This page intentionally left blank

FROM THE TETRARCHS TO

THE THEODOSIANS

An integrated collection of essays examining the politics, social net-

John Matthews
 YALE CLASSICAL STUDIES

VOLUME XXXIV

FROM THE TETRARCHS TO

THE THEODOSIANS
Later Roman History and Culture,
284–450 CE

edited by
SCOT T M C GILL
Rice University, Houston

CRISTIANA SOGNO
Fordham University, New York

EDWARD WAT TS
Indiana University
CAMBRIDGE UNIVERSITY PRESS
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore,
São Paulo, Delhi, Dubai, Tokyo

Cambridge University Press

The Edinburgh Building, Cambridge CB2 8RU, UK

Published in the United States of America by Cambridge University Press, New York

www.cambridge.org
Information on this title: www.cambridge.org/9780521898218
© Cambridge University Press 2010

This publication is in copyright. Subject to statutory exception and to the

provision of relevant collective licensing agreements, no reproduction of any part
may take place without the written permission of Cambridge University Press.
First published in print format 2010

ISBN-13 978-0-511-71312-5 eBook (NetLibrary)

ISBN-13 978-0-521-89821-8 Hardback

Cambridge University Press has no responsibility for the persistence or accuracy

of urls for external or third-party internet websites referred to in this publication,
and does not guarantee that any content on such websites is, or will remain,
accurate or appropriate.
For John Matthews, on the occasion of his 70th birthday
 Contents

Acknowledgments page ix

Introduction 
Scott McGill, Cristiana Sogno, and Edward Watts

part i politics, law, and society

 The unity of the Roman Empire 
David Potter
 Roman patronage 
Peter Garnsey
 Roman matchmaking 
Cristiana Sogno
 Constantine the lawgiver 
Jill Harries
 Constantine answers the veterans 
Serena Connolly

part ii biography and panegyrics

 Three generations of Christian philosophical biography 
Edward Watts
 The education of Paulinus of Pella: learning in the
late empire 
Josiah Osgood

vii
viii Contents

 Another man’s miracles: recasting Aelius Donatus in Phocas’s

Life of Virgil 
Scott McGill
 Gregory of Nazianzus’s Life of Julian revisited (Or.  and ):
the art of governance by invective 
Susanna Elm

part iii faces of theodosius i

 Liar in winter: Themistius and Theodosius 
Peter Heather
 Moments of truth: Gregory of Nazianzus and Theodosius I 
Neil McLynn
 Reinventing Constantinople: Theodosius I’s imprint on the
imperial city 
Brian Croke
 Reinventing history: Jerome’s Chronicle and the writing of
the post-Roman West 
Mark Vessey

References 
Index 
 Acknowledgments

We owe debts of gratitude to a range of people who have made this project
possible. Christina Kraus and the Yale University Department of Classics
have been enthusiastic in their support of our initiatives. Michael Sharp and
Elizabeth Hanlon of Cambridge University Press have been important
resources throughout the process of preparing and assembling this volume.
The Departments of Classics at Rice University and Fordham University
have also made our efforts easier through their financial support. We also
must thank Veronika Grimm who offered good advice even while keeping
our activities a secret.

ix
 Introduction
Scott McGill, Cristiana Sogno, and Edward Watts

One had only to open the pages of Ammianus to see that this was
a source for late Roman history that . . . was a wonderfully effective
introduction to a new age, combining the unexpected features of
this new age with a more or less traditional way of describing them.
After the well-practiced regularities of early imperial history, what was
striking about the later Roman empire was its richness and diversity,
and its massive and varied documentation; and here was a writer
prepared to address it in the familiar terms of the Classical historian.
The work of John Matthews can be described in much the same terms that
he uses to introduce the historian Ammianus Marcellinus in the second
edition of his book The Roman Empire of Ammianus. Across a long and
distinguished career, Matthews has framed late antiquity in classical terms,
but with an eye to bringing out the distinctive contours of the new age.
Like Ammianus, Matthews never pretends that the structures and routines
of the high empire survive unchanged into late antiquity. Yet, again like
his most famous subject, Matthews also recognizes the advantages of using
classical tools to draw upon the great range and relative abundance of
sources available to reconstruct the history of the later empire.
Matthews’s Ammianian approach has made significant contributions to
a fertile period in late antique studies. Early in Matthews’s career in the
s and s, Peter Brown led a push to bring scholarship on the period
beyond the pessimistic Gibbonian paradigm as well as the contemporary
(and more deliberate) historical model of A. H. M. Jones. In their stead
came a series of studies highlighting the period’s cultural and religious
dynamism as well as its continuities with the Roman imperial past. Recent
years have seen a reaction against this trend, with the history of (particularly
western) political and economic disruption again being prioritized. While
 Matthews : ix–x.
 Jones a. The most notable of Brown’s early work tending in this direction is Brown .
 E.g. Heather  and Ward-Perkins .


 scott m c gill, cristiana sogno, and edward watts

these larger discussions have certainly influenced Matthews, his scholarship

connects to them by implication: his career is marked not by expansive
arguments about the character, shape, and limits of late antiquity, but by a
Symian close exploration of particular elements of the later Roman world.
In a field of study where grand military, economic, and religious narratives
often predominate, Matthews stands out as a proponent of close, analytical
history, and as someone who has succeeded in bringing a great number
of individual events, people, and texts into view or into clearer focus. His
work, with its own richness and diversity, as well as its own extensive and
varied documentation, has made late antiquity more intelligible and vital
to new generations, just as Ammianus opened up that age to him. The
Roman empire that emerges neither marches inexorably toward its decline
and fall nor strides confidently along the path of continuity and adaptation.
Its history instead moves more naturally, and therefore more irregularly,
as the inclinations, interests, and interactions of individuals determine the
course of events.
Though the power of Matthews’s approach was already apparent in a
series of articles in the late s and early s, his first book, Western
Aristocracies and the Imperial Court, AD 364–425, fully demonstrated its
promise upon publication in . In its focus upon both the political and
personal characteristics of the later Roman aristocracy, Western Aristocra-
cies described in an original way a world in which later Roman political,
cultural, and religious narratives intertwined with one another across indi-
vidual lives and careers. The book reveals an age in which elites from across
the empire found themselves drawn towards the imperial court. The court
functioned as the center of their ambitions and a facilitator of their rela-
tionships, but ultimately became captive to the private interests of these
aristocrats. Nevertheless, Western Aristocracies serves not as a narrative of
decline, but as a careful profile of a broad and extremely important social
milieu and network. As one reviewer commented at the time, the ability
of Western Aristocracies to outline the enduring concerns of the ruling elite
alongside their changing religious sensibilities creates a picture in which one
part is “as reminiscent of Tacitus as is the other of Augustine of Hippo.”
One finds not a world unchanged, but one in which change progressed at
a human pace and gradually moved along lines that those familiar with the
cultural and political landscapes of the Roman empire could recognize and
understand.

 Wormald : .

 Introduction 

Not only was Western Aristocracies “one of the most influential and
challenging studies of Late Antiquity to appear in the s,” as the many
and significant reviews that it received both in English and in several
European languages show, but it also remains a fundamental study for late
antique scholars and students alike. A clear sign of the sustained appeal of
Western Aristocracies was its reprint in  with only minor corrections and
a substantial Postscript. While addressing the points raised by reviewers of
the first edition, Matthews refused to engage in what “the awful parlance
of modern Universities” would call a “thorough reappraisal.” That would
have called for the circular scrutiny of a vast body of literature that was
ultimately inspired by Western Aristocracies itself.
The same feeling Matthews displays in Western Aristocracies for the
human dynamics that underlie institutions, events, and movements
informs his second monograph, The Roman Empire of Ammianus ().
Inspired in part by Syme’s Tacitus, Matthews’s Ammianus offers a mon-
umental treatment of the history of Ammianus Marcellinus and the world
that gave it shape. The study is divided into two sections, which elucidate
the life and times of the author in different fashions. The first (Res Gestae)
interprets and expands upon Ammianus’s narrative by offering a detailed
reconstruction of the events and personalities featured in the historian’s
text. The historian himself stands as the central figure in this inquiry, as
the focus rests upon events as he experienced them. The second section
of the book (Visa vel Lecta) develops a set of topics “for which Ammianus
is a source and on which his views are of interest.” These thematically
organized sketches provide a comprehensive view of much of the fourth-
century Roman world. They touch upon topics as diverse as the position
and functions of the emperor, the foreign and domestic enemies of the
Roman order, social relationships in the Roman world, and Ammianus’s
religious attitudes. Matthews then concludes with a chapter investigating

 Browning .
 In addition to Wormald’s review cited above, see Fontaine , Giardina –, and Rosen
. The book also received notice beyond the world of academe and was reviewed by John Wilkes
in The Times Literary Supplement,  May .
 With characteristic humor, Matthews remarks that the decision to reprint Western Aristocracies was
prompted by the desire “simply to assist in making available once more a book which . . . has enjoyed
the doubtful accolade of being stolen both from University libraries and from the bookshelves of
colleagues.” Matthews : .
 Matthews : .
 Between Western Aristocracies and Ammianus, Matthews produced Matthews and Cornell . In
the interest of space, however, we treat only Matthews’s monographs in this introduction.
 Syme .  Matthews : x.
 scott m c gill, cristiana sogno, and edward watts

Ammianus’s own conception of his work, which further explores his histo-
riographical method and what we can learn from it and through it.
Ammianus has been praised for its “great originality of both substance
and style,” its richness of detail, and its “immensely entertaining and
stimulating” content; but it has also proven the most controversial of
Matthews’s books. The three areas of greatest controversy seem to be
Ammianus’s place of origin, his religious attitudes, and the degree to
which Ammianus’s rhetorical aims have distorted the “real” Roman world
in his history. In response to the first of these issues, Matthews him-
self mounted a thorough defense of Ammianus’s Syrian origins in a 
article that treats in detail the reasons for identifying Ammianus Mar-
cellinus as the recipient of Libanius’s Ep. . Despite some dissent,
meanwhile, Matthews’s nuanced perspective on the historian’s religious
attitudes continues to be influential. Finally, regarding the possible distor-
tions introduced by Ammianus’s rhetorical aims, Matthews has recently
argued that “factual description and rhetoric are not clearly separable fea-
tures of Ammianus’s writing,” owing to the author’s great skill in abruptly
moving between the two. He thus suggests that the presence of rhetoric
need not preclude a record of facts, and that one can responsibly recover
the latter while acknowledging the former. In whatever way these debates
are eventually resolved, Ammianus remains a monument to Matthews’s
extraordinary erudition and care.
Matthews’s  book Laying Down the Law: A Study of the Theodosian
Code, finds him moving from the writing of Roman history to the writing
of Roman law. The study appeared during a period of increased scholarly
interest in the Theodosian Code, a collection of laws published in  CE
under Theodosius II. Matthews examines not only the content of the
Code, but also the processes whereby the collection was put together and
promulgated. The clarity that Matthews is able to achieve on these matters
is all the more admirable when one considers the density and complexity of
the Code itself. The talent he displays for taking the formidable and making
it accessible – a process that, not irrelevantly, finds him often responding to

 Bowersock b: .  Acknowledged even by Barnes : .

 Drinkwater : .
 Arguments against an Antiochene origin were first raised by Charles Fornara in a  Oxford
lecture and repeated by others in subsequent venues (e.g. Bowersock b, Barnes : ).
 E.g. Barnes : ; Barnes .  This is the overriding argument of Barnes .
 Matthews .  E.g. G. Kelly :  n. .  Matthews : xii.
 A recent article by Sabbah () that offers an overview of both Ammianus’s work and secondary
scholarship endorses Matthews’s views in all three areas of controversy.
 See the studies of Honoré  and Harries .
 Introduction 

Mommsen’s work on the Code – can also be seen as a talent for extracting
significant historical information from what might appear to be inhos-
pitable sources. As Laying Down the Law vividly exemplifies, Matthews
reconstructs ancient history from much more than ancient historiography.
Despite his (good-humored) concern that it will be the “least popular”
of his books (p. xi), Laying Down the Law has become essential reading
in the study of Roman jurisprudence: as one reviewer put it, Matthews
provides “a firm foundation on which to build historical discussions from
the Code.” Scholars interested in the history of legal codification in
Rome, the circulation of laws in the Empire, and, more locally, the sources,
substance, and uses of the Code itself now have Matthews to guide them
through what remains a disciplinary labyrinth. By also rooting the Code so
firmly in its historical context, Matthews makes Laying Down the Law an
invaluable resource for material on late Roman administrative history, the
place of law in late Roman society, and the political and legal landscape of
fifth-century Constantinople.
Matthews’s most recent book, The Journey of Theophanes: Travel, Busi-
ness, and Daily Life in the Roman East, shows him once more finding history
in unexpected places. Theophanes, for which Matthews received the Amer-
ican Historical Association’s Breasted Prize in , uses materials from
a papyrus archive to reconstruct the journey from Hermopolis Magna in
Egypt to Syrian Antioch that the lawyer and public figure Theophanes and
his entourage undertook in the early s. It is a mark of Matthews’s great
scholarly imagination and acuity that, when encountering a set of sources
that largely document travel expenses, he did not equate the quotidian
with the pedestrian and the insignificant, and instead saw the archive as a
valuable source for social and cultural history. Not only does the study
offer insights into what it was like for individuals to live (including in no
small measure what it was like for them to eat) in the late antique world,
but it also connects the Theophanes archive to a broader context, so that
a more panoramic view of the period emerges from its details. Matthews
impressively shows how much historical information is contained in every-
day objects and events. Along with advancing our understanding of ancient
travel, Theophanes joins with his other scholarship in enabling his audi-
ence to continue to discover late antiquity anew.

 Lenski b: . Other reviews include those of Ando  and Humphries .
 Humphries :  makes these points similarly.
 R. J. A. Talbert expresses a similar admiration for Matthews in his insightful review of Theophanes
(Talbert ).
 For a recent work that deals with late antique travel, see Ellis and Kidner .
 scott m c gill, cristiana sogno, and edward watts

This volume marks John Matthews’s scholarly accomplishments, but it

would not exist without his equally impressive achievements as a teacher,
mentor, and colleague. He possesses a remarkable ability to instill in those
around him the same love of intellectual experimentation, attention to
detail, and sensitivity to the contours of individual experiences that typify
his own work. The assertion of distinct historical schools founded by
particular scholars has (perhaps rightly) been discouraged in recent years,
and we do not claim that Matthews has established such a school or that
he ever aimed to do so. Indeed, this very notion would likely embarrass
him. And yet the work of those he has taught and influenced reflects his
belief that an understanding of the later Roman world benefits from the
consideration of diverse and sometimes obscure sources, and his concern
to reconstruct the Roman past from the ground up, through the close
analysis of the primary material. Matthews has made us all appreciate that
the philological tools of the classicist remain indispensable to scholars of
late antiquity.

