-213.
Schmidt (1881)-Schmidt, J., "Aus Constantinopel und Kleinasien," Ath. Mitt. 6
(1881) 132-153.
Schuchhardt (1886)-Schuchhardt, C., "Kolophon, Notion und Klaros," Ath. Mitt.
11 (1886) 398-434.
Schwertheim (1974)-Schwertheim, E., Die Denkmiiler orientalischer Gottheiten
im romischen Deutschland mit Ausnahme der iigyptischen Gottheiten, Leiden
1974.
Schwertheim (1975)-Schwertheim, E., "Ein neues Weihrelief fiir Men und seine
Mutter aus Lydien im Museum von Izmit," Mitt. des Deutschen Arch. Inst.,
lstanbul25 (1975) 357-365.
Schwertheim (1978)-Schwertheim, E.• "Denkma1er zur Meterverehrung in Bi-
thynien und Mysien," in Studien zur Religion und Kultur Kleinasiens. Fest-
schrift fiir Friedrich Karl Dorner ... , Leiden 1978, 2.791-837.
Schwertheim (1979)-Schwertheim, E., "Mithras. Seine Denkma1er und sein Kult,"
Antike Welt 10 (1979) 2-76.
Scranton (1944)-Scranton, R. L., "Two temples of Commodus at Corinth," Hes-
peria 13 (1944) 315-348.
Scroggs (1980)-Scroggs, R., "The sociological interpretation of the New Testa-
ment: the present state of research," New Testament Studies 26 (1980) 164-179.
Seston (1956)-Seston, W., "La conference de Carnuntum et le 'dies imperii' de
2JO Bibliography

Licinius," in Carnuntina. Ergebnisse der Forschung ii.ber die Grenz.provinzen

des romischen Reiches. Vortriige beim internationalen Kongress der Alter-
tumsforscher Carnuntum 1955, ed. E. Swoboda, Graz 1956, 175-186.
Seyrig (1941)-Seyrig, H., "Antiquites syriennes 38: Inscriptionsgrecquesde l'agom
de Palmyre," Syria 22 (1941) 223-270.
Seyrig (1970)-Seyrig, H., "Antiquites syriennes 89: Les dieux armes et les Arabes
en Syrie," Syria 47 (1970) 77-112.
Seyrig (197la)-Seyrig, H., "Antiquites syriennes 93: Bel de Palmyre; 94: Quatre
images sculptees du Musee d'Alep," Syria 48 (1971) 85-120.
Seyrig (197lb)-Seyrig, H., "Antiquites syriennes 95: Le culte du soleil en Syrie a
l'epoque romaine," Syria 48 (1971) 337-373.
Simon (1972)-Simon, M., "Theos Hypsistos," in Ex orbe religionum: Studia Geo
Widengren oblata, Leiden 1972, 1.372-385.
M. Smith (1971)-Smith, M., "Prolegomena to a discussion of aretalogies, divine
men, the gospels and Jesus," ]ourn. Bibl. Lit. 90 (1971) 174-199.
M. F. Smith (1978)-Smith, M. F., "Fifty-five new fragments of Diogenes of Oeno·
anda," Anatolian Studies 28 (1978) 39-92.
M. F. Smith (1979)-Smith, M. F., "Eight new fragments of Diogenes of Oenoanda,"
Anatolian Studies 29 (1979) 69-89.
Snyder (1940)-Snyder, W. F., "Public anniversaries in the Roman empire ... ,"
Yale Class. Studies 7 (1940) 223-317.
Sodano (1958)-Sodano, A. R., Porfirio, Lettera ad Anebo, Naples 1958.
Sokolowski (1955)-Sokolowski, F., Lois sacrees de l'Asie Mineure, Paris 1955.
Sokolowski (1962)-Sokolowski, F., Lois sacrees des cites grecques. Supplement,
Paris 1962.
Sokolowski (1969)-Sokolowski, F., Lois sacrees des cites grecques, Paris 1969.
Speidel (1978)-Speidel, M.P., The Religion of Iuppiter Dolichenus in the Roman
Army, Leiden 1978.
Sperber (1966)-Sperber, D., "Costs of living in Roman Palestine," journal of
Econ. and Soc. Hist. of the Orient 9 (1966) 182-211.
Spinazzola (1953)-Spinazzola, V., Pompei alla luce degli scavi nuovi di Via del-
l'Abbondanz.a, 2 vols., Rome 1953.
Stahelin (1948)-Stiihelin, F., Die Schweiz im romischer Zeit, 5 Basel 1948.
Stambaugh (1978)-Stambaugh, J. E., "The functions of Roman temples," ANRW
II 16, 2 (Berlin 1978) 554-608.
Starcky (1949)-Starcky, J., "Autour d'une dedicace palmyrenienne a Sadrafa eta
Du'anat," Syria 26 (1949) 43-85.
Stein (1925)-Stein, A., "Zu Lukians Alexandros," in Strena Buliciana. Commen·
tationes gratulatoriae in honore F. Bulic, Zagreb 1925, 257-265.
Steinleitner (1913)-Steinleitner, F., Die Beicht im Zusammenhange mit der sakra-
len Rechtspflege in der Antike ... , Leipzig 1913.
Stocks (1937)-Stocks, H., "Studien zu Lukians De Syria dea," Berytus 4 (1937)
1-40.
Storch (1968)-Storch, R. H., "The 'absolutist' theology of victory: its place in the
Late Empire," Classica et mediaevalia 29 (1968) 197-206.
Strohm (1952)-Strohm, H., "Studien zur Schrift von der Welt," Mus. Helv. 9
(1952) 137-175.
 Bibliography 231

Siiss (1972)-Siiss, L., "Ein gallo-romischer Achtecktempel in Friedberg," Ge-r-

mania 50 (1972) 153-174.
Swoboda (1964)-Swoboda, E., Carnuntum. Seine Geschichte und seine Denk-
miiler4, Graz and Cologne 1964.
Syme (1972)-Syme, R., "Lawyers in government: the case of Ulpian," Proc. Am.
Philos. Soc. 116 (1972) 406-409.
Szilagyi (1961 )-Szilagyi, J., "Beitrage zur Statistik der Sterblichkeit in den west-
europaischen Provinzen des romischen Imperiums," Acta arch. acad. sci.
Hung. 13 (1961) 125-156.
Tate (1929)-Tate, J., "Cornutus and the poets," CQ 23 (1929) 40-45.
Tate (1930)-Tate, J., "Plato and allegorical interpretation," CQ24 (1930) 1-10.
Taus and Berardi (1972)-Taus, G., and P. Berardi, "Iscrizioni 1atine inedite di
Roma ne1 Cenobio diS. Paolo," Epigraphica 34 (1972) 77-88.
Teixidor (1977)-Teixidor, J., The Pagan God: Popular Religion in the Greco·
Roman Near East, Princeton 1977.
Ternes (1965)-Ternes, C. M., "Les inscriptions antiques du Luxembourg," He·
mecht. Zeitschrift fiir Luxemburger Geschichte 17 (1965) 267-481.
Ternes (1973)-Ternes, C. M., "Les sanctuaires des Trevires," Caesarodunum 8
(1973) 6-15.
Themanakis (1893)-Themanakis, "Marbres decouvertes afiaA.ui," BCH 17 (1893)
208-209.
I. T6th (1971)-T6th, 1., "Sacerdotes Iovis Dolicheni," Studium 2 (1971) 23-28.
L. T6th (1974)-T6th, L., "Eine Doppelheit der Geschichte des Isis- und Sarapis-
kultes in Pannonien," in Studia Aegyptiaca: Recueil d'etudes ... aV. Wessetz·
ky, Budapest 1974, 345-360.
S. T6th ( 1967)-T6th, S., "Zur Frage des Ursprungs und des sozia1en Hintergrunds
des Silvankultes in Dazien," Acta class. univ. scient. Debrecen. 3 (1967) 77-
84.
Toutain (1892)-Toutain, J., "Le sanctuaire de Saturn us Balcaranensis au Djebe1
Bou-Kournein (Tunisie)," MEFR 12 (1892) 3-124.
Toutain (1907-1920)-Toutain, J., Les cultes paiens dans l'empire romain, 1: Les
provinces latines, parts 1-3, Paris 1907-1920.
Toynbee (1944)-Toynbee, J. M. C., Roman Medallions (Numismatic Studies 5),
New York 1944.
Tran Tam Tinh (1964)-Tran Tam Tinh, V., Essai sur le culte d'Isis Pompei,a
Paris 1964.
Tran Tam Tinh (1971)-Tran Tam Tinh, V., Le culte des divinites orientales a
Herculanum, Leiden 1971.
Tran Tam Tinh (1972)-Tran Tam Tinh, V., Le culte des divinites orientales en
Campanie en dehors de Pompei, de Stabies et d'Herculanum, Leiden 1972.
Troiani (1974)-Troiani, L., L'opera storiograjica di Filone da Byblos, Pisa 1974.
Tudor (1969-1976)-Tudor, D., ed., Corpus monumentorum religionis equitum
Danuvinorum, 2 vo1s., Leiden 1969-1976.
Turcan (1975)-Turcan, R., Mithras Platonicus. Recherches sur l'hellenisation
philosophique de Mithra, Leiden 1975.
Turcan (1978)-Turcan, R., "Note sur Ia liturgie mithriaque," Rev. de l'hist. des
religions 194 (1978) 147-157.
2J2 Bibliography

VanWinden (1971)-VanWinden, J.C.M., An Early Christian Philosopher. justin

Martyr's Dialogue with Trypho, Chapters One to Nine . .. , Leiden 1971.
Velkov (1974)-Ve1kov, V., "Der kult der Diana Plestrensis in Moesia inferior," in
Actes du JXe Congres international d'etudes sur les frontieres romaines . ..
1972, Bucharest 1974, 151-153.
Vermaseren (1956-1960)- Vermaseren, M. J., Corpus inscriptionum et monumen-
torum religionis mithriacae, 2 vols., The Hague 1956-1960.
Vermaseren (1966)-Vermaseren, M. J., The Legend of Attis in Greek and Roman
Art, Leiden 1966.
Vermaseren (1969)-Vermaseren, M. J., "Paganism's death struggle. Religions in
competition with Christianity," The Crucible of Christianity ... , ed. A. Toyn-
bee, London 1969, 235-260.
Vermaseren (1971)-Vermaseren, M. J., Mithriaca 1: The Mithraeum at S. Maria
Capua Vetere, Leiden 1971.
Vermaseren (1977)-Vermaseren, M. J., Cybele and Attis, the Myth and the Cult,
trans. A.M.H. Lemmers, London 1977.
Vermaseren and van Essen (1965)-Vermaseren, M. J., and C. C. van Essen, The
Excavations in the Mithraeum of the Church of Santa Prisca in Rome, Leiden
1965.
Veyne (1967)- Veyne, P., "Au tour d'un commentaire de Pline le Jeune," Latomus
26 (1967) 723-751.
Vidman (1965)- Vidman, L., "Die Isis- und Sarapisverehrung im 3. Jhdt. u. Z.," in
Neue Beitriige zur Geschichte der alten Welt, II: Romisches Reich, Berlin 1965,
389-400.
Vidman (1966)-Vidman, L., "Trli~r des Isis- und Sarapiskultes in der romischen
Provinzen," Eirene 5 (1966) 107-116.
Vidman (1969)- Vidman, L., Sylloge inscriptionum religionis lsiacae et Sarapia-
cae, Berlin 1969.
Vogliano (1933)-Vogliano, A., "La grande iscrizione bacchica del Metropolitan
Museum," A]A 37 (1933) 215-231.
Vorbeck and Beckel (1973 )-Vorbeck, E., and L. Beckel, Carnuntum. Rom an der
Donau, Salzburg 1973.
Vuckovic-Todorovic (1964-1965 )-Vuckovic-Todorovic, D., "Le sanctuaire de J u-
a
piter Dolichenus Brza Palanka," Starinar 15/16 (1964-1965) 173-182.
Waltzing (1895-1900)-Waltzing, J.P., Etude historique sur les corporations pro-
fessionnelles chez les Romains . .. , 4 vols., Louvain 1895-1900.
Wagenvoort (1966)-Wagenvoort, H., "Auguste et Vesta," in Melanges d'archeo-
logie, d'epigraphie et d'histoire of/erts a ]. Carcopino, Paris 1966, 965-
978.
Warmington (1954)- Warmington, B. H., The North African Provinces from Dio-
cletian to the Vandal Conquest, Cambridge 1954.
Weber (1911)- Weber, W., Drei Untersuchungen zur iigyptisch-griechischen Re-
ligion ... , Heidelberg 1911.
Weinreich ( 1909)-Weinreich, 0., Antike Heilungswunder. Untersuchungen zum
Wunderglauben der Griechen und Romer, Giessen 1909.
Weinreich (1912)-Weinreich, 0., "O&oi tm\ICOOl," Ath. Mitt. 37 (1912) 41-68.
 Bibliography 2JJ