Inspired by John Matthews’s example, our volume gathers essays that coher-
ently explore topics figuring prominently in his interpretation and recon-
struction of later Roman society. These include politics and elite culture in
late antiquity; late antique historiography; ancient legal theory and juris-
tic texts; late antique authors’ engagements with the classical past; and
the interplay between classical culture and Christianity. The chapters will
largely focus upon the period from the age of the emperor Diocletian (–
) to that of the emperor Theodosius II (–), a period that we call
“the long fourth century.” By this we do not mean to evoke the oft-debated
concept of a “long late antiquity.” Instead, we seek to reveal different
transformations from those that generally interest advocates of that idea.
In the process we follow Matthews in highlighting specific features of a
late classical world, rather than of a postclassical one. For our contrib-
utors, this is a matter of viewing historical events, cultural institutions,
and literary texts against precedents both within the era in question and
across Roman history more broadly, so as to bring out different examples
of social, political, and literary evolution in the period from  to .
 Also deserving mention are Matthews’s abilities as a prose stylist: the clarity, force, and wit with
which he writes call to mind one of his favorite writers, Gibbon. (The influence of Gibbon on the
Matthewsian footnote is in itself a subject worthy of exploration.)
 For a recent example, see Marcone .
 For a notable case where late antiquity is conceptualized as “postclassical,” see Bowersock, Brown,
and Grabar .While Matthews is certainly attuned to what is unique about the period, mean-
while, he is consistent in treating it as the later Roman empire rather than as a postclassical age.
 Introduction 

The volume is organized into three Parts. The first deals with political life
and elite experience in late antiquity. The contributors together convey that
in spite of the rise of Christianity and the emergence of alternative sources
of political and personal authority, many elements of Roman political and
social life remained only slightly changed. Part I begins with David Potter’s
division of Roman history into three distinct phases of self-definition that
shape both its political behaviors and its ability to respond to crises. Potter
attributes the eventual dissolution of the western part of the Roman empire
to the inability of a state increasingly beholden to regional aristocratic office
holders to respond flexibly to the demands of Goths and other outsiders.
And yet, Potter argues, even this later Roman world grew slowly and
organically out of earlier models of imperial self-definition.
Peter Garnsey follows Potter with another far-reaching essay, this time
on the institution of Roman patronage. Garnsey argues that patronage,
long thought to be a vestigial part of Roman social life by the fourth
century, retained its vibrancy as an organizing principle for Roman personal
and political interactions. Like Potter, Garnsey sees institutional evolution
across Roman history; but patronage’s changes proceed more gently and
result in a less dramatic outcome. In a related paper, Cristiana Sogno uses
the correspondence of Symmachus to show how marriage alliances created
and cemented political relationships within the Roman aristocracy much as
they had done since at least the late republic. Sogno shows how the general
model that Garnsey traces actually operates in cementing and securing
particular sorts of later Roman social relationships.
Two other contributions show that the slow and steady evolution of
social and political institutions across the period had the effect of muddy-
ing popular perceptions about the working of the Roman system. In the
first, Jill Harries looks at Constantinian and Theodosian assumptions about
the practical mechanisms for communicating and enforcing law. Harries
goes on to explore the possibility that when political change occurs slowly,
as in the period she examines, administrators may not recognize the degree
to which their social and institutional context diverges from that of their
predecessors. Part I concludes with Serena Connolly’s study of a Constan-
tinian law that shows the emperor working to uphold his traditional duty
to respond to the requests of his subjects while adapting the structure of
his responses to a new ceremonial context. In this, Connolly shows an
emperor responding to the regional concerns that Potter identifies as so
typical of the later imperial system.
Part II analyzes cultural developments in the long fourth century by
focusing upon varied kinds of biographical texts. The aim is to offer fresh
 scott m c gill, cristiana sogno, and edward watts

insights into how authors from the Tetrarchy to Theodosius II worked

with traditional modes of discourse, but adapted them to new material
and new functions in ways that brought about changes to inherited forms.
What emerges, therefore, is a focused picture of late antique innovation, in
which texts evolve in light of particular literary and cultural developments.
Edward Watts begins with a paper describing philosophical biography
as a literature defined by a particular rhetoric that illustrates the practical
application of ideals of conduct. Watts illustrates a broad structural simi-
larity between fourth- and fifth-century pagan and Christian philosophical
biography in which Christian texts like the Life of Antony and Theodoret’s
Life of Simeon Stylites develop the traditional rhetorical strategies of pagan
philosophical biography to define idiosyncratically Christian behaviors in
philosophical terms. Augustine’s Confessions, by contrast, subtly alters this
rhetoric to argue simultaneously against the positive moral effects of classi-
cal philosophy and for the redemptive power of its Christian counterpart.
Each of the remaining chapters of Part II centers on the biographical
work of a single author. In one, Josiah Osgood’s study of Paulinus of Pella’s
Eucharisticon, Augustine continues to figure prominently. Osgood argues
that Paulinus follows the rhetorical lead of the Confessions and uses his
hexameter Eucharisticon to suggest that, in a Christian context, traditional
education stands among other unnecessary worldly trappings. Osgood con-
tends, however, that this rhetoric hides a more nuanced approach to the
classics, in which Paulinus simultaneously disavows their importance and
appropriates their language. Like Osgood, Scott McGill examines a bio-
graphical poem with links to the classical past and the world of education.
In a chapter on the grammarian Phocas’s hexameter biography of Virgil,
McGill investigates how the author deviated from his chief model, Aelius
Donatus’s fourth-century Virgilian Vita, and crafted a new profile of Vir-
gil’s infancy. McGill goes on to explore why Phocas adapted Donatus as he
did, a topic that leads him to consider the biography’s fiction, function, and
possible audience. Finally, to conclude Part II, Susanna Elm investigates the
pragmatics of biographical discourse in Gregory Nazianzen’s two orations
against the emperor Julian. Elm argues that Gregory’s speeches should be
read not as indictments, but as a metaphorical stele that publicly displayed
the emperor’s shameful deeds. While Gregory’s orations attack Julian, they
also suggest a set of ideal behaviors that a Christian emperor ought to
adopt. The aim is to instruct through polemic, rather than through the
positive, prescriptive biographical paradigm explored by Watts.
Part III of the volume works in much the same way as John Matthews’s
most important studies. It explores the broader political, cultural, and
 Introduction 

religious consequences of the challenges presented at one important his-

torical moment, the return of Theodosius I to Constantinople following
his unsuccessful campaign against the Goths in . Peter Heather’s chapter
begins Part III by outlining the precarious political and military position in
which Theodosius found himself during the winter of /. He describes
how, despite his reputation as a resolute Christian champion, the weakened
emperor turned to the Constantinopolitan pagan senator Themistius for
help in rhetorically reinforcing his position. Echoing Potter’s portrait of an
inflexible empire beholden to regional aristocracies, Heather then describes
how the Gothic settlement for which Themistius provided rhetorical cover
created ambiguities and challenges that the Roman administration proved
unable to resolve. The resulting rebellion of Alaric that resulted from this
particular set of circumstances began the process that culminated in the
end of Roman control of the west.
In chapter , Neil McLynn evokes Gregory Nazianzen to show the
measured and ambiguous actions connected to Theodosius’s assertion of
Nicene control over the churches of Constantinople. McLynn’s Theodosius
resembles the careful and politically astute emperor described by Heather
and displays a similar concern for effecting change without angering local
elites. McLynn goes on to show how Theodosius’s deft handling of public
opinion also left him respected by Nicenes, Arians, and Gregory himself.
The political savvy of this emperor is further revealed in Brian Croke’s ensu-
ing chapter. Croke demonstrates how deliberate changes to the physical
and ceremonial space of Constantinople encouraged popular recognition
of a new political order. Croke outlines how, in the s, Theodosius and
his family instituted a calendar of political rituals that reaffirmed both
Constantinople’s status as an imperial capital as well as the important role
that the Theodosian sovereign played in affirming its primacy.
Part III and the volume conclude with Mark Vessey’s discussion of
Jerome’s Chronicle. Vessey locates the Chronicle’s composition at just the
moment when the Latin literary and Roman political contexts began to
make impossible the classicizing, triumphalist historiography of authors
like Ammianus Marcellinus. He argues that the projects of Jerome and
Ammianus differed most significantly in how they sought to shape time
and define the identities and achievements of the age’s central players.
Yet even Jerome’s remarkable redefinition of historiography’s scope and
content shows a deep reliance upon sources and themes drawn from earlier
periods. Together, these contributions show that the first years of the
Theodosian dynasty saw significant innovation in how imperial power
operated, how it was presented, and how it was commemorated. Yet the
 scott m c gill, cristiana sogno, and edward watts

essays also emphasize that change need not entail rupture, and that the
new Theodosian age maintained deep ties with the past’s political practices
and textual forms.

Throughout his distinguished career, John Matthews’s work has shown

particular sensitivity to the unique aspects of later Roman politics, elite
culture, law, and literature, while remaining appreciative of how the period
so often saw the assimilation and modification of existing models, struc-
tures, and practices. To draw out this late antiquity, viewed as a complex
of individuals and of individual episodes, institutions, and texts, has
been his chief enterprise. This he has done by practicing inductive history
and telling specific stories that fill in a world piece by piece. Matthews’s
method complements the grand narrative approach to later Rome and
works toward producing several localized narratives of the period. At the
same time, Matthews’s scholarship is of a scope that includes traditional
oppositions like east and west, imperial center and provincial periphery,
and polytheism and Christianity. The wide-ranging work he has produced
treats historiography, epigraphy, prosopography, poetry, law, and even an
ancient travel ledger. This volume honors Matthews by gathering a range
of papers from his former students and friends that reflect his interest in
accumulating varied information and giving diverse, detailed accounts of
the politics, society, and culture of late antiquity. The essays are also meant
to convey the deep gratitude that their authors feel towards someone who
has offered guidance, support, and compassion throughout their careers
and lives.
mentis agitator meae,
Consilia nullus mente tam pura dedit.
Ausonius, Prof. xv.–

 The language here echoes Scourfield : .

 Matthews’s concern, as he puts it in Matthews : xii, to “integrate the elements of political,
social, and cultural history into a narrative structure” marks his work as a whole, and certainly
remains keenly evident in his latest book on Theophanes.
 part i

Politics, law, and society

 chapter 1

The unity of the Roman Empire

David Potter

While on an embassy to the court of Attila in , Priscus of Panium

reported that his party encountered an embassy to the Huns from the
“western Romans.” These Romans had issues with the Huns that were
completely independent of those that had brought Priscus to the lands
north of the Danube, and it seems that he regarded them as members of
a different state. He referred to his own people as “the eastern Romans,”
and, as various sources preserve his words, writes of Rome as a geographical
area ruled by multiple “kings.” Priscus’s language is scarcely unusual for
this era. In a very different context, Anicius Achilleus Glabrio Faustus,
who received the great law code commissioned by Theodosius so that
he could bring it to the senate at Rome, is described as a man who was
honored in “both courts.” Theodosius himself said that he had conceived
the project of the code as a gift to “his world,” which, by implication,
only included the world ruled by Valentinian III once that emperor had
agreed. A generation earlier, the general Arbogast had refused to accept
that another Valentinian (Valentinian II) could dismiss him from office on
the grounds that he was not responsible for his appointment, while nearly
fifteen years before, the uncle of that same Valentinian had not presumed

I am grateful to my colleagues Arthur Verhoogt and Nicola Terrenato for comments on an earlier
draft of this paper. My debt to the honorand of this volume is hopefully evident throughout.
 Priscus fr. , line  (Blockley): pr”sbeiv to©nun par‡ %et©ou kaª toÓ basileÅontov tän —sper©wn
ëRwma©wn . . . ktl. (see also l. ; the placement of Aetius on equal footing with Valentinian III is
also telling). See also fr.  Àti Þv  gg”lqh t
%ttžl t¼n Markian¼n –v t‡ kat‡ tŸn ›v ëRwma·k‡
parelhluq”nai bas©leia met‡ tŸn Qeodos©ou teleutžn,  gg”lqh d• aÉt
kaª t‡ tv ëOnwr©av
p”ri gegenhm”na, pr¼v m•n t¼n kratoÓnta tän —sper©wn ëRwma©wn ›stelle toÆv dialexom”nouv
mhd•n ëOnwr©an plhmmele±sqai (the fact that the same usage appears in two different sources
suggests that the language is that of Priscus rather than the epitomators). So too in the case of ëRÛmh,
which in fr.  (–pª tv basile©av Qeodos©ou kaª OÉalentinianoÓ AÉgoÅstwn –str†teusan
kat‡ ëRÛmhv kaª Kwnstantinoup»lewv . . . ktl.) designates the geographical area of the empire, a
usage similar to that in fr. : ¨ke g†r tiv ˆgg”lwn t¼n %ttžlan to±v kat‡ tŸn ëRÛmhn –piq”sqai
basile©oiv.
 ILS , –: utriusque inpe|rii iudicii sublima|to.
 For this ceremony, see Matthews : –.