Weisgerber (1975)-Weisgerber, G., Das Pilgerheiligtum des Apollo und der Sirona
von Hochschied im Hunsriiclt, Frankfurt 1975.
West (1971)-West, S., "A new reading in PSI XI 121l=Aeschylus, Myrmidons,"
ZPE 7 (1971) 9~-96.
Wey (1957)-Wey, H., Die Funlttionen der bosen Geister bei den griechischen
Apologeten des 2. ]ahrhunderts n. Chr., Winterthur 1957.
Wheeler and Wheeler (19~2)-Wheeler, R. E. M. and T.V., Report on the Excava·
tion of the . .. Site of Lydney Parlt, Gloucestershire, Oxford 19~2.
Wiegand (1982)-Wiegand, T., Zweiter Bericht tiber die Ausgrabungen in Perga·
mon 1928-1932: Das Asltlepieion, Berlin 1982.
Wikander (1950)-Wikander, S., "Etudes sur les mysteres de Mithras," Arsbolt,
Vetensltaps-Societeten i Lund 1950, 5-46.
Wilcken (1885)-Wilcken, U., "Arsinoitische Tempelrechnungen aus dem j. 215
n. Chr.," Hermes 20 (1885) 480-476.
Wilcken (1912)-Wilcken, U., Grundziige und Chrestomathie der Papyrusltunde,
2 vols., Leipzig and Berlin 1912.
Wild (1968)-Wild, H., "Les danses sacrees de l'Egypte ancienne," in Les danses
sacries . .. (Sources orientales VI), Paris 1968, 88-118.
Wilhelm (1948a)-Wilhelm, A., "Beschluss zu Ehren des Demetrios 6 J.ttrac;," ]OAI
85 (194~) 157-163.
Wilhelm (1943b)-Wilhelm, A., "Zwei Inschriften auf Didyma," ]OA/35 (1943)
164-169.
Will (1951)-Will, E., "Le sanctuaire syrien de Delos," Annales arch. de Syrie I
(1951) 59-79.
Will (1975)-Will, E., "Le rituel des Adonies," Syria 52 (1975) 93-105.
Williams (1978)-Williams, G., Change and Decline: Roman Literature in the
Early Empire, Berkeley 1978.
Wilson (1973)-Wilson, D. R., "Temples in Britain. A topographical survey,"
Caesarodunum 8 (1978) 24-44.
Windisch (1920)-Windisch, H., "Der Barnabasbrief," in H. Lietzmann, Die apo-
stolischen Vater, Tiibingen 1920, 299-413.
Winnett (1957)-Winnett, F. V., Safaitic Inscriptions from jordan, Toronto 1957.
Wiseman (1971)-Wiseman, j., "Excavations at Stobi, 1970," A]A 75 (1971) 395-
411.
Wiseman (1973)-Wiseman, j., "Gods, war and plague in the time of the Anto·
nines," in Studies in the Antiquities of Stobi, ed. j. Wiseman, I, Belgrade 1973,
143-188..
Wissowa (1912)-Wissowa, G., Religion der Romer! (Handbuch der klass. Alter-
tumswiss. 5, 4), Munich 1912.
Witt (1971)-Witt, R. E., Isis in the Graeco-Roman World, London 1971.
Witt (1975)-Witt, R. E., "Some thoughts on Isis in relation to Mithras," in Mith·
raic Studies. Proceedings of the First International Congress of Mithraic Stud·
ies, ed. J. R. Hinnells, Manchester 1975, 2.479-493.
Witt (1979)-Witt, R. E., "john Dillon: The Middle Platonists ... ," Gnomon 51
(1979) 882-385.
Wolters (1914)-Wohers, P., "'AlCoa{," Hermes 49 (1914) 149-151.
214 Bibliography