 david potter

to appoint an emperor to replace his own uncle, Valens, as ruler of the

east once Valens had fallen victim to the Goths at Adrianople. Still, that
empire had been one empire with different emperors leading parts of it,
ranked according to seniority. The strict territorial division between east
and west – despite the fact that the practice of listing “legitimate emperors”
in order of seniority still obtained when legal decisions were made – had
only become obvious after the reign of Theodosius II. Olympiodorus, the
great historian of the period before Priscus, had spoken only of Romans in
his history rather than the Romans of east and west, and ended his history
with a triumphant vision of unity as Theodosius II placed Valentinian III
on the throne in .
The formal division of the Roman state between east and west was a
formula for dissolution. What mattered it to the ruler of one Roman state
if the other fell apart? The singular lack of note as to the significance of the
year  at the time is a reflection of this attitude. Indeed, why should the
failure of an Italian regime be especially relevant? The emperor of the east
recognized that there was still an emperor in the western portion, as well as
the new patrician in Italy whose claims were no more extreme than those
of earlier Germanic kings who had carved out realms for their peoples.
The issue that I would like to explore is, rather, how “portions” became
independent states and how these “portions” differed from earlier ways of
describing Roman power.
Methods of describing the power structure of the Roman empire reflect
the value system of the governing class. Empires, in any age, are built in
part through force, in part through the cooperation of the conquered, and
in part by the ability of the governing class to articulate a justification for
its existence that is amenable to those who have come under its control. In
the case of Rome, in all three phases of self-definition that will be outlined
below there is some evidence for people who actively sought to join the
Roman polity as subjects, either as a form of protection from even more
noxious neighbors or because the rulers of Rome were arguably better than
their own. In the third phase, the failure to admit “voluntary” members
of the empire, the lack of imagination that stemmed from the dominant
administrative mentalité provided ready excuse for leaders such as Alaric,
Attila, and Geiseric to do such immense harm to the Roman state that
it could, in the end, no longer retain its integrity. That was not usually
 Zos. .. (Arbogast); for the circumstances surrounding the rise of Theodosius, see Errington
a.
 For Olympiodorus’s understanding of the Roman world, and triumphant vision, see Matthews .
 See, most importantly on this point, Croke .
 The unity of the Roman Empire 

the case under earlier systems of definition, though it is arguable that a

similar failure of imagination caused the crisis that engulfed Italy during
the last years of the second century BCE and set republican government
irrevocably upon the course that culminated in the principate.
The three phases of Roman imperial self-definition may be described in
the broadest terms as being “ethical,” “legal,” and, finally, “administrative.”
The first phase extends from roughly the middle of the fourth century BCE,
when the Roman state achieved the competence and coherence to dominate
the Italian peninsula, to, roughly, the middle of the second century BCE,
when Rome began to come to grips with the need to provide a “modern”
form of administration that was (notionally) based on law. This second
phase elides into the third, “administrative,” phase in the course of the late
third and early fourth centuries CE. In all these cases, it needs also to be
noted that I am talking about the way that the institutional exercise of
power was understood. This is not to deny the importance of the rhetoric
that accompanied the exercise of power but merely to note that, while it
was typical for inhabitants of the Roman empire to think that their place
on earth was assured through divine sanction and for the language of cult
to be incorporated into discussions of mundane power, dominant modes
of describing the way that the empire worked depend upon institutions.
Arriving at the first mode of imperial self-definition is made somewhat
problematic by the nature of the source material. The problem is not
simply that Livy and others often made things up out of whole cloth –
that is the nature of ancient historiography – but rather that the factual
framework over which that rhetorical silk is draped was itself often the
product of highly questionable scholarship. Livy lived in an age when
the engines of scholarly endeavor throbbed at an extraordinary pace as
Romans sought, often for profoundly important contemporary reasons,
to discover the origins of their institutions. The process of the unification
of Italy raised real questions about the status of communities and their
legal relationship to Rome. In the face of this it mattered that no one, it
seems, actually knew the way that municipal status had come into being
during the fourth century BCE. Ignorance promoted vigorous scholarly
invention. Likewise, no one, it seems, really understood the magisterial
structure of the fourth century (e.g., that the officials whom the tradition
calls consuls were called praetors at least until a third was created as a result
of the Lician–Sextian reforms of ), or that the senate did not exist in

 Bispham : –.

 david potter

anything like the form that it would in the late fourth century. Nor can
we now know whether, for instance, the lex Ovinia recognized an existing
state of affairs or whether it created essentially a new institution when
the Roman people granted the censors, rather than the consuls, the power
to select members of the senate. Similarly, scholars would debate such
topics (without self-evidently satisfactory resolution) as the development
of the manipular legion, or the practice of devotio. They could also make
impressive errors such as attributing the grant of a yearly allowance in
denarii to Campanian equites to the period of the Latin wars in the s
(nearly a century before the first denarius was minted). For all of this,
one practice does seem to have descended more or less unchanged from
the fourth century to the first, and to have done so because its persistent
use was a sign of its profound importance in the creation of the empire.
This was the deditio in fidem.
The one reasonably complete text of a deditio in fidem reads as follows:
In the consulship of Gaius Marius and Gaius Flavius, the people of the Saenoci
handed over themselves and all their worldly goods to Lucius Caesius, son of
Lucius, the imperator. Lucius Caesius, son of Lucius, the imperator, after he had
received them in his power consulted his consilium about what he should instruct
them to do. In accord with the advice of the consilium, he ordered that they hand
over the arms, hostages, deserters (?), captives, stallions, and mares that they had
taken. They handed all of these over. Then Lucius Caesius the son of Gaius ordered
that they should be free and restored to them the fields and buildings, laws and all
other things that had been theirs on the day before they handed themselves over
and still existed so long as the Roman senate and people agreed, and, concerning
this matter he ordered that they send ambassadors: Crenus the son of [ . . . ? . . . ],
Arco the son of Catonus were the legates.

 On the lex Ovinia, see Cornell ; Oakley : – (it may be significant that Livy does not
mention this law directly but records the lectio senatus carried out by Appius in  that presumes
passage of the law at ..).
 On both points see the excellent discussions in Oakley : –, –.
 Livy ..; the source is a bronze tablet. For the issues see Oakley : –, who may be too
generous to Livy on this occasion. Likewise the immediately preceding notice in Livy about the
treaty with the Laurentes might preserve a variant tradition from that alluded to on ILS ,
–, honoring Spurius Turranius, the pater patratus populi Laurentis foederis | ex libris Sibullinis
percutiendi cum p.R.
 Hölkeskamp , esp. pp. –, arguing for a link with fetial procedure, but see, on the contrary,
Watson : –. For the role of diplomacy during the fourth-century period of expansion, see
Auliard : –, –. For the importance of fides in personal relationships, see Garnsey, in
this volume.
 AE  no. : C. Mario vac. C. Flavio vac. [cos.] | L. Caesio C. f. imperatore populus Saeno[corum
se suaque] | dedit. L. Caesius C. f. imperator postquam [in dicionem] | accepit ad consilium retolit
quid est im[perandum | censerent. De consili sententia inperav[it arma obsides transfugas?] | captivos,
equos, equas quas cepisent [dederent. Haec | omnia dederunt. Deinde eos L. Caesius C. [f. imperator
 The unity of the Roman Empire 

Implicit in this document is the assumption, on the part of the peo-

ple surrendering, that, once Caesius consulted with his concilium, he
would reach a specific conclusion – that they could have their city back.
Indeed the deditio had the rather curious effect of imposing an obligation
upon the victorious power to behave decently toward the weaker. The
notion that fides was built upon respect of reciprocal power relationships
between victor and vanquished is linked with both public and private
relationships that were of very long standing. This is not to say that the
Romans were incapable, on occasion, of being perfectly bestial to people
who came into their power (just because something is supposed to work
in a certain way does not mean that it will always work in that way), but
it is to suggest that appalling conduct (at least as the Romans might define
it) was the exception rather than the rule.
At first sight the assertion that a document of the late second century
BCE offers a path back to the creation of the early empire in Italy might
appear as dubious as any: is it not the case that a document such as this
might be the result of exactly the sort of antiquarian experimentation that
we have already seen to have been the case? Worry might stem as well
from the fact that the practice was profoundly interesting to leaders of the
Roman state in the first century BCE. Thus the fact that Virgil alludes
to the requirement that conquered peoples be spared – as indeed does
Augustus in his Res Gestae – can be deeply disturbing. So, too, can the
overtly self-justificatory and defensive assertion of interest in the principle
manifested by earlier imperialists in the middle of the first century BCE
be cause for concern. Caesar, after all, made the claim that Rome had
a direct interest in the affairs of Gaul because Quintus Fabius Maximus
had defeated the Arverni and Ruteni in  and, even though the Roman
people had forgiven them, “if it is right to look to the most distant past, the
imperium of the Roman people was most rightfully in Gaul, and if it was
right to look to the judgment of the senate, Gaul ought to be free because
the senate had wished it, once conquered in war, to be under its own laws.”
His statement here is a description of the result of a deditio. In the very
next year he said that the Remi asked that “they and all their property be

liberos | esse iussit agros et aedificia leges cete[raque omnia | quae sua fuissent pridie quam se dedid[issent
quaeque | extarent eis redidit, dum populus [senatusque | Roomanus [sic] vellet; deque ea re eos [mitti
et Romam]| eire iussit vac. legatos Cren[um (?) . . . f. | Arco Cantoni f. vac. Legates [sic].
 Virgil, Aen. .; RGDA .; see also Hor. Carm. Saec. –.
 BG ..–: Bello superatos esse Arvernos et Rutenos a Q. Fabio Maximo, quibus populus Romanus
ignovisset neque in provinciam redegisset neque stipendium posuisset. Quod si antiquissimum quodque
tempus spectari oporteret, populi Romani iustissimum esse in Gallia imperium; si iudicium senatus
observari oporteret, liberam debere esse Galliam, quam bello victam suis legibus uti voluisset.
 david potter

allowed into the faith and power of the Roman people.” While one may
wonder from whom they learned the precise language of a deditio, Caesar
seems clearly aware of the long tradition of states summoning Rome to
their aid by making precisely such a request. A few decades earlier, the
army of Lucullus had been outraged when the general accepted deditiones
from cities in Bithynia since it meant that they would not be allowed to
sack them. Lucullus seems thus to have been aware of the condemnation
of Marcus Popillius Laenas for his slaughter of the Ligurians after their
deditio in . Cicero also commented upon the importance of showing
fides to those who surrendered.
Discussions of this sort reflect antiquarian views on the nature of deditio,
but they may also have been directly relevant to the status of some com-
munities into the later part of the second century as the terms of a deditio
would have defined the terms of that city’s relationship to Rome. The fact
that the concept was alive and well in late republican Rome is therefore not,
in and of itself, sufficient to condemn all Livian (and other) reconstruction
of mid-republican negotiations in which the notion of the deditio and fides
dominate. This is even more so the case because there is independent evi-
dence for the importance of the concept in the mid-republic, and because
very similar statements about the importance of the deditio appear in the
work of two Greek authors, both considerably closer in time to the late
fourth century BCE than was Livy. One is Polybius, whose account of
the Mamertine appeal to Rome during  BC was presumably extracted
from the work of Fabius Pictor. In this case Polybius writes that when
the Mamertines made their deditio to Rome as a way of bringing Roman
power into Sicily, the senate debated the point without reaching resolution,
allowing the consul, Appius Claudius, to put the matter before the people.
The point at issue seems to have been whether or not a deditio should be
accepted from a people as fundamentally unattractive as the Mamertines.

 BG ..: Eo cum de improviso celeriusque omnium opinione venisset, Remi, qui proximi Galliae ex
Belgis sunt, ad eum legatos Iccium et Andebrogium, primos civitatis, miserunt, qui dicerent se suaque
omnia in fidem atque potestatem populi Romani permittere.
See also BG ..: omnes maiores natu ex oppido egressi manus ad Caesarem tendere et voce significare
coeperunt sese in eius fidem ac potestatem venire neque contra populum Romanum armis contendere, with
the discussion in Hölkeskamp : , on the gestures that went with a deditio (also seemingly
familiar to the Bellovaci, if Caesar is to be believed).
 Plut. Lucull. . with Auliard : –.
 For Laenas, see Liv. .; for outrage, see also with Hölkeskamp : –; .
 Cic. Off. ..  On this point, see esp. Hölkeskamp .
 Pol. ..–: o¬ d• pr¼v ëRwma©ouv –pr”sbeuon, paradid»ntev tŸn p»lin kaª de»menoi bohqžsein
sf©sin aÉto±v ¾mofÅloiv Ëp†rcousin. ëRwma±oi d• polÆn m•n cr»non  p»rhsan di‡ t¼ doke±n
–x»fqalmon e²nai tŸn ˆlog©an tv bohqe©av.
 The unity of the Roman Empire 

On their side, the Mamertines seem to have been fully aware of the fact
that the acceptance of a deditio placed Rome under a moral obligation.
When they did accept the Mamertine invitation, we are told that Hiero of
Syracuse pointed out that the action made a mockery of the Roman claims
to be devoted to the concept of fides:
Hiero replied that the Mamertines, who had laid waste Camarina and Gela and
had seized Messana in so impious a manner, were besieged with just cause, and
that the Romans, harping as they did on the word fides, certainly ought not to
protect assassins who had shown the greatest contempt for good faith; but if, on
behalf of men so utterly godless, they should enter upon a war of such magnitude,
it would be clear to all mankind that they were using pity for the imperiled as a
cloak for their own advantage, and that in reality they coveted Sicily.
The importance of Hieron’s observation is that it seems to have figured in
the mid-third-century BCE history of Philinus of Agrigentum and is thus
as early a description of Roman diplomacy as we have, confirming that the
picture of Roman diplomacy based upon fides that Livy offers is not simply
a creation of the first century BCE.
Fides was one of the four virtues celebrated by the Romans in a group
of temples to divinities concocted out of self-congratulatory abstractions
upon the instruction of the Sibylline oracles in the first half of the third
century (the others are Concordia, Victoria, and Spes). It was also a
principle underlying crucial legal and quasi-legal institutions within the
Roman state as structures emerged in the course of the sixth and fifth
centuries BCE to stabilize and expand the collection of aristocratic gentes
that had existed, seemingly, since the city began to take shape. Indeed,
just as the fifth-century law code of the Twelve Tables institutionalizes
horizontal practices of maintaining property within gentes, so too does it
institutionalize the vertical structures within a gens that are based upon the
relationship between patroni and clientes even as it also creates a framework
for a state to operate independently of those relationships. Thus, while
the first and second tables establish the framework for public trials and
distinguish the categories of assiduus and proletarius within the overall
 Diod. .: ¾ d• ëI”rwn ˆpekr©nato di»ti Mamert±noi Kam†rinan kaª G”lan ˆnast†touv
pepoihk»tev, Messžnhn d• ˆseb”stata kateilhf»tev, dika©wv poliorkoÓntai, ëRwma±oi d”,
qrulloÓntev t¼ tv p©stewv Ànoma, panteläv oÉk ½fe©lousi toÆv miaif»nouv. For discussion
of the source of Diodorus here, see, in general, the discussion in P. Goukowsky’s recent Budé
edition, though some may feel that the case in favor of Silenus rather than Philinus of Agrigentum
as the direct source had long since been put to rest by Walbank  (it should be noted that the
volume reviewed here will still inform those who do not believe that Silenus was Diodorus’s source
on many other points).
 Orlin : .  See, in general, Smith : –, –.
 david potter

body of civites, the tenth table orders that public trials be held before a
full meeting of the people; the eighth also states that if a patronus should
defraud a cliens, he shall be sacer, a stark contrast to the class relationship
envisaged in the sixth, which appears to refer to nexum (debt bondage).
Implicit in the statement of the eighth table is a relationship of fides between
patronus and cliens that placed a significant burden on the patronus. So,
too, the deditio was not simply a surrender; it was the establishment of a
bond that placed significant responsibility upon the Roman state.
The deditio is not the only device in the toolbox of mid-republican
expansion that reflected relationships within Roman society. So, too, did
the procedures linked with the declaration and ending of war. The under-
lying principle of the fetial procedure employed to declare war was that
a foreign people must have done something to the Romans that required
redress. The procedure itself, with its protolegal language summoning
the gods to witness that the Romans were wronged and its assertion that
the Romans had tried to avoid conflict by having the offending state make
amends for what had happened, conveyed the notion that Rome would
not act without divine sanction. Similarly, the concept of impietas and
discourse, even at an early date (if it is right to think that Macrobius’s
quotation of discussion of the role of impietas in determining the outcome
of the battle of Allia is from a fourth-century source), on the catastrophic
consequence of irritating the divine would seem to have offered power-
ful support to a diplomatic tradition based upon the assertion of divine
approval. It is perhaps of broader historical interest to note that the fetial
procedure implied de facto that Rome could not develop a doctrine of
preemptive warfare and succeed in becoming a great power.
Although treaties in all Mediterranean societies, virtually by definition,
were sworn with gods as witnesses, the Roman notion of the foedus extends
to some degree beyond this in its assertion of moral probity on the part
of the participants. Other peoples did not stress that participants should
act without bad faith (sine dolo malo), a phrase that is a telltale indication
of Roman draftsmanship, or use a term derived from the concept of good
faith (foedus derives from the archaic form of fides), as opposed, e.g., to
terms indicative of friendship – fil©a – or joining together (e.g., sunqžkh).
The ethical significance of the foedus is perhaps most strongly stated by