Wortmann (1968)-Wortmann, D., "Neue magische Texte," B]bb 168 (1968) 57-
Ill.
Wortmann (1969)- Wortmann, D., "Ein Mithrasstein a us Bonn," B]bb 169 (1969)
410-423.
Wright (1895)-Wright, J. H., "A votive tablet to Artemis Anaitis and Men Tia·
mu ... ," HSCP 6 (1895) 55-74.
Wunsch (1909)-Wiinsch, R., "Deisidaimoniaka," Archiv fur Religionswiss. 12
(1909) 1-45.
Yoyotte (1960)-Yoyotte, J., "Les pelerinages dans l'Egypte ancienne," in Les
pelerinages (Sources orientales 3), Paris 1960, 18-74.
Zotovic (1968)-Zotovic, L., "Les conditions historiques du developpement des
cultes orientaux dans les provinces romaines du territoire de la Yougoslavie,"
Starinar 19 (1968) 59-74.
Zwicker (1934-1936)-Zwicker, J., Fontes historiae religioniscelticae, 3 parts in one
vol. (paged continuously), Berlin 1934, Bonn 1935, 1936.
 Index

Abonuteichus. See Alexander of Abonuteichus Apollodorus, 10

Abraham, 92, 94 Apollonius of Tyana, 42, 92
Achaea, 30 Apologists, Christian, xii, 20, 75, 77, 96, 182n9
Achilles, 35, 60 Apu1eius, 15, 83, 97
Adamna, 18 Apulia, 114
Adonis, 18, 37, 91, 153n33 Apulum, 104, 110, 159n79
Aegeai, II, 158n75 Aquileia, 4, 36, 99, 115, 159n79
Aelius Aristides, 15, 42, 60, 97 Arabia(n), 84, 99, 143nl8
Africa, 5, 7, 81, 116, 133 Arausio, 104
Aglibol, 4, 40, 93 Arcadia, 32, 173n33
Ainigmata, 78, 183nl9, 187n49 Arcadius, 134, 135
Alexander of Abonuteichus, 72, 98, 106, 121, Aretalogy, 14, 68, 148n67, 178n30
122, 178n25 Argos, 10, 143n18
Alexander Rhetor, 17 Ariphron, 15
Allegorical interpretation, 78-79 Army. See Soldiers
Altbachtal, 20 Arrian, 60
Ammon, 11, 68 Arsinoe, 100, 153n33
Ammonius, 11, 68 Arsou, 142nl8
Amphilochus, 55, 62 Art, religious, II, 30, 31, 36, 41, 91, 156nn61-
Amyzon, 19 63, 165nn40 and 44, 189n62
Andania, 12, 165n44 Artemidorus, 60, 156n55
Angels, 61, 79, 82, 110, 184n22, 185nn28 and Artemis,6, 7,19,21,92,107,169n9,andpassim
31 in notes
Animal gods, 75, 76, 142n17, 181n4, 182n12, -Anaitis, 32
183n19 -of Ephesus, 20, 21, 26, 35, 37, 42, 106, 107,
Anna Perenna, 24 142nl7, and passim in notes
Antinous, 60, 91, 102 -Nanaia, 19
Antioch (Pisidia), 92, 142n17 -of Perge, 4, 26, 52, 142n17, 143n22
Antioch (Syria), 27, 100, 133, 143n27 - Thermia, 31
Antoninus Pius, 103, 127, 194n36 Arval Brethren, 35, 36, 45, 106, 128, 161nl2
Anubis, 1, 22 -- Ascalon, 35
Aoth, 185n33 Asceticism, 12, 43, 55, 63, 73, 97, 146n51, 147
Apamea, 25 n63, 166n7, 191n5
Aphrodisias, 35, 143n19 Asclepius, 4, 6, 7, 15,16, 30,35,42,49,61,110,
Aphrodite, 7, 35, 36, 60, 90, 142nl7 120, 133, 143n25 and passim in notes
Apis, 1 -at Epidaurus, 5, 20, 35, 42, 66, 107
Apollo, 6, 7, 78, 82, 86, 88, 93, 100, 108, 142 -at Pergamon, 4, 15, 16, 20, 21,27,28,42,66,
n17 and passim in notes 143n25 and passim in notes
-Lairbenos, 32 Astarte, 5, 90
-of Glaros, 11, 12, 13, 17, 62, 63, 77. 87, 92, Atargatis, 19, 24, 34, 35, 36, 73, 143n18
108, 157n67 and passim in notes Atheism, 2, 62, 63, 67, 141n10
-of Didyma, 13, 40, 56, 72, 102 Athena, 1, 7, 90, 148n73
See also Delphi Athens, 12, 24, 42, 108, 133, 143n25