 Crawford : no. . .: adsiduo vindex adsiduus esto. Proletario ?civi? quis volet vindex esto; .:
de capite civis, <ni> maximus comitatus <est>, ne ferunto; .: si patronus clienti fraudem fecerit,
sacer esto; .: cum ?faciet nexum? mancipiumque, uti lingua nuncupassit, ita ius esto.
 Garnsey, in this volume.  Watson : .  Macrob. Sat. ..–.
 For the religious aspect of the foedus, see still Mommsen , vol. : –.
 The unity of the Roman Empire 

Cicero, in his observation that the foedus that Romulus struck with Titus
Tatius established the integration of former enemies into the Roman polity
as a basic way that Romans did business. In this period as well, the oaths
that would sanctify a treaty would ideally be sworn by a fetial, who would
sacrifice a pig as the oaths were sworn, a scene illustrated, for instance,
on a stater issued between  and , depicting a Janiform version of
the Dioscuri on the obverse and two soldiers standing on either side of a
fetial who holds the soon-to-be sacrificed pig between them. The meaning
of the scene stresses the centrality of the alliance system to the Roman
state and the importance of the notion that the war with Hannibal was
a cooperative effort between Rome and the Italian allies. Hannibal had
recognized that this was the key to Rome’s strength (as is clear from the text
of the treaty with Philip V), but he could not undo centuries of mutual
obligation through mass slaughter.
Well before Caesar exploited the language of archaic Roman diplomacy
to further his designs in Gaul, and even as the threat of Hannibal was
fading from view, the Roman empire ceased to be based upon the ethical
principle of fides and became rooted in the legal definition of imperium.
The change was gradual and, perhaps, cannot be seen to be fully complete
until the generation prior to that of Marius and Sulla, for it is only in the
second half of the second century that definitions of imperium are firmly
tied to provincial boundaries and that imperium assumes its dual meaning
of magisterial power and the total geographical region wherein the power
of Roman magistrates was exercised. The usage seems to have developed
only gradually in the course of the second century BCE – the earliest extant
usage of imperium to designate a “kingdom” appears to be a fragment of
Accius – and to have been preceded by more than a century by the change
in standard usage of provincia from “task” to the area where the “task” was
carried out. By the end of the second century, we also have firm evidence

 Cic. Balb. : Illud vero sine ulla dubitatione maxime nostrum fundavit imperium et populi Romani
nomen auxit, quod princeps ille creator huius urbis, Romulus, foedere Sabino docuit etiam hostibus
recipiendis augeri hanc civitatem oportere; Rep. .: qua ex causa cum bellum Romanis Sabini intulissent,
proeliique certamen varium atque anceps fuisset, cum T. Tatio rege Sabinorum foedus icit, matronis
ipsis quae raptae erant orantibus; quo foedere et Sabinos in civitatem adscivit sacris conmunicatis, et
regnum suum cum illorum rege sociavit – with Auliard : .
 Crawford : no. /. See also ILLRP , the inscription honoring Duilius, who stresses Rome’s
defense of allies and the role of the allies in the defeat of Carthage.
 Pol. .. The full implications of this document cannot be discussed here, but see the summary in
Potter : –.
 Richardson  remains an essential guide to the subject.
 On this point, see especially the excellent discussion in Bertrand . For the significance of Accius
– (Ribbeck), see Richardson :.
 david potter

that the senate was taking steps to limit the ability of a governor to exercise
authority while in transit to or from his province or to move an army
outside the fines without permission. The most important evidence in this
case stems from the language of a lex Porcia that might date to either 
or  BCE. A section of that law, quoted in the Law on the Praetorian
Provinces of  BCE, states that
No-one, in contravention of those measures that are in the statute which
M. Porcius Cato passed three days before the Feralia, is knowingly, with wrongful
deceit, to draw up (an army) or march or travel outside his province, for whatever
reason, or whenever he shall arrive, nor is any magistrate or pro-magistrate to travel
or proceed outside the province in command of which it shall be appropriate for
him to be in accord with this statute, except according to the decree of the senate,
except for reasons of travel or for reasons of state, and he is, without wrongful
deceit, to restrain his staff.
The lex Porcia, whatever its date, appears to have been relatively close in
time to the composition of Artemidorus’s description of Spain. Even if the
papyrus that purports to present a section of that work is a nineteenth-
century forgery, the surviving fragment, upon which the forgery was based,
is concerned with the boundaries of the Roman provinces in Spain and is
the earliest direct evidence for the importance of fines in defining magisterial
authority.
The link between imperium and space that was established in the course
of the second century BCE provided one of the most important sup-
ports underpinning the Augustan regime. Documents of the new regime
reveal explicitly that the technical authority required to command the
armies – the imperium maius – was defined in terms of the provinces.
 Crawford : no. , Cnidos Copy Col. iii, –: mžte tiv toÅtoiv to±v pr†gmasin Ëpenan|t©wv
to±v —n täi n»mwi Án M†arkov P»rkiov | K†twn strathg¼v –kÅrwse pr¼ ¡m”rwn gì t|än
Fhral©wn –kt¼v tv –parce©av –ktas|s”tw mžte ˆg”tw tiv mžte poreu”s|qw tiv diì  –k†s[tote]
–p†xei e«dÜv d»lwi |ponhräi mžte tiv Šrcwn mžtì ˆnt†rcwn | –kt¼v tv –parce©av, <–f’> ¨v
aÉt¼n –parce©av ka|t‡ toÓton t¼n n»mon e²nai de± £ dežsei, | e« mŸ ˆp¼ sugklžtou gnÛmhv,
poreu”s|qw mžte proag”tw, e« mŸ diapore©av ™ne|ken £ dhmos©wn c†rin pragm†twn, toÅv
te —|autoÓ kwlu”tw {e«dÜv}Šneu d»lou ponhroÓ. For a different view, arguing that the notion
that provinces had fixed fines was not common until the middle of the first century see Richardson
: .
 Stephanus. s.v. ìIbhr©ai: ˆp» d• tän Purhna©wn ½rän ™wv tän kat‡ G†deira t»pwn kaª tän
–ntot”rwi klim†twn ¡ sÅmpasa cÛra sunwnÅmwv ìIbhr©a kaª ëIspan©a kale±tai die©rhtai dì
Ëp¼ ëRwma©wn e«v dÅo –parce©av kaª tv m•n prÛthv –stªn –parce©av ¡ diate©nousa ˆp¼ tän
Purhna©wn ¾rän pasa m”cri tv Kainv Karchd»nov kaª KastÅlwnov kaª tän toÓ Ba©tov
phgän tv d• deut”rav –stªn –parce©av t‡ m”cri Gade©rwn kaª t‡ kat‡ tŸn Luseitanian. The
text first published in Gallazi and Kramer  has been attacked as a forgery in Canfora ;
Richard Janko has recently discovered evidence that would support Canfora’s case, but Ludwig
Koenen, who examined the papyrus when it first appeared on the market in the early s, tells
me that he saw sand in the ink, which would suggest that the ink was indeed ancient.
 The unity of the Roman Empire 

Thus, in his funeral oration for Marcus Agrippa, Augustus said of his
former colleague and son-in-law that “it was ordained by a law, that no
one should have greater imperium than you in any of those provinces
into which the common affairs of the Roman people should call you.”
The precise meaning of this phrase, which can be traced, verbatim, to
the language of a motion connected with the passage of a law grant-
ing Gnaeus Pompey control of the grain supply of Rome in  BCE, is
confirmed by its appearance in the decree of the senate settling the case
of Gnaeus Calpurnius Piso. The senate stated that Piso’s crimes arose
when
he should have remembered that he had been given as a support for Germanicus
Caesar (who had been sent by our Princeps in accordance with the authority of
this order to settle overseas affairs that required the presence of either Ti. Caesar
Augustus himself or of one or the other of his two sons), ignoring the majesty of
the imperial house, and also ignoring the law of the land – having been attached
as a proconsul to a proconsul, about whom a law had been carried in a public
assembly providing that in whatever province he entered he had greater imperium
than the province’s proconsular governor, with the proviso that in every case Ti.
Caesar had greater imperium than Germanicus Caesar.

Although Tacitus might note the passage of measures to confirm

imperium on new emperors only in passing (and inaccurate) fashion in
the course of his narrative of AD , and although he might occlude the
significance of the formal meeting of the senate after Augustus’s funeral
when he did not stress the passage of a relatio conferring powers upon
Tiberius that he lacked and were part of the statio of his father, republican
precedent clearly mattered. Tacitus notes, for instance, that the command

 P. Koln. : –: kaª e«v{v}Œv džpo|t” se Ëparce©av t‡ koin‡ tän ìRw|ma©wn ™f”lkoito, mhqen¼v
™n –|ke©naiv –xous©an <e²nai> m. e. ©.zw. tv sv ™n. | n»mwi –kurÛqh.
 SCP –: cum deberet meminisse adiutorem se datum | esse Germanico Caesari, qui a principe nostro ex
auctoritate huius ordinis ad | rerum transmarinarum statum componendum missus esset desiderantium |
praesentiam aut ipsius Ti. Caesaris Aug(usti) aut filiorum alterius utrius, neclecta | maiestate domus
Aug(ustae), neclecto etiam iure publico, quod adle<c>t(us) pro co(n)s(ule) et ei pro co(n)s(ule), de quo |
lex ad populum lata esset, ut in quamcumq(ue) provinciam venisset, maius ei imperium | quam ei qui
eam provinciam proco(n)s(ule) optineret, esset, dum in omni re maius imperi|um Ti. Caesari Aug(usto)
quam Germanico Caesari esset. For the late Republican precedent, see Cic. Att. ..: alteram Messius,
qui omnis pecuniae dat potestatem et adiungit classem et exercitum et maius imperium in provinciis
quam sit eorum qui eas obtineant. Messius’s bill did not pass, but the fact that he used this formula
suggests that it was available to legislators of the s more generally.
 See Tac. Hist. ..; .. (shown to be imprecise by the acta Arvalia for that year, on which
see now Scheid : no.  and Scheid . On the matter of the relatio passed in AD  and
mentioned in Tac. Ann. ., I am indebted to discussion with J. F. Matthews. For the history of
laws conferring power on emperors otherwise see Brunt .
 david potter

of Corbulo in the east was couched in terms redolent of the lex Gabinia,
and he had himself composed the remarkable discussion asserting that the
generation of Augustus was but one removed from his own. The contin-
ued relevance of republican precedent is reflected not only in the reading
list of Aulus Gellius, concentrating as it does upon works composed prior
to the time of Augustus, but also in the scholarship of Festus and other
intellectuals of the imperial period. The same cultural bias is reflected
in legal thinking when, for instance, Tacitus’s younger contemporary, the
jurist Pomponius, wrote that
More recently it has happened that there are fewer paths for establishing law; this
seems to have happened through stages under the force of circumstance, so that it
is necessary for the state to be governed by one man (for the senate is not able to
govern all the provinces properly). A princeps therefore was appointed, to whom
was given the right that what he had decided be considered law. (D. ..)

Nearly a century after Pomponius wrote these words, Ulpian wrote, in

what was his version of a basic textbook on Roman law, that
Whatever the emperor decides has the force of law since the populus confers upon
him all of its imperium and potestas through a lex regia, which is passed concerning
his imperium. (D. ..)

What is significant in both cases is that authority within the empire

is envisioned as being based on law that conferred imperium, and even
if Ulpian’s lex regia is a fiction, the point remains that imperial author-
ity was extrinsic to the person of the emperor who is seen as working
within the confines of the larger state. Moreover, not only was the exis-
tence of the empire independent of the existence of the emperor; so, too,
by this period, seem to have been some properties that belonged to the
patrimonium.
The fact that empire (and imperial property) had an independent exis-
tence meant that the concept of imperium that the Augustan regime inher-
ited from the republic could be variably defined as necessary. Just as the
geographical imperium had partes – Velleius, for instance, points out that
“while everything was going well in this part of the imperium” disaster
struck in another – so, too, could the imperium maius be limited by region

 Tac. Ann. ..: scribitur tetrarchis ac regibus praefectisque et procuratoribis et qui praetorum finitimas
provincias regebant, iussis Corbulonis obsequi, in tantum ferme modum aucta potestate, quem populus
Romanus Cn. Pompeio bellum piraticum gesturo dederat; Dial.  passim.
 D. ..– with Millar : .
 The unity of the Roman Empire 

(as was plainly the case with the grant to Germanicus). The partibil-
ity of imperium maius had important consequences for the unity of the
empire, for it enabled the ad hoc creation of large regional commands.
Consequently, from the first to third centuries, the practice of appointing
deputies with commands greater than a single province is well attested.
Leaving aside men who governed more than one province at a time (e.g.,
Pertinax’s governorship of both Moesian provinces in  and the gover-
norship of both those provinces and Daciae tres in ), but including men
such as Aulus Plautius, Vespasian, and Petilius Cerialis whose command
for a specific bellum involved the ability to give orders to other provincial
governors, there are no fewer than nine such men between the death of
Augustus and that of Antoninus Pius.
Ideology is in these cases the expression of the corporate interest of
the governing class. The frequency of the appointments of what were in
effect “deputy emperors” throughout the first two and a half centuries CE
without infringing on the essential unity of the imperium highlights the
significance of the change in expression after Diocletian’s abdication in
, and it raises the question of why Marcus’s power-sharing arrangement
with Lucius Verus proved so very different from that of Diocletian and
Maximinus.
The transformation was not the result of sudden change, but just as the
alteration from “ethical” to “legal” conceptions of empire had occurred
in response to gradual circumstance, so, too, did the change from “legal”
to “administrative” definition. The rise of the “administrative” definition
of imperial power was the result of the integration of local ruling classes
into the lower echelons of the imperial governing class. Early signs of
the process appear in such texts as the foundation of Salutaris at Ephesus
or of Demosthenes at Oenoanda, where former imperial officials who
either returned to home cities or set up residence at provincial centers
invoked imperial protection for self-commemorative foundations. That
such protections might actually work is confirmed by a letter written
by Hadrian in response to an embassy from the International Theatrical
Association of the Artisans of Dionysus, Sacred and Crowned Victors
stating that “I order that all the games shall be held and that a city is
not to divert the funds established for games according to law, decree, or

 Vell. Pat. ..: sed dum in hac parte imperii omnia geruntur prosperrime; see also Tac. Agr.  with
Richardson : .
 For a list, see Potter ; the omission of Petilius Cerealis’s command in the war to suppress the
Batavian revolt from the list of those with imperium maius was a mistake.
 Dmitriev : .
 david potter

testamental foundation for some other expenditure nor do I permit some

building project to be constructed with the money that ought to pay for
prizes for the contestants or allowances to victors.”
Effective though this intervention may have been, it does not mean
that agents of the state were in direct control of cities (it simply means
that people who knew how to appeal to the personal taste of an emperor
could get their way). Likewise, the appointment of curators, sometimes
of local origin, to oversee civic finances would not inevitably result in the
co-optation of local governing classes into the bureaucracy. Indeed, as late
as the Severan period, Ulpian quoted Septimius as having written, on the
subject of gifts that a governor was allowed to take “as far as it concerns
gifts, this is what we think: there is an old proverb, ‘not everything, nor
everyday or from everyone.’ For it is inhuman to accept nothing, but to
accept all things indiscriminately is vile and greedy.”
The point here is that while the governor is still seen as an individual
who is distinct from local society, he had to be able to deal on civil terms
with leading members of that society. In doing so, he was still something
of an outsider, and the rule against appointing a man to be an official in his
own province annunciated by Marcus Aurelius after the revolt of Avidius
Cassius, just like the rule imposed after the troubles on the Rhine in
– CE that auxiliaries should not serve in the district from which they
were recruited, shows that emperors genuinely recognized the dangers of
local particularism to the empire as a whole. It is not until the post-Severan
period that there began to be indications that the imperial government was
losing control of the lower echelons of its own government and that offices
within the administration could take on a life of their own. Indeed, while
ties of patronage declined in the late republic, they were strengthened in
the course of the imperial period as entry to the lower rungs of imperial
administration that depended on the patronage of those who had already

 Petzl and Schwertheim : lines –: ToÆv ˆgänav p†ntav Šgesqai keleÅw kaª mŸ –xe±nai
p»lei p»rouv ˆgänov kat‡ n»mon £ kat‡ yžfis|ma £ diaqžkav ˆgom”nou[v] metenenke±n e«v
Šlla dapanžmata oÉd• e«v ›rgou kataskeuŸn –f©hmi | cržsasqai ˆrgur©, –x oÔ qla t©qetai
ˆgwnista±v £ sunt†xeiv d©dontai to±v ne[i]kžsasin. For further discussion of the text see now
AE  no. .
 D. ...: non vero in totum xeniis abstinere debebit proconsul, sed modum adicere, ut neque morose in
totum abstineat neque avare modum xeniorum excedat. quam rem divus Severus et imperator Antoninus
elegantissime epistula sunt moderati, cuius epistulae verba haec sunt: “quantum ad xenia pertinet,
audi quid sentimus; vetus proverbium est ‘oÎte pŽnta oÎte p†ntote oÎte par‡ p†ntwn.’ nam
valde inhumanum est a nemine accipere, sed passim vilissimum est at omnia avarissimum.” Contrast
D. ..: Lex Iulia repetundarum pertinet ad eas pecunias, quas quis in magistratu potestate curatione
legatione vel quo alio officio munere ministeriove publico cepit, vel cum ex cohorte cuius eorum est.
Excipit lex, a quibus licet accipere: a sobrinis proprioreve gradu cognatis, uxore (Macer).
 The unity of the Roman Empire 

begun the ascent. Lacking much in the way of an actual government,

there had been little in the way of offices that the grand aristocrats of the late
republic could offer supporters, whose loyalties were notably fickle. As the
principate wore on, the number of prizes to be won increased and control
of access to those prizes became more significant. Increasing size meant,
as well, that emperors could not be expected to know much about those
who were appointed primus pilus in their army or as minor procurators
in the domains. It is likely that efforts to expand imperial control of local
finances in the middle of the third century foreshadowed the development
of late imperial patterns of government, while the internal wars between the
capture of Valerian and the accession of Diocletian resulted in the creation
of imperial bureaucracies that could survive regime changes in different
parts of the empire.
In the matter of imperial finance, it can be detected on papyri that dur-
ing the reign of Philip, as Peter Parsons noted some forty years ago, a pair
of officials named Marcellus and Salutaris engaged in major overhauls of
the landholding, taxation, and liturgical systems of the province and, per-
haps more significantly, that Marcellus held the new post of rationalis.
In another reform that briefly presaged developments later in the cen-
tury, Philip created two supraregional commands: one for a kinsman
named Severianus in the Balkans; the other for his brother, Priscus, in
the east. Twelve years later, of course, the empire split into three sections
and remained divided for more than a decade, although, in the east, the
reigning emperor at the center, Gallienus, retained the authority to appoint
provincial governors while granting some sort of general authority over the
region to Odaenathus of Palmyra. When Gallienus’s successor, Claudius,
appears to have failed to recognize that Palmyrene interest in continuing the
arrangement after Odaenathus’s death, resulting in a Palmyrene invasion
of Egypt, we suddenly get important evidence for the way that admin-
istrative regimes had been evolving. Unlike the civil wars of the Severan
period, in which cities and their leaders paid heavy penalties for choosing
the wrong side (or even guessing wrong about the direction of politics in
the court), civil wars of the later third century appear to have resulted in
few changes below the uppermost echelons of government. This pattern
seems to have continued into the fourth century. Aside from the reign of
Constantius II, whose agent, Paul, “the Chain,” seems to have controlled a
genuinely effective repressive apparatus, symbolic execution of a few senior

 On this point, see Brunt : –, and Garnsey, in this volume.  Parsons .
 david potter

(and especially loathed) officials seems to have been more common than
widespread reprisals in the wake of a regime change.
Among the important pieces of evidence for the absence of large-scale
eliminations of serving staff is a recently published Michigan papyrus
containing an appeal from a man named Aurelius Thosius for the transfer
of some property, supported by the prefect of Egypt in a process that
appears to correspond exactly to the Palmyrene invasion, suggesting that
the change of regime had no impact on administrative practice (including
the composition of a part of this document in Latin). So, too, at the other
end of this brief interlude, Statilius Ammianus, who served as prefect of
Egypt both under the Palmyrenes and under Aurelian. Statilius succeeded
Julius Marcellinus, a deputy prefect who was appointed prefect after the
death of Tenagino Probus, who had attempted to resist the invasion in .
The earliest date for Statilius is year  of Aurelian, the Egyptian year /,
in the Julian year . He was still in office in July/August of . Two
other pieces of evidence, one a papyrus and the other a passage that survives
into the Codex Justinianus from the Codex Gregorianus, likewise reveal that
Aurelian changed as little as possible in recovering the eastern provinces
from Zenobia. The papyrus records an edict of the prefect Hadrianus
Sallustius in which it is clear that members of the Boule in Alexandria
remained in office after the reconquest, while the passage from the period
of Diocletian refers to property that had been seized under the Palmyrenes
and never returned to its original owners. What is evident on the “micro”
scale in these texts would also seem to lie behind events recorded on
the “macro” scale in what pass for the histories of this period. Thus, for
instance, the Syrian army refused to accept the proclamation of Florian by
the army that had once served directly under Aurelian and later under the
emperor Tacitus after the latter’s assassination. The eastern army won, but
 For instances of Severan repression, see, e.g., Herod. ..; HA V. Sev. .; Mal. .; Dio ..
with Potter : , ; for Constantius, see Potter : –.
 Gagos and Heilporn .
 Lewis : , establishing the presence of Statilius Aemilianus on the text and reading Aurelian
as emperor; see also Rea  and his discussion on P. Oxy. . Statilius is also prefect on P.
Wisconsin  (May  of either  or ). P. Oxy. , where he is also in office, has no regnal
year; it may fall in the period of the reconquest.
 P. Oxy. .–: %dri†nov SalloÅstiov ¾ diashm»tat[ov ›parcov] | A«gÅptou l”geiá
prostacq•n Ëp¼ .[. .] . . . . . . .[. .] | tän bebouleuk»twn ™kaston ™n t. λ. amprot†t | %lexan-
dr”wn p»li ˆp¼ toÓ deut”rou ›touv tv | AÉrhli<an>oÓ toÓ –n qeo±v basile©av, diatreib»n-
twn | te oÉk –pª tv p»lewv m»nhv, ˆll‡ kaª –n Œpasin to±v tv A«gÅptou nomo±v, t†lanton
e«skom©sai pr¼v tŸn tän . . .  letters . . . ].wn q”rmwn ™pi|skeužn; CJ ..: Impp. Diocletianus
et Maximianus AA. et CC. Agrippae. Cum cognatum tuum ingenuum, factum Palmyrenae factionis
dominatione velut captivum, distractum esse dicas, praeses provinciae ingenuitati suae reddi eum efficit.
S iiii Id Ian AA conss.
 The unity of the Roman Empire 

Probus appears to have had a great deal of difficulty enforcing his authority
in the western reaches of his empire, areas that had once been part of the
imperium Galliarum. Although that regime had been created by an officer
whose origins we do not know (Postumus), it ended up under the control
of men whose roots were implanted within the aristocracy of northern
Gaul. Carus, who overthrew Probus, was himself of Gallic extraction, and
seems to have been little loved by the eastern army, even though he led
it to victory in Mesopotamia; that much may be divined – if not from
the dubious tale that his tent was incinerated by a thunderbolt – from the
fact that the officials in that army murdered his son and heir, Numerian,
somewhere on the road between Antioch and Nicomedia, and plainly had
no interest in submitting to the authority of his elder son, Carinus.
Diocletian evidently recognized the perils of “regionalism” in asserting
that the empire remained a whole under the rule of the senior Augus-
tus, but this point could not overcome the tendency to recruit locally
into expanded bureaus of government. Local groups of imperial admin-
istrators could be highly resistant to outside pressure, and, within two
years of Diocletian’s abdication, two of them had thrown up “usurpers”
against the authority of the senior Augustus. Although we lack the ability
to reconstruct the membership of one of these regimes – Constantius I’s
administrative group in  – it is possible to do this to some degree for
that of Maximian. In the course of his years of power, it appears that Max-
imian had come to rely heavily upon officials of Italian or North African
extraction, and that these people were willing to support his son, Max-
entius, when he raised the standard of rebellion. Of Maxentius’s three
known praetorian prefects – Manlius Rusticianus, C. Ceionius Rufius
Volusianus, and Ruricius Pompeianus – one, Ceionius, is known to have
been a member of one of the most powerful senatorial families of the third
through fifth centuries and was himself very likely related to Nummius
Tuscus, who served as urban prefect in . At the same time, the list of
prefects of Rome reads like a virtual Who’s Who of the upper echelons of the
Italian aristocracy. The list includes C. Annius Anullinus (–), Attius
Insteius Tertullus (–), Aurelius Hermogenes (–), the Ceionius
Volusianus who also served as praetorian prefect (–), Aradius Rufi-
nus (), and, for religious reasons, Anullinus a second time (Maxentius

 For the events of these years, see Potter : –.