215
2.16 Index

Attica, .'12 Coins, imperial, advertise cults, 4, 84, 85, 86,

Attis, 1, 18, 55, 80, 84, 91, 92, 147n62, and 92, 127
passim in notes -municipal, 25, 81, 151nl8, 201nl9
Augustus (-a), 103, 104, 169n7 Colophon, 17, 157n67
Aurelian, 85, 86, 127 Comana (Cappadocia), 100
Aventine, 1.'1, 91, 101 Comana (Pontus), 100
Aziz, 84 Comum, 4
Concord, I08
Ba'al, 5, 16lnl2, 162nl8 Constantine, 43, 71, 85, 110, 127, 132, 137,
Baalbek, 34, 35, 84, 101, Ill, 166n7 165nl
Baal-shamin, 19, 109 Coptos, 32
Bacchus. See Dionysus Corinth, 37
Baliddiris, 142nl7 Cornutus, 10, 145n42
Bathing of images, 27, 44, 45, 154n41 Corybas, 18
Bebelahamon, .'18 Cult associations, in eastern provinces, 12,
Beggars, 24, 37, 42, 48, 98, 192nl2 37, 143nl8, 146nn57-58, 163nn23-24
Bel, 37, 81, 147n66, 187n45 -in Italy and west, 39, 93, 109, 124, I95n41
Belenus, 4, 5, 192nl3 -meals and drinking in, 38, 39, 47, 57, 161-
Belhammon, 109 164 notes, passim
Bellona. See Ma -rules for, 13, 41, 104, 105, 146n57, 16In9,
Bendis, I 164nn36-38
Benefal, 38 Cumont, Franz, xii, 80, 116, 122, 123. 200nll
Birds, sacred, 35, 147n64 Cybele (the Great Mother), 6, 7, 1I, 16, 21. 24,
Bithynia, 34, 41, 84, 104, I69n9 35, 90, 93, 99, 100, 104, 106, 116, 120
Bona Dea, 101, 159n79 Cynic(ism), 62, 156n55
Britain, 16, 153n33
Byblos, 106, 154n38 Dacia, 38, 69, 99, 104, 113, 114, 142nl8
Dalmatia, 1.'1, 127
Cabiros, 20, 142nl7 Damascus, 26, 108, 162nl8
Cadiz, 29 Dancing, cult, 20-25 passim, 43, 44, 152n22,
Caelestis, 5, 85, 108, 157n63 163n25
Caesarea (Palestine), 95 Daphne, 27
Caesarea (Pontus),, 95 Dea Dia, 35, 16ln!O
Calendars, religious, 43, 47, 93, 100, 106, 107, Delos, 19, 23
110, 19.'1n27, 197n68 Delphi, II, 29, 82
Campania, 114 Demeter, 12, 14, 72, 90, 106
Canopus, 100, 107, 158n77 Demons, 79, 82, 96, 157n66
Capitol (Capitolium), 35, 44, 99, 100, 109, Diana, 13, 22, 198-99n4
157n66 -Plestrensis, 198n70
Capitoline Triad, 1, I 10, 127 -of Tifata, 113, 197n64
Capua, 34, 160n2 Didyma, 4, 21, 142nl7
Caracalla, 4, 100 Dii Mauri, 14In2, 142nl7
Caria, 4, 19, 35, 73, 90 Dio Chrysostom, .'II, 74
Carnuntum, 4, 36, 118, 157n64 Diocletian, 4, 127, 174n40
Carthage, 8, 19, 84, 156n63 Diogenes of Oenoanda, II, 62
Celtic, 6, 28, 41, 91, 130, 165n39 Diogenianus, 10
Ceres, 92, 109 Dionysus (Liber Pater, Bacchus), 6, 14, 15, 19,
Christ, 81, 82, 84, 92, 94 21, 44, 93, 109, 124, 142nl7, and passim in
Christians, 2, 39, 40, 62, 129, 1.'1.'1, 144n34 notes
Cicero, 59, 70, 101 -fertility god, 52, 87
Cilicia, 90, liS, 142nl7 -identified with other gods, 84, 86, 91
Clitumnus, 33 Dioscuri, 60, 81, 91, 184n26
Cnidus, 36 Dodona, 56
 Index 237

Doliche, 38, 192n17 Hadrian, 4, 102, 107, 108, 149n77. 184n24,

Domitian, 103, 194n39 194nn!16-!17
Dougga, 108 Haemon, 18
Dreams, 17, 32, 49, 60, 61, 67, 9!1 Hawran, 19
Druids, 2 Healing, 15, 42, 49, 51, 57, 60, 67, 158n71,
Ducavavius, 1 158nn75-79, 160n82, 177n21
Dura, 19, !17, 80, 81, 119 Hecate, 88, 90, 92, 120, 142nl7
Dusares, 1, 5, 99, 160n!l -of Lagina, 14, 21, 46, 47, 48, 106
Helios, 80, 82, 142nl7, and passim in notes
Echo, 14 -identity with other gods, !18, 51, 58, 8!1, 84,
Egypt, 21, !18, 68, 14!1n 18, and passim in notes 86, 87, 88
Elagabalus, 85, 86, 104, 14!1n18 Hephaestus, 75, 18ln4
Eleusis, 5, 20, 21, 24, 27, 156n6!1 Hera, 1, 7, 22, 46, 78, 1!1!1, 142nl7
Elis, 10 Heraclitus, 10, 76
Emesa, 84, 85, 86 Herculaneum, 107, 156n6!1
Emona, 1!1!1 Hercules, 7, 9!1, 104,127,142nl7, and passim in
Emperor worship, 17, 46, 100, 105, 149n86, notes
194n!l6 Hermes, 16, 17, 58, 68, 162n 17, and passim in
Ephesus, 105. See also Artemis notes
Epicurean(s), 10, 56, 62 Heroes, 11, 60, 61, 82, 175n4!1, 19ln2, 195n51
Epidaurus, 14!1nl8. See also Asclepius Hesiod, !1, 8, 68, 78. 145n41
Epiphanies, !12, 46, 66, 159n81, 167nl8 Hestia, 14, 90, 170nl2
Epona, 91 Hierapolis (the Holy City), 28, 29, !15, 106,
Eros (Cupid), 16, 36, 157n6!1 164n!l2
Etepancus, I Hierocaesarea, 99
Etruria, 114 Hispellum, 105
Eutecnius, Julius, 97, 98 Homer, !1, 9, 68, 74, 78, 145n41
Exegetes, 29, 30, 32, !13, 60, 61, 144n!16 Horus, I
Exorcism, 50, 95, 96 Hostels, 28, 42, 105, 155nn49-50
Hygieia (Hygia), 12, !19, 48, 52, 124, 125, 160nl
Festivals, 25-29 passim, 40, 47, 146n51 Hymnodes, 17, 100, 149n84, 19!1n28
Fish, sacred, 12, 35, 36 Hymns, 15-24 passim, 44, 45, 48, 91,100,102,
Fortuna, 7, 109, 14ln2 178n!IO, 184n2!1, 19!1nl9, 202n28
-Primigenia, 14!1n26, 157n63, 197n64
Fregellae, 25 lacchus, 14
Fronto, 67, 143nl8 lamblichus, 75, 79, 19ln2
larhibol, !12, 9!1, 142nl7, 158n70
Gaul(s), 8, 18, 97, 117 Iciodurum, 109
Gaza, 95, 133, 14!1nl8 lguvium, 11!1
Genius of the Roman People, 127 Iliad, !I, 145n41
Gerasa, 19, l69n7 Ilium, 4, 143n25
Germany, 119, 144n29 Immortality (after-life), 1!1, 53, 54, 55, 56, 123,
Glycon, 121, l22,198nl. See also Alexander of 1!16
Abonuteichus Imperial cult. See Emperor worship
Gnosticism, 68, 79, 98, 168n2 Initiation, 18, 20, 5!1, 54, 56, 58, 145n47
Gorgos, 11 Inscriptions, painted, !II, !12, 157n64, 159n78
Gorsium, 99 -recut, 12, !II, 157n65
Graces, 120 -statistics of, 5, 6, 7, 114-19 passim, 144n!ll
Grand, 19, 17!1n3!1 lntercisa, 11!1, 143nl8
Groves, sacred, !15, 36, 161nn9-ll, 166n!l Ionia, 23
Guilds. See Cult associations Isis, 6, 27, 69, 80, 90
-aretai of, 32, 33, 49, 53, 58
Hadad, 19 -cult in Egypt, 29, 68
2J8 ltulex