 PLRE, s.v. “Volusianus” () = PIR2 R  with further discussion – see Christol . For Nummius
Tuscus, see PLRE, s.v. “Nummius Tuscus” (); PIR2 N . Manlius Rusticianus is probably to be
identified with the Manilius Rusticianus who was praefectus annonae at Rome under Maximian, see
PIR2 M .
 david potter

seems to have thought he was a good-luck charm whose presence in office

would guarantee victory in the battle of the Milvian bridge). The fam-
ily of Anullinus (possibly the very family rumored to have once owned
Diocletian as a slave) can be traced back as far as the late Severan period,
when C. Annius Anullinus Geminus Percennianus, who owned extensive
estates near Theveste, appears in the ranks of the Arval brethren. Insteius
Tertullus, whose family likewise supplemented their Italian fortune with
extensive holdings in North Africa, was another man who could trace his
senatorial lineage to the Severans. So, too, Aradius Rufinus was a member
of a family whose dignity was Severan in origin, and estates linked them to
North Africa. Aurelius Hermogenes was of somewhat less exalted lineage,
being the descendant of a third-century equestrian official, and atypical in
this group in that he had been governor of an eastern province (albeit
Asia) in the time of Diocletian. For only two of Maxentius’s prefects –
Statius Rufinus (–) and Iunius Flavianus (–) – is there insufficient
information to enable us to know more of their careers. Moreover, in the
cases of those whom we do know, we also know that their families contin-
ued to flourish in the post-Maxentian period. Indeed, the entire reign of
Maxentius, with its extensive building projects in and around Rome, looks
at times like an extended effort to buy the loyalty of the capital (the one
exception to this being a measure to force some members of the senate to
contribute to the grain supply). The overall pattern, linking North Africa
with Italy, would also continue to be much in evidence during the rest of
the century as the Italian aristocracy continued to assert some influence
over increasingly distant emperors.
In their relationship with the Italian aristocracy and, given the strong
resistance on the part of this group to the attempt of Galerius and Severus
to assert their authority – and, one suspects, the authority of officials with
different backgrounds – the regime of Maximian and Maxentius offers
a paradigm for the relationship between imperial houses and regional
aristocracies that staffed regional bureaucracies in the course of the fourth
 PLRE, s.v. “Anullinus” (); PIR2 A  (note the connection with Diocletian asserted in Eutrop.
Brev. ..; Anon. de Caes. .; Zon .); see also PIR2 A  (Anullinus Geminus). He was also
the consular colleague of Nummius Tuscus (see no. ). See also Christol : .
 PLRE, s.v. “Tertullus” (); see also PIR2 I –; .
 PLRE, s.v. “Rufinus” () with PIR2 A – (family).
 PLRE, s.v. “Hermogenes” (); see also PIR2 A ; PIR2 A  (his ancestor); see also Christol
: –. His father was involved in mysterious ways with the death of Odaenathus.
 For Maxentius in Rome, see, in general, Curran : –; for the attribution of senators to the
navicularii, see with Corcoran a: .
 See on this point Matthews : –; note also the discussion on p.  of how the pattern of
landholding had indeed acted to unify court and senate starting in the early fourth century.
 The unity of the Roman Empire 

century. It was also the alliance between local officials and bureaucrats that
would, years later, set a man named Theophanes, memorably armed with
his martini, from Antinoopolis in Egypt to plead some cause at Antioch
in Syria. The further significance of these alliances is underscored by
Ammianus, especially in the period after the death of Julian, when imperial
bureaucrats wrested effective control of different portions of the empire
from emperors whose main qualification for office was dependence upon
those same bureaucracies in the wake of the all-too-powerful personalities
of Constantius II and Julian. It is testimony to the continuing power of
these groups that even Constantine – whose understanding of authority, on
earth as in heaven, tended to the severely unitarian – devised a succession
scheme that would have redivided the empire into four parts.
It is to Ammianus that we owe much of our ability to trace the decline
of imperial power in the wake of Julian’s demise. This is perhaps most
striking in the case of Valentinian when, as Ammianus points out, his
officials forbade him to come to the aid of his brother when it looked as if
Procopius might drive him from the throne. Valens ended up owing his
throne not so much to his own ability as to the influence of Arbitio, once
magister militum of Constantius. Likewise, Ammianus thought it con-
ceivable that Theodorus might inform Valens of the prophecy that he had
received concerning the succession, while Ammianus also makes it quite
clear that the prime figure in government at this time and the man most
concerned with picking a successor appears to have been the praetorian
prefect. Valens himself had no say in the choice of Gratian as successor to
Valentinian and, more significantly, no ability to react (any more than did
Gratian) to the blatant act of bureaucratic aggrandizement that resulted
in the appointment of Valentinian II as emperor by Meroboduus after
Valentinian I’s sudden expiration. And it is, of course, to the pen of
Ammianus that we owe the memorable picture of Petronius Probus, a man
whom even the emperor feared to move against openly.
Olympiodorus ended his history with a paean of triumph for the unifi-
cation of the east and west, with victory in a campaign where he may have
played a significant role in negotiating the outcome, and with a vision of
Helion, the eastern magister officiorum, placing the imperial robe on the

 Matthews : –.

 For possible linkage between Constantine’s vision of imperial power and the Nicene creed, see
Potter : .
 Amm. ..– with Matthews : .  Amm. ...
 Amm. ..;  with Potter : .  Amm. .. with Matthews : .
 Amm. . with Matthews : –.
 david potter

shoulders of Valentinian. Theodosius tried to unite the realms still fur-

ther with the symbolic issuance of a great law code. The time for symbolic
action had, however, long since run its course; by this point the institutions
that offered some capacity to rebuild a unified society were the Church and
the chariot factions. Both provided venues where the emperor could, and
did, encounter his subjects, but only in the east. While fifth-century coun-
cils certainly did not bring about peace within the Church, they did at least
allow for some direct communication between the emperor and his subjects
in their own language; similarly, while the circus was often at the epicenter
of protest, it was also an area where voices from outside the bureaucracy
were heard. Neither worked so well in the west, which lacked both the
institutional and economic resources to support either on the scale upon
which they existed in the east.
When it came to the relationship between the “western Romans” and the
“eastern,” the power of aristocratic office holders had effaced law, ethics, or
pleasure as binding forces to unite the Roman world. Appeals to principles
in government are not merely a sham and a screen, they are often statements
of values that can bind societies together, agents of mythologies that are all
the more powerful for the ability of people to make of them what they will.
When power is substituted for principle, a state must fail, and so, after so
many centuries, did the empire of Rome.

 Olympiodorus fr.  () (Blockley).

 chapter 2

Roman patronage
Peter Garnsey

a peculiar institution?
There is a problem about patronage. Patronage has traditionally been
viewed as a defining feature of Roman society, an institution of pivotal
importance, offering a vital clue to the relative stability of that society
over time. It was an institution as quintessentially and centrally Roman
as patria potestas, the authority of a Roman male head of household.
Nowadays, however, historians are saying that patronage was much less
significant than it has been made out to be, especially in the sphere that (in
their judgment) really mattered, politics. That is, in the first instance, for
the period of the republic, when political office was highly regarded and
fiercely competed for within the framework of a republican constitution –
before monarchy arrived and transformed the nature of politics. It used
to be thought, to put it simply and starkly, that the Roman aristocracy
under the republic controlled the voting population of citizens through
patronage, that what counted in elections was the quantity and quality of
one’s clients. The experts don’t believe this any more. They follow Peter
Brunt in his conviction that patronage relations were fragile, peripheral,
short-lived, and did not count for very much – especially, but not only, in
the late republican period.
As to the principate, again there is an old idea, now regarded as an
anachronism, that with an emperor in charge private patronage became
redundant. The emperor was universal patron, patron of all citizens, patron
 De Ste. Croix  stands out as a model study of the historical development of patronage; it
is relatively little known, perhaps because it appeared in a journal of sociology, but its value is
appreciated by some, including John Matthews. See Matthews :  n. . John’s own breadth
of interest and knowledge was already visible in Matthews . It gives me pleasure to acknowledge
here the inspiration of his scholarship and the warmth and continuity of his friendship. I am grateful
to Myles Lavan, Arnaldo Marcone, Neville Morley, and Greg Woolf for their enlightening comments
on an earlier version of this paper. Of writers on patronage who have influenced my research (and
their names are easily traceable in the footnotes and bibliography), I would single out Richard Saller.
 Brunt : ch. .


 peter garnsey

of all subjects. Private patronage simply lost its raison d’être. Brunt is on
the other side here. Glancing forward to the principate, he opined that
while patronage was marginal under the republic, it enjoyed a heyday
under the principate and flourished under the late empire. With regard to
the latter period at least, some historians have arrived at a rather different
conclusion. They think that old-style patronage was overshadowed by new
ways of gaining access to office and power, in particular – money.
Here is a definition of patronage:
First, patronage involves the reciprocal exchange of goods and services. Secondly, to
distinguish it from a commercial transaction in the market place, the relationship
must be a personal one of some duration. Thirdly, it must be asymmetrical, in the
sense that the two parties are of unequal status and offer different kinds of goods
and services in the exchange – a quality which sets patronage off from friendship
between equals.

The essential elements of the definition are three: patronage is between

unequals, it involves the exchange of benefits, and it is enduring. This can
only be a working definition, and not just because there is no definition
to be found in the surviving Roman sources. This definition is derived
from the anthropological literature, and from the study of traditional rural
societies, particularly in the Mediterranean. In a model society of such a
type, a large landowner is surrounded by small farmers, peasant proprietors,
or tenants working the big man’s land; nothing much changes; the structure
of society remains fixed. For all we know, Rome might have been such a
society at some early stage. But such a model is inappropriate for any
period of historical Rome from the middle republic to the late empire –
and we must allow for changes in the nature and the functions of patronage
relationships over time.
But also, it is possible that Roman patronage was rather individual, as
patria potestas was acknowledged to be: “Also in our potestas,” writes Gaius
the jurist, “are the children whom we beget in civil marriage. This right
is unique to Roman citizens; for scarcely any other men have over their

 Saller : . Patronage could be more than a dyadic relationship. See below, p. .
 There is a description rather than a definition of clientage in Dig. ... The jurist Proculus states
in passing that clients are inferior in terms of auctoritas, dignitas, and vires, that is, prestige, status,
and resources.
 See, e.g., Gellner and Waterbury .
 For that matter, patria potestas did not remain the same. In particular, if fathers ever had the power of
life and death over children – and for arguments against, see Shaw  – then this had gone by the
late republic. Note also that the rules surrounding emancipatio of the filius familias were gradually
relaxed. See CTh ..,; with Arjava .
 Roman patronage 

sons a power such as we have.” There are some indications that Roman
patronage was, and was held to be, a “peculiar” institution.
Claude Eilers provides a clue. The evidence for the exercise of patronage
by Romans over Greek cities, which is his subject, comes from inscriptions
that concentrate around the period of the late second century BCE to
the middle of the first. The Roman nobles who entered into patronage
relationships with particular cities were usually provincial governors. Eilers
noticed that the Greek cities in question used the Latin words for patron
and patronage to describe the relationships: they wrote patron or patreuein,
passing over in the process the various terms that they had at their disposal
for describing foreign benefactors – like euergetes, soter, proxenos. This
practice, says Eilers, “does not sit easily with the idea that patronage is a
cultural universal.” Later he describes the phenomenon as “the introduction
of a Roman social institution into the Greek East.”
An inquiry into the singularity of Roman patronage might begin with
ritual, in the first instance, salutatio. This was the early morning greeting
of patron by client – familiar from authors from Cicero through Seneca,
Martial, and Juvenal to Ammianus and beyond – and was the springboard
for a series of acts in which patrons and clients participated on a daily
basis. Galen of Pergamum, in his treatise On How to Recognize the Best
Physicians, alludes to salutatio in a way that implies that the word, if not
the institution, was foreign to him: he claims to have devoted himself
to the study of medicine instead of wasting his time in “what is called
‘salutation.’”
To take the matter further we need to inquire into the content of
patronage and the status of the patronage relationship. Eilers encourages
this approach when he suggests that the special Roman element is the level
of commitment to the relationship on the part of the superior party. This
might imply that the patron took rather a lot onto himself and, perhaps,
that he took risks in doing so.
Eilers was perhaps taking his cue from Dionysius of Halicarnassus,
the Greek historian and antiquarian who wrote imaginatively about early
Rome. Dionysius found Roman patronage (introduced, in his story, by
Romulus) to be different from and preferable to some Greek counterparts.
 Gaius, Inst. ..
 Eilers : , . Lavan () makes a different point, but also one relevant to Rome’s relations
with provincials, in demonstrating that Romans (to judge from sources such as Cicero and Livy)
made use of the concepts of patrocinium and clientela (in addition to the paradigm of slavery) in
representing the imperial relationship.
 Not only clients attended. See Yakobson : –, with refs.
 Iskandar : sect. .  Dion.Hal. .–.
 peter garnsey

It had a fine name, a lofty aim (protection of the poor and lowly), and
imposed obligations on both parties. It also brought positive results: patron-
age was the source of the social concord enjoyed by Romans until the arrival
on the scene of Gaius Gracchus. Dionysius was living in a dreamworld if
he thought patronage relationships could ever have been as balanced and
harmonious as he envisaged them. However, his account of the mutual
services associated with patronage is not without use and might bear some
resemblance to contemporary patterns of behavior or, at least, to patron-
age as a model or ideological construct as formulated by the Romans and
understood by this Greek observer of Roman society. Patrons (according to
the law of Romulus) were to give legal aid and material assistance; clients
were to provide their patrons with financial support when and as neces-
sary in their private lives and political careers. The list of services is not
comprehensive. Dionysius is mainly interested in asserting that Roman
patronage was not exploitative, in contrast with his Greek exemplars, in
that superiors gave as well as received (but only the financial obligations
of the client are spelled out). On the nature and level of commitment of
superiors, Dionysius is interesting: “For both patrons and clients alike it
was impious and unlawful to accuse each other in lawsuits or to bear wit-
ness or to give their votes against each other or to be found in the number
of each others’ enemies.”
Law or legal sanction finds no place in the definition of patronage with
which I began. It is normally held that patronage in Rome was a matter of
fides, grounded in social custom and tradition, not law. Now, Dionysius, as
we saw, thought that patronage originated in a law of Romulus. Patronage
might in fact have been at first grounded in law, but one would not
want this to hang on the testimony of Dionysius. It happens that there
are vestiges of a law of patronage in the Twelve Tables (mid-fifth century
BCE), where it is stipulated that a patron who defrauds his client is liable to
the death penalty. He is “sacer,” potentially a sacrificial victim to the gods,
and as such anyone can kill him with impunity. We find further links
between patronage and the law in the late-second-century law on extortion
associated with Gaius Gracchus. One clause has to do with the appointment
of an advocate; another, with the summoning of a witness. Neither was
permitted if there was a relationship of fides between the accused and a

 Cornell : ff., on warfare as a client’s duty.