Isis (continued) Mactar, 55

-cult elsewhere in east, 120, 143nl8 Magic, 9, 51, 70, 72, 82, 83, 84, 203n40
-in west, 91, 93, 99, 101, 103, 107-108, 114- Mainz, 117, 199n6, 201n16
17, and passim in notes Maioumas, 19, 21
Malakbel, 4, 38, 40, 162n20
jerusalem, 28 Manaval, 38
jews (and judaism), 2, 3, 40, 58, 98, 141n9, Marcus (heretic), 95, 184n22
192n9 Marcus (missionary), 95
Julian, 132, 133, 135 Marcus Aurelius, 66, 67, 94, 95, 143n25
juno, 1, 7, 44, 80, 91 Mars, 4, 6, 16, 109, I 10, 127, l43n23, 166nl2,
-of Populonia, 113 and passim in notes
-Sospita, 107 - Iovantucarus, 52
jupiter, 5, 7, 13, 22, 44, 51, 69, 81, 113, 127, -Lenus, 117, 177n11
198nn2-3 -Segomo, I 18
-Apenninus, 113 Matuta, 69
- Axoranus, 39 Mauretania, 16, 190n68
-Capitolinus, 100 Maximin Daia, 85, 89
-Dolichenus, 7, 38, 61, 66, 70, 81, 91, 92, 93, Megalodemonia, 83, 89, 184n23
99, 114, 116, 118, 127, 141n2 Melkart, I
-the Free, 11 Men, 25, 32, 80, 92, 153n32, and passim in
-Heliopolitanus, 111, 147n66, 198n70 notes
juvenal, 53 Menander Rhetor, 17
Mercury, 6, 7, 52, 141n2, 170nll, 199n6
Kakasbos, 4 -Cissonius, 92
Karpokrates, 14 Mesomedes, 15
Kastel, 100 Metempsychosis, 54, 55, 56, 170nl3
Kore, 12, 14, 55, 90, 101, 147n59 Milan, 117, I 73n33
Kronos, 18, 69, 78 Miletus, 13, 62, 143n18
Minerva, I, 4, 44, 51, 92, 109, 141n2
Labeo, Cornelius, 87 Miracles (except healing), 32, 33, 59, 60, 69-
Labraunda, 35, 162n17 70, 73, 96, 167nnl7 and 22, 175n42, 177nl6,
Lagina, 21, 108, 167n19 191nn1-2
Lambaesis, 7, 109, 110, 116, 141n2 Missionaries, 97, 98, 99, 110, 111, 114, 192nn9
Laodicea, 97 and 12
Lebanon,4 Mithra, 101, 106, 107, 122-27 passim, 141n2,
Lectistemia, 39, 45, 149n83, 151n14 and passim in notes
Leto, 4, 5, 90 -beliefs of, 54, 58, 84, 92, 122
Lex sacra. See Cult, rules of; Sacrifices, rules -cult practices of, 39, 45, 53-54, 99, 122
for -in Africa, 109, 110, 198n70
Liber Pater. See Dionysus -in Italy, I, 7, 83, 99, 118-19
Libraries, 11, 68 -on Danube, 91, I 10, 118-19, 133
Licinius, 85, 110, 127, 197n69 -in east, 7, 118, 119
Limyra, 25, 153n32 -on Rhine, 4, 116, 119, 198n10
Love, divine, 53, 75, 98, 177nll Moesia, 117
Lucian, I, 74, 75, 181n4 Monotheism, 83-89 passim, 110, 187n45
Luna, 1,88,90 Mopsos, I, 96
Lycaonia, 58 Moses, 3
Lycia, 90, I65n2 Mother (goddess), 4, 5, 15, 28, 32, 73, 143n22
Lydia, 29, 32, 35, 58 Music, 16, 20, 22, 24, 38, 40, 48, 152n28
Lyon, 25, 62, 97, 100, I 16, 144n29, 193n25 -instruments for, 15, 16, 23, 24
-notation for, 17, 148n68
Ma, 25, 43, 64, 90, 100, 166n4, 189n56, and Myndrytus, I
passim in notes Mysia, 32, l43nl9
 Index 239