 Twelve Tables .: patronus si clienti fraudem fecerit sacer esto; Brunt : –. C. Murgia, per
litteras, says that the key terms in Servius (who quotes the clause), namely, fraudem, clienti, and
patronus, are likely to be authentic, because Servius would have used, probably through Aelius
Donatus, a good ancient source.
 Roman patronage 

proposed advocate or witness or if such a relationship had existed among

their ancestors. The mutual obligations of patron and client are given
legal recognition and force here. The legal sources for the classical period
have nothing similar. They give scant attention to patronage. There is
no legal definition of patronage, and, a telling point, there is no case
law associated with patronage. Roman lawyers were never called upon to
offer learned opinions on patronage because patronage disputes never came
before the law courts. One might conjecture that a connection with the law
(likely enough in the early period) faded out, in response to an increasing
reluctance among upper-class patrons to expose themselves to the higher
risks and other disadvantages that went with a legal obligation to clients.
There is, however, an anomaly. One thing that does get a lot of coverage
in the legal texts is the relationship of an ex-master to his ex-slave, and the
former is invariably patronus. This relationship receives attention from the
jurists because it is a spin-off of slavery, a by-product of the manumission
of slaves. The granting of freedom to a slave was a momentous act, and
it was not done lightly. Strings were attached, and that’s what all the
juristic discussion is about – the strings. It should be stressed that the
emancipation of slaves in quantitatively significant numbers was a special
feature of Roman society and gives a whole new dimension to Roman
patronage as well as underlines its social specificity.
That is, if the kind of patronage we are talking about is patronage at all. In
the texts the superior is patronus and the inferior libertus, “freedman,” never
cliens. Most scholars are unwilling to call this patronage. Eilers has some
light to shed on this issue. He identifies two kinds of patronage in the Greek
East: one voluntary, the other involuntary. Voluntary patronage is that
entered into with a provincial city by a Roman noble on the city’s initiative.
Eilers sees this as modeled on vertical patronage between noble and free
plebeian, classic patronage, if you like. Involuntary patronage involves the
conquering general and the city that has surrendered unconditionally to
him. The relationship is well attested but has also proved problematic
for some historians. Eilers suggests an appropriate analogy would be with
another kind of patronage in Rome, that connecting patronus and libertus.
There is an untold story here (and this is not the occasion to tell it). It
concerns a period in Roman history when status differences among social
 Lex repet. (Crawford ) no., line .
 The obligations were not entirely one-sided. See, e.g., Dig. .. (the patron must provide alimenta
if the freedman cannot maintain himself ); also Dig. .., , ,  (operae, “services,” should not
be life-threatening, or damaging to the reputation of the freedman/woman, and the obligation to
perform them ceases where the freedman/woman has risen in social status or produced two or more
children).
 peter garnsey

inferiors were less marked, when slave and filius familias and freeborn client
and freedmen were all on a spectrum, neither wholly subject nor wholly
free. There are markers of this earlier era in the legal sources, for example
in their discussions of the legal positions of sons and freedmen. The jurists
often couple the two, as in the Digest section “On the obedience to be
offered to parents and patrons.” Sons and freedmen appear to have a
considerable amount in common. The surprise is not so much that the
freedman is (rather) like the filius emancipatus but that the filius emancipatus
is (rather) like the freedman. This presumably goes back to some early time
when in the household there was little difference between sons/daughters
and slaves, except in expectations. Such social arrangements could not
survive the substantial changes that Roman society underwent from the
second century BCE on, including the introduction of chattel slavery on a
massive scale. The second type of patronage whose existence I am proposing
was a victim of these changes, which created a society highly differentiated
at the bottom as well as at the top.
However, slaves, selected slaves, continued to be manumitted – and
on a considerable scale. The stock of patroni and liberti was continually
being renewed. And this leads to another point (which again I cannot
develop here). The patronus/libertus relationship could be, and sometimes
was, transitional to the “standard” patronage relationship, between patron
and freeborn client. Whatever restrictions bound the freedman did not
apply to a son. The patron or patron’s son and the freedman’s son could be
involved in a fully fledged, orthodox, patronage relationship. I think this
happened routinely. If a freedman was fortunate enough to found a family,
then that family might move easily into a patronage relationship with the
family that had permitted it to evolve in the first place.
We should bear in mind that a freedman’s son was freeborn: he did
have a father. The profile of the freedman’s son that I am talking about
is of someone who has become the beneficiary of extremely warm and
mutually beneficial relations between his father and his father’s former
master. The freedman/father had served his patron loyally and done him
services beyond those he was obliged to do and beyond the call of duty: in
terms of roles and behavior he was hardly distinguishable from a client. In
consequence, the two families came together in a patronage relationship.
 Cf. Crook : .  Dig. ..
 Under the republic such freedmen might play significant supporting political roles. See Treggiari
: –. One thinks of Butas, in Plut. Life of Cato, or Phania and Cilix, freedmen of Appius
Claudius; see Cic. Ad fam. .ff. See also Cic. Pro Sestio on “libertini optimates.” For counterparts
under the empire, see, e.g., Tac. Hist. ., .; Fronto, Ad L.Verum .. In such texts liberti and
clientes often appear together. In Pliny, Hist. Nat. ., a freedman’s devotion to his patron is one
of four exempla pietatis. See Saller : .
 Roman patronage 

We have come upon one way in which patronage reproduced itself,

and a peculiarly Roman way at that. The patron/freedman relationship,
whether or not we grace it with the title “patronage,” was a fertile seed-
bed for “regular” patron/client relationships. And this phenomenon has a
relevance that extends beyond the lower rungs of the social ladder, given
the opportunities for upward mobility that Roman society offered those
with the right connections. Take P. Helvius Pertinax – an extreme example,
perhaps, but one that serves to make the point that patronage had the
capacity both to reproduce itself and to serve the greater cause of the repro-
duction of the governing class. The father of Pertinax was a slave who was
later manumitted. Pertinax had his first important break, an appointment
into the centurionate, by courtesy of his father’s patron, one Lollianus
Avitus, who happened to be a leading senator. Avitus was not the patron of
Pertinax’s father by virtue of having once owned him as a slave. He was a
later acquisition, as it were. How that patronage relationship developed is
not known, but it seems that Pertinax was born and grew up on an estate
belonging to the family of the great man. Later on, Pertinax benefited from
the patronage of someone even more influential, Claudius Pompeianus,
son-in-law of the emperor Marcus Aurelius. Pertinax reigned as emperor
from  December  to  March .
If the sources were kinder to us, we would be able to construct a fuller
picture of the characteristic features of the Roman version of patronage.
This is an appropriate way of putting it, because patronage was by no
means peculiar to Rome; rather, there were, and are, other versions of
patronage that might differ from the Roman version. In defense of the
definitions employed by social anthropologists, one might say that they
serve the purpose of heuristic devices to facilitate comparative analysis,
without any implied claim that patronage is a “cultural universal” whose
local versions are all identical.
Meanwhile, there is a pressing task at hand, namely, to inquire into
the importance of patronage within Roman society. For, as I indicated at
the beginning, this is something that can no longer be taken for granted.
A full treatment of the issue is out of the question, and the limitations
of my discussion will be obvious enough. Most conspicuously, the first
two sections that follow are concerned largely with the city of Rome alone
 SHA Vita Pert. –. Purcell (), in a model article, shows how patronage lay behind the impressive
social mobility of assistants of magistrates, apparitores, who were typically of freedman status.
 The reticence of the sources is an issue that emerges clearly in connection with the role of patronage
in the late republic. P. Brunt, the founder of what might be called the current orthodoxy, is
apparently willing to talk about patronage only when the sources talk about it explicitly, whereas
one might reasonably trace the workings of patronage even when there is no such explicit mention
or when a relationship is expressly not described in these terms. See next section.
 peter garnsey

(especially in the republican period). Further, this was a society undergoing

significant changes and disruption. If there were such a thing as “normal”
Roman patronage, then it would be unlikely to be on show between the
end of the second century BCE and the mid-first century CE. Ironically,
late antiquity might turn out to give us a better view of the “norm” in
patronal relationships or, at least, help us to see how patronage, both as
conceptualized by the Romans and in practice, was distorted under the
particular circumstances of late republican politics and the establishment
of the principate.

patronage in late republican rome

Not long ago it was the accepted wisdom that the Roman nobility of the
republic had a stranglehold over Roman politics through its networks of
patronage and friendship. This consensus was smashed by Peter Brunt.
He found that patronage relationships were unstable and of marginal sig-
nificance, especially in the last century of the republic. The senate, in
which most of the powerful patrons were gathered, had lost control. Their
supposed connections with the equestrians, the municipal gentry of Italy,
and the little people of Rome did not count for anything when the chips
were down. Political alliances of all kinds were made and unmade. Inher-
ited friendship could not be leant upon, let alone patronage relationships.
Brunt’s arguments have been very influential.
Late republican Rome was a fiercely competitive and disintegrating soci-
ety, in which established values and practices were progressively disregarded
and pushed aside. This was not an environment in which patronage could
flourish. That it did not have a role to play at all, and was altogether
insignificant, is more uncertain. John North, in reviewing Brunt’s book,
made two pertinent observations. He pointed first to the “sweeping neg-
ativity” of Brunt’s treatment, which is “overwhelmingly powerful for the
purpose for which it is designed” – to show that the role of patronage in
politics had been vastly overrated – but does not appear to prove the
contrary, which is that “client relations . . . were altogether of negligible

 As regards the imperial period, if Roman patronage is (as Dionysius thought) markedly different
from any Greek model, then it would be interesting to follow the effects of Greek incorporation into
the empire on Greek social behavior. Likewise, it would be worth asking how far Roman practices
were exported to the western provinces and, if so, how they might have interacted with traditional
forms of dependence and social relationship.
 Millar : ; Yakobson ; Mouritsen : –, –; Morstein-Marx : ; etc.
 North : .
 Roman patronage 

importance in Roman Republican society.” In other words, Brunt’s contri-

bution was essentially destructive, was directed at a very narrow target, and
was successful largely on those terms. North goes on to note that Brunt
presents his position as the product of his reading of the sources. Of his
conviction that patronage was inefficacious and irrelevant, Brunt writes,
“This is the only interpretation of Roman politics which is commended
by the ancient sources.” But this position is, at least to a degree, based
on an argument from silence. According to Brunt, clients are much less
conspicuous in the sources than they ought to be, should the thesis that he
is attacking be correct. North comments:
Of course, Brunt is right to point out the disproportion between the direct evidence
of patron/client relations and the importance given to them in modern theories;
but that does not itself settle the question. It is quite possible that the basic
relationships of society, so familiar to contemporaries, should be assumed and
rarely referred to in contemporary texts; it must be asked therefore whether we
can reasonably expect clientship to obtrude frequently and overtly into the kind
of discourse that survives from the Republican period.
There is room for a reconsideration of the role of patronage in Rome
that moves beyond the narrow (if crucially important) matters of voting,
elections, and political success. Andrew Wallace-Hadrill promisingly shifts
the scene from the hub of politics to, more broadly, society, ideology, and
cultural experience. He begins:
To stand at the door of an upper-class Roman house of the late republic or
early empire is already to glimpse something of the centrality of patronage in
Roman society. From the threshold a vista opens out, elaborately and symmetrically
framed. . . . At the focal point, in the centre of the tablinum, belongs the master of
the house, seen as Cicero pictures the great noble of an earlier generation: “both
walking about in the forum and sitting at home on his throne, he was approached,
not just about questions of law, but about marrying a daughter, buying a farm,
cultivating the land, in fact about any point of social obligation or business.”
From an evocation of the physical ambience of patronage and of the way
a Roman noble saw himself and wanted to be seen Wallace-Hadrill directs

 Brunt : .

 In fact, the sources are by no means silent on patronage, as Brunt’s own footnotes show. Among
the more interesting texts are Auctor ad Her. ., .; Cic. Post red. In Sen. ; Parad. St. –; De
part. orat. ; Pro Mur. –; Post red. ad Quir. ; Ad fam. ..; Ad Quint. Frat. .., ..; Ad
Att. ... See, in general, Deniaux .
 Wallace-Hadrill . Nor should the economic aspects of patronage relationships be ignored. See
Verboven .
 Wallace-Hadrill : , citing Cic. De orat. ..
 peter garnsey

us to the wider role of patronage in promoting social cohesion and social

integration. The focus is on cross-class vertical relationships, which are
paradigmatic for patronage (and to this aspect of patronage the Cicero
passage is an appropriate introduction).
Is patronage in the late republic, then, properly the subject matter of
social and cultural rather than political history? The division is an artificial
one. “The Roman noble felt himself almost naked without an entourage
of dependants, who acted as a visible symbol of his social standing.” “The
Roman noble was above all a homo politicus, driven by the ambition to
reach, hold and exercise power.” The link between these two claims is
not as direct and straightforward as was once thought, because, thanks to
Brunt and others, we can see that patronage was not in itself a ticket to
political success.
But a link exists. First, patrons looked to clients for general support of
their political activities. Roman magnates used lower-class clients (often
together with freedmen) to enhance their public image and to advance
their political ambitions. We should not, of course, look to them to be
open, generous, and grateful over such services, and they are not. Thus, for
example, in a tendentious passage from the Pro Murena, Cicero appears to
see nothing between, on the one hand, a lower-class clientele of “men of
small means” who have little to offer a patron besides constant attendance
and, on the other, “friends” in the senate and the equestrian order with
whom it was possible to practice a meaningful reciprocity – and Cicero is
speaking the language of patronage here. He writes:
Men of small means are only able to earn favors from our order or pay us back in
one way, and that is by helping us and following us about when we are candidates
for office. It is not possible and it cannot be asked of us senators or of the Roman
knights that they should attend for whole days their friends who are candidates. If
they come in large numbers to our houses and on occasion accompany us down
to the forum, if they condescend to walk with us the length of a public hall
we think we are receiving great attention and respect. It is the poorer men with
the time available who provide the constant attention that is habitually given to
men of standing and to those who confer benefits . . . . As they are always saying,
they cannot plead for us, stand surety for us, or invite us to their homes. They ask
us for all these favors, but think that they can only repay us for what they receive
from us by personal service.
Underlying the “help” (opera) that lower-class clients might give a can-
didate besides forming the core of his escort is a whole raft of services that

 The first statement is Wallace-Hadrill’s; the second, mine.  Cic. Pro Mur. –.
 Roman patronage 

are alluded to in other literature, including acting as agent, manager, and

financial contributor to a candidate; Cicero predictably is not interested in
spelling out such services. Similarly, we may suspect that among the “needy
but good” men whom Cicero in De officiis deemed to be suitable clients –
and who are compared favorably with the rich and ungrateful – were some
who had rather more to offer a patron than mere gratitude.
Second, a step or more above the “respectable plebs” just considered
were clients who might also be friends (at this social level the two terms
flowed together). The Younger Seneca writing under Nero provides an
interesting detail about the way the great men of late republican Rome
graded their visitors at the salutatio: “Gaius Gracchus, and a little later,
Livius Drusus, were the first to set the fashion of classifying the crowds of
their followers, and of receiving some in privacy, some in company with
others, and others en masse. These men thus had chief friends, ordinary
friends, but never true friends.” Such “friends” would have included
members of at least the lesser aristocracy: junior senators, equestrians,
ambitious municipal notables, people on the make who had property and
were nourishing the prospect of social and political advancement. They
were in the hunt for men of power and influence to promote their cause.
They were, in other words, looking for patrons. As Wallace-Hadrill puts
it, in a rhetorical question: “How effectively could any individual or group
gain access to the resources controlled by the Roman state (of whatever sort,
judgment, privilege, status, power, money) except through personal links of
patronage?” One may profitably look for such people, for example, among
the staff of a provincial governor; the support group that Cicero himself
assembled for his proconsulship of Cilicia can be reconstructed from his
letters. Cicero’s correspondence is a useful guide to his patronal activity as a
whole, in particular the letters of recommendation, of which a small sample
survives. As patronal correspondent Cicero can surely stand proxy for the
whole senatorial class of his time. Looking ahead, we can see that Cicero’s
letters of recommendation set a pattern that recurs, with variations, in the
works of the Younger Pliny, Fronto, and Symmachus, among others. The
friends peppered with such letters were typically officials in a position to
distribute favors, for example as provincial governors.
Brunt was right to remind us that patronal and other personal links
were fluid and unstable in the late republic. One need only think of the
Roscii of Amerinum, who “had many patrons and hospites inherited from
 Cic. De off. .–.  See, e.g., Saller ; Deniaux : –.
 Sen. De ben. ..  Wallace-Hadrill : .  Cic. Ad fam. , from – BC.
 See Cotton ; at p.  n.  she cites the correspondence of Gregory the Great.
 peter garnsey

their ancestors whom they ceased to cultivate and pay observance to, and
betook themselves to the fides and clientela of Chrysogonus.” But we need
to look beyond the success or lack of it of individual patrons in advancing
their protégés, or even in holding their allegiance, to the highly significant
(though far from exclusive) role played by patronage in reproducing the
governing class of republican Rome.

patronage under the principate

Was private patronage incompatible with the rule of one man who was
universal patron? A monarch could not be patron of all in the same way as
he could be patron of a few. In a patronage relationship a few individuals
and families are favored. If everyone is a client, no one is. If we talk of
the emperor as universal patron, we are using the word in a loose sense,
of the man with the most power and resources in the state and with the
unequaled capacity to use them for the benefit of his subjects, for example,
as the giver of bounty. And, in fact, liberalitas is the main virtue of emperors
who are acting as patrons on a large scale. It is on display, for example, in
the provision of free grain or in cash handouts to hundreds of thousands
of Romans. Only an emperor could afford this; or, to put it in another
way, only an emperor could divert state funds toward these ends. Further,
liberalitas had to be an imperial monopoly. Anyone else who indulged in
it would be competing with the emperor and challenging his rule.
The emperor’s liberalitas undercut personal patronage to the degree that
it reduced the responsibility of private patrons for the material welfare of
their clients. But there was still a lot for private patrons to do. Imperial and
private patronage complemented rather than competed with each other.
The mutual exchange of services that (ideally) characterized patronage rela-
tionships remained. Patronage was integral to the lifestyle of the nobility.
Emperors knew this well: they were themselves, after all, promoted nobles.
So private patronage could subsist under monarchy. Brunt held that
it did not just survive but did well: though weak under the republic, it
flourished under the principate. This is a paradoxical result. Patronage
became more prominent as the political role of the citizenry became less
meaningful. For Brunt, “meaningful political role” boils down to voting

 Cic. Pro Roscio Amer. , quoted by Brunt (: ). Chrysogonus was a notorious freedman of
Sulla.
 An emperor might have his own private clients. See Suet. Div. Aug. .; Pliny, Pan. ..
 According to Brunt (: –), clients needed patronage networks more under the principate
to enhance their careers; and citizens were more vulnerable because citizenship counted for less.
 Roman patronage 

power. The republic was dead and buried, its voting system defunct. And
Brunt thought that clients could do nothing of any political value for
patrons under the principate. How then can patronage be said to have
flourished? In fact there is plenty of evidence that clients were part of
the support system that was essential for any ambitious politician under
the principate. Such clients included plebeians as well as people on the
fringes of the nobility. They are likely, however, to have been men of some
respectability, means, and status. In what follows I focus, rather, on the
role of aristocratic patrons in advancing their clients, more specifically, in
the general context of the recruitment of the governing classes of Rome and
the empire. This is the main subject of a model study by Richard Saller.
There was a constant need for the co-optation of new members into the
Roman aristocracy: in the first instance the senate, but also, and increas-
ingly, the equestrian order. The high mortality rates that were normal and
inevitable in premodern societies, and that applied across the whole social
spectrum, lie at the bottom of this. Life expectancy at birth lay in the
range of twenty to thirty years. In the case of the aristocracy, there were
additional factors: unnatural deaths of individuals who fell foul of emper-
ors and the not-unrelated retirement into private life of some members of
noble families.
So how were replacements to be found? The upwardly mobile had always
needed patrons to enhance their careers. But under the principate there
was a modification in the structure of the patronage system, at any rate
at the top levels of society and in this, the public, sphere. Whereas under
the republic the efforts of patrons to promote clients into or within the
governing class had to be put to the test in the ballot box, under the
principate the promotion system rested on the decision of someone in
authority, the emperor or one of his friends or trusted advisors. Political
patronage now operated within a pyramidal structure: the emperor was at
the top, with a chain of dependence leading down from him. The final
arbiter was the imperial will. However, in practice the emperor was in no
position to choose personally anyone much beyond his own entourage.
He necessarily leaned on the counsel of the people closest to him, friends,
top officials, and other courtiers. Subimperial patronage was absolutely
essential in the bringing forward of the next generation of aristocrats and
administrators.
 See, e.g., Tac. Hist. ., ., ., .; Ann. .; ..  See, e.g., Tac. Hist. ..
 Saller .  See Scheidel , with bibliography.
 Of course, magistrates and promagistrates routinely distributed posts that were at their disposal and
other favors to clients or clients of friends, without reference to the emperor.
 peter garnsey

Saller called this class of intermediaries “brokers” and those they helped
“protégés,” taking his cue from the anthropological literature. The use of
this terminology carries with it the acknowledgment that patronage is not
solely a dyadic relationship. B has a continuing relationship (of patronage
or friendship) with each of A and C and acts as the link between them. The
system is recognizable as patronage, despite the lack of a direct connection
between A and C. By the same token, the overt description in a text of a
relationship as patronage (notably by the use of patronus and/or cliens) is
not a necessary condition for the relationship to be classifiable as such. A
letter written by Fronto to Lucius Verus, brother of Marcus Aurelius and
coemperor, soliciting high office and financial support for a fellow senator
who is his junior, Gavius Clarus, is pertinent in this regard. Fronto is
suitably unwilling to call Clarus a client, choosing to say instead that his
protégé’s deferential behavior approximated to that of clients and worthy
freedmen. When he does select a word to characterize their relationship,
he opts for the safe term amicitia, “friendship.”
This is one of several texts that demonstrate that Roman aristocrats
were coy about using the terms “patron” and “client” to describe their
(own) relationships. This is the inference to draw from Fronto’s delicate
circumlocutions, not that the relationship in question was not a patronal
one, and even less that it is inappropriate to treat any such relationship as
an instance of patronage. Saller was challenged on this point, his critics
holding that a defining characteristic of a patronage relationship is that it
is vertical, by which they meant that it was a cross-class relationship, one
that connected the upper classes to the lower classes, the propertied to the
nonpropertied. To be sure, much of the old debate on patronage was
carried on with this assumption: did the nobles or did they not control the
plebs through patronage?
Let us consider briefly the situation of Saller’s protégés. These were peo-
ple who at the time when we glimpse them do not have status and power
comparable to their champions. There is asymmetry and verticality, even

 Boissevain ; .  Fronto, Ad L. Verum ..

 Cf. Cic. De off. .–; Sen. De ben. ...
 As distinct from such relationships between others. See, e.g., Tac. Ann. ., ., .; Hist. .
(Vespasian as client of the Vitellii), .. These texts refer in the main to relations within the Roman
aristocracy.
 Saller  is a reply to his critics.
 In some instances there might come a time when the position of the two individuals and families
concerned are equalized or even reversed. A classic case is C. Marius. See Plut. Mar. .. Marius
declared that the claims to a patronage relationship over him of a noble, Herennius, were redundant,
now that he had acquired a senior magistracy.
 Roman patronage 

if they are not as marked as in the case of upper-class/lower-class rela-

tionships. The protégés might have stuck precisely where they were – or
backpedaled – without the intervention of their social superiors. And not
all protégés improved their position. Ronald Syme devoted an article to
“Pliny’s less successful friends.”
Syme’s paper reminds us that entry into and promotion within the
governing classes of Rome were strongly competitive. There were rela-
tively few positions – well under a thousand – and the catchment area of
potential recruits expanded as local elites prospered under Roman rule and
increasingly looked beyond their cities, regions, and provinces; in this they
received encouragement and support from Roman aristocrats who were
themselves increasingly of municipal or provincial origins and developed
patronal connections with their home cities. What marks off the promo-
tion system of the early from that of the late empire is not, therefore, an
absence of competition in the former; it is rather the dominance of the
traditional patronage system. There was as yet no bureaucratic ideology
to compete with and undermine the traditional, aristocratic value system
in the selection and advancement of officials. “Seniority” and “merit” did
not yet count as against the aristocratic virtues of birth, excellence, and
wealth. Again, if a position of any rank was purchased, this was regarded
as irregular and scandalous and was typically the product of the machina-
tions of people close to the emperor of lowly status and inordinate power,
such as the freedmen of Claudius.

patronage in the late empire

Ammianus Marcellinus in the course of digressing twice on social life in
Rome shows us a noble holding court at home. The great man receives
the crowds who come to greet him every day, separating off “his own,” who

 Syme .  Millar : ; Saller : ch. . And see below, p. .
 Cassius Dio ..; cf. Millar : . However, buying of offices was standard practice among the
apparitores, that is, the assistants of Roman magistrates (scribes, messengers, lectors, heralds, etc.),
and this might be said to have created a precedent for the late imperial bureaucracy. See Jones 
and Purcell . In addition, money (and other goods of value) changed hands in other contexts;
for example, it went into the pockets of provincial governors in return for services rendered (or
sought). For a famous case, see Pflaum  (the Thorigny marble). If the official concerned was
“unfortunate,” he might be faced with an indictment for extortion. See Brunt . Again, advocates
were not supposed to take fees, according to the traditional aristocratic value system, but in practice
might do so. See Crook : –.
 Amm. . and ., with Matthews : –. On patronage in the late Roman empire, the
older discussions of de Ste. Croix () and Jones (a: –) are still valuable. For more
recent full (and contrasting) treatments, see Krause ; MacMullen ; and C. Kelly .
 peter garnsey

are “counted in,” from others who include “strangers,” such as Ammianus
would have been. Due recognition is given to status differences among
the callers, as symbolized in the manner in which they are permitted to
greet the great man: “Some of these men, when one begins to salute them
breast to breast, in the manner of threatening bulls turn to one side their
heads, where they should be kissed, and offer their flatterers their knees
to kiss or their hands, thinking that quite enough to ensure them a happy
life.” He dispenses favors (handouts, dinner invitations, and so on) and
enjoins duties (attendance during the day), as appropriate. John Matthews
writes: “The element of satirical distortion in Ammianus’ portrayal needs
no emphasis, nor the visually exuberant, even theatrical, quality of his
choice of detail. The result is a caricature of the social manners of the
classes of Rome, resembling not so much serious analysis as the parodies
of popular comedy, less photographic archive than portfolio of cartoons.”
And, concerning Roman senators in particular: “It is impossible to state
any ‘scientific’ conclusion as to the moral standards and general seriousness
of the senatorial class in Rome.” Still, Ammianus’s historian succeeds in
reading him with sympathetic intelligence (entertaining his readers along
the way). We can note that Ammianus’s account of the salutatio contains
good detail and is laced with technical terms. He had clearly observed the
event, and may even have taken part in it, though evidently with no great
pleasure.
Ammianus came to Rome in the early s to write history. If he knocked
on the doors of Roman nobles, it was in search of literary patronage, not
career advancement. His career – as protector domesticus in the emperor’s
mobile army – was long over. Naturally he was an unquestioning supporter
of the traditional system of patronage, of which the salutatio was a central
institution, and in this he was representative of his class. “It would not
have occurred to Ammianus or his contemporaries to deny the propriety
of influence – suffragium – in the promotion of perceived and reliable
virtue.” Ammianus was himself a beneficiary of the system, having been
selected by the general Ursicinus for his staff. This was another case of
nepotism, “that is, of the normal workings of the system.”
The wheels of patronage continued to turn in the late empire and
the continuities are conspicuous. Letters of recommendation from the
 Matthews : .
 For indirect evidence that salutatio was “alive and well” in the late Roman empire, we can turn to
the archeological record of reception rooms of great houses in diverse parts of the empire. See, e.g.,
Ellis ; ; ; with bibliography.
 Matthews : –, for political patronage in Ammianus. The citation is from p. .
 Matthews : , with the conjecture that Ammianus’s father was known to Ursicinus.
 Roman patronage 

Roman senator Symmachus composed in the late fourth century recall

those from the pens of Cicero, Pliny, and Fronto. Here are two specimens,
both addressed to powerful praetorian prefects, Nicomachus Flavianus and
Flavius Neoterius, respectively:
The merits of Sexio, formerly governor of Calabria, are well spoken of by many,
who have therefore requested that I should recommend him to your patronage
[suffragium]. It is part of the generosity habitual to your nature, to honour with
your affection those whom others have found congenial. I ask you, then, that if you
find no obstacle against satisfying the wishes of those who make this request, you
allow Sexio to draw the benefit of my words and the hopes of his many admirers.
I am delighted that my friend Drinnacius has proved acceptable to your judgment;
for it conveys very great honour on me also, when our sentiments are in accordance.
He will therefore enjoy the richest possible rewards for his blameless integrity of
character, having won my affection by his proper attention to social duties, and
your respect by the strictness of his professional conduct.
If neither Sexio nor Drinnacius is heard of again (though the latter, it
seems, had at least got out of the starting blocks), this merely confirms
that patronage was still a competitive business. Libanius of Antioch, whose
copious correspondence reveals his energetic activity as a patron, would also
have had “less successful friends.” The other side of the picture is revealed in
the fact that among the officials on whom Libanius pressed his pupils and
other clients and friends were a number who had reached their positions
with the aid of his testimonials. The fruitful operation of patronage
can be traced in the entry into the imperial administration of a number
of relatives of Ausonius, the rhetorician and poet of Bordeaux who was
tutor to Gratian; the arrival of a group of Pannonians under Valentinian I;
and the advance of Spaniards under Theodosius I. Looking ahead to the
early sixth century, we can follow through the writings of John Lydus the
successful promotion of a number of Lydians in addition to himself, with
Arcadius’s prefect Zoticus of Philadelphia, John’s hometown, at the head.
Continuity with the past is similarly visible in the exercise