Mysteries, 2!1, !19, 58, 106, 145n!l8, 172nn19-20 Pausanias, 20, !12, 78
-of Asclepius, 42 Pautalia, J!l!l
-of Cybele, 2!1 Pentateuch, !l
- Eleusinian, 24 Perasia, 90
-of Glycon, 122 Pergamon, 28, 29, 31, 100, 106. See also As·
-at Panamara, 45, 57 clepius
Myths (life-stories of gods), 14-15, 17, 18, 56, Perge. See Artemis
64, 67, 69, 75-78 passim, 87, 101-02, 106, Petra, 99
19!ln!l0, 194n!l5 Philae, 29, l65n44
Phileas, 95, 96, l9lnl
Narbo, 105 Philippi, 1, 9!1, 15ln14
Nemesis, 5, l!l!l, l4!ln25 Philodemus, 10, 182n9
Neoplatonism, 12, 68, 86, 170nl!l Philosophers, philosophy, 11, 1!1, !10, 71, 79,
Neopythagoreanism, 12 87, 89, 97, 182n15
Neptune (Poseidon), 64, 91, 169n7, l70n11, Phlius, 11
197nn64 and 68 Phoenicia(n), 155n47
Nicomedia, l, l6!ln29, l7!ln!l5 Phrygia(n), 18, !12, !15, 52, 84, 118, 169n9 and
Novels, religion in, 10, 24, !19, 61, 68, 8!1, passim in notes
l6!ln29, 17!ln!l5, 178n!l0 Pilgrims, 29, 42, 155n48
Numenius, 10, 68, 145n42 Pisidia, 25, 169n9 and passim in notes
Plato, 10, 75, 79, 86, 87, 88, 145n41
Obdos, 1 Pliny the Younger, !l!l, !14, 104
Oea, 72 Plotinus, 68, 79, 175n4!l
Oenomaus, 10, 176n56 Plutarch, !l, 57, 58, 59, 69, 74
Olbia, 7!1 Pluto, 165n4!l, 184n24
Olympia, 26, 28, !11, 7!1, 9!1, 107 Poetovio, 110
Olympus, I, 76 Polemo, !II, 154n40, 178n25, 20!ln40
Oracles, 6, 10, 61, 62, 6!1, 82, 15!ln!l2, 174nn Polyonymy (many-named-ness), 12, 86, 87,
!19-40 88, 90, 91
-responses of, 12, !l!l, 60, 147n65 Pompeii, 7, 6!1, 105, 107, 108, 114, J96n61
See also Amphilochus; Apollo of Claros; Pontus, 21, 2!1, 25, 62, 95, 100
at Didyma; Delphi; Dodona; Limyra; Mop· Populonia, 64
sos; Trophonius Porphyry, !l, 10, 1!1, 68, 86
Orphic hymns, 16, 68, 78, 87 Poseidon. See Neptune
Osiris, 14, 18, 68, 80, 9!1, l94n!l!l, and passim Priests, appearance, 4!1, 64, 1!1!1
in notes -as donors, 47, 66, 99, 162n!l2, 17!ln29,
-death of, 27, 4!1, 76, 78, 172n22 197n64
Ostia, 7, 92, 99, 115, 116, 119, 144n29 -as interpreters, 29
Ousoros, 1 -careers, 104, 106, 109, 1!12
-devout, 64,66-67, 72-7!1, 9!1, 99, 109, 166n7,
Palestine, 60, 178n28 177-78n21
Palmyra, 1!1, !17, 81, 86, 99, 109, 142nn17-18, -duties of, 4!1, 44, 47, 99, l46n50
and passim in notes. See also Aglibol; Aziz; Prometheus, 75, 181n4
larhibol; Malakbel Proserpina (Persephone). See Kore
Panamara, 28, 57, 142nl7. See also Zeus Prostitution, temple, 154n!l8
Panegyriarch, 15!ln!l2, l54n!l7 Ptah, 14
Pannonia, 99, 106, 127, l4!ln18 Punic cults, 41, ll!l, 1!10, l42n17
Pantheus (·a), 187n47 Puteoli, 4, 142n18
Papa, 18, 84 Pythagoras, 96, 122
Parades, religious, 16, 19, 27-28, 45, 47, 149
n78, l54n40 Rhea, 18
Patrai, 21 Rhine, 18, 119, 201n17
Patrodus, 60 Rhodes, 86, 91, 14!ln18
 240 Index

Rhone, 4, 95 Stobi, 30
Rider Gods, 36, 81, 91, 92 Stoicism, 10, 56, 66, 86, 87, I83n19
Rome, 38, 44, 85, 100, 101, 102, 116, 142nl8. Stratonicea, 46, 47, 48, 73, 106, 108, 151nl5
See also Aventine; Capitolium Sun. See Sol
Superstition, 70, 72, 73, 74
Sabazius, I, 54, 14In2 Sur, 19
Sacrifices, animals for, 12, 21, 28, 37, 41, 46, Syncretism, 92, 93, 94, 101, 190n71
129, 147n66, 204n49 Syria(n), 5, 18, 35, 37, 73, 80, 106, 110, and
-human, 76, 182nl4 passim in notes
-rules for, 12, 13, 15 -cults abroad, 19, 37, 54, 114, l43n18
Sahr, 19
Salonae, 13 Tabus, 12,58, 174n36
Salus, 109 Tanit, 41, 85, 160n6
Samothrace, 18, 23, 29, 99, 17Inl9 Tars, 4
Sanxay, 19, 155n50 Tarsus, 4, 81
Sarapis, 6, 31, 44, 49, 93, 100, 143nl8, and Taurobolium, 103, 105, 17lnl5, 172n24, 204
passim in notes n43
-identified with other gods, 51, 83, 84, 91 Taurus, 5
Saturn, 5, 42, 101, 106, 127, 141n2 Telmessos, 4
Scaptopara, 27, 105 Teos, 105, 142nl7
Seeia, 19 Tetrarchs, 85, 89, 118, 127, 132, 188n53, 194
Selene. See Luna n36
Seneca the Elder, 10, 63, 145n42 Teucrus of Cyzicus, 66
Seneca the Younger, 56, 87 Teutates, 143n27
Severus, Alexander, 92 Theaters, cult-, 18-27 passim, 46,153n33, 167
Severus, Septimius, 84, 194n39, 195n40 niB
Sicyon, 35, 52, 160nl -secular, 18-24 passim, 30, 143n27
Sidon, 106 Thebes (Boeotia), ll, 20
Silvanus, 6, 7, 51, 93, Ill, 113, 141n2 Theodosius, 133, 205n3
Sinope, 90 Theologues, II, 17, 146n52, 156n56, 183nl9
Sinuri, 17 Theophilus, 182n15
Slaves, 8, 47, 51, 69, 114, 196nn62-63 Thermopylae, 27
Smyrna, 104, 109, 143n25 Thoeris, 162n23, 177nll
Snakes, sacred, 12, 34, 35, 91, 120, 121, 122 Thrace, I, 105, 159n80
Sol, 4, 7, 58, 81, 84-94 passim, 104, 110 Tiberias, 84
Soldiers, cults acts by, 100, llO-Il, ll3, 114, Tifernum, 104
117, 118, 119,124,143nl8,167nl5, 175n47, Tithe to temples and temple-funds, 32, 41,
197n64, 198n70, 199n7, 200nl5, 20lnl6, 46, 48, 99, 105, 154n34, 164n36, 192nl7
202nn23-24 Tithorea, 26, 32
Sophocles, 15 Tomis, 120, 121
Souchos, 142nl7 Tourists, 11, 30, 73
Souregethes, I Toutain, Jules, 116, ll7
Springs, sacred, 28, 33, 58, 155n50 Treves, 52
State religion, alleged, 102, I 04, 110, Ill, 187 Troezen, 32, 162nl6
n42, 198n70, 20lnl8,202n27 Trophonius, 28
Statues and idols for worship, 31, 43, 47, 59, Tyche (Fortune), 7, 81, 120, 133
76, 156-57n63, 157-58n67, 175n42,182nl3, Typhon, 87
193n26 Tyre, 4
-miracles by, 48, 60, 175n42
-votive, 32, 42, 60, 156-57n63, 157-58n67, Uchi Maius, 109
165n44, 194n36 Ulpian, 5, 197n68
 Index 241

Utica, 110 -as worshipers, 8, 47, 52, 69, 115, 116, 117,
144nS7, 155n52, 161nl4, 170nll
Val di Non, I See also Priests
Valens, C. Domitius, I~
Valentinianism, 79 Xenophanes, 75, 181n5, 182nl0
Varro, 10
Vatican Hiii, 100 Zalmoxis, I
Venetia, 5, 114, 115 Zeus, I, 4, 7, 10, 18, !II, 80, 8~. 84, 87, 91, 92,
Venus, 6, 7, 22, ~6. 6~. 64,92, 165n44 108, 1~3. 142nl7 and passim in notes
Vertumnus, I -Brontaeus, l58n67
Vesta, 7, 6~. 110, 127, 128, 194n~6. and passim -Casius, 4, l84n24
in notes -Galactinos, 9~. l69n9
Vicarello, 28 -Larasios, 154n!18
Vintius, 4 -Osogos, 165n2
Virunum, ~~~. 162n21 -of Panamara, 14, 46, 47, 48, 57, 106
Vulcan, I, 64, 92, 156n~ -Protomysius, 196n55
- Thamaneitanos, 4
Women, 161nl4, 170nl1 - Trosus, 158n70
Zizima, 4
CPSIA infonnation can be obtained at www.ICGtesting.com
Printed in the USA
BVOW08sl738100616

451387BV00003B/5/P
History / Classics

Paganism in the "MacMullen . . . has published several books in

recent years which establish him, rightfully, as
Roman Empire a leading social historian of the Roman Em-
pire. The current volume exhibits many of the
Ramsay MacMullen characteristics of its predecessors: the presen-
tation of novel, revisionists points of view ... ;
discrete set pieces of trenchant argument
which do not necessarily conform to the
boundaries of traditional history; and an im-
pressive, authoritative, and up-to-date docu-
mentation, especially rich in primary sources .
. . . A stimulating and provocative discourse
on Roman paganism as a phenomenon worthy
of synthetic investigation in its own right and
as the fundamental context for the rise of
Christianity."-Richard Brilliant, History
"MacMullen's latest work represents many fea-
tures of paganism in its social context more
vividly and clearly than ever before."-Fergus
Millar, American Historical Review
"The major cults .. . are examined from a so-
cial and cultural perspective and with the aid
of many recently published specialized studies .
. . . Students of the Roman Empire .. . should
read this book."-Robert J. Penella, Classical
World
"A distinguished book with much exact obser-
vation .... An indispensable mine of erudi-
tion on a grand theme."-Henry Chadwick,
Times Literary Supplement

Ramsay MacMullen, Dunham Professor Emeri-

tus of History and Classics at Yale University,
is also the author of Christianizing the Roman
Empire, A.D. 100-400, Corruption and the
Decline of Rome , Roman Social Relations , 50
B.C. to A.D . 284, and Christianity and Paganism
in the Fourth to Eighth Centuries.

I SB N 978- 0 -3 00 - 02 984-0
9000
YALE UNIVERSITY PRESS
11111111
9 78 0 3 00 0 2 9 84 1 1 1 1 1 1 ~1
O> NEW HAVEN AND LONDON
http://www. yale.edu/yup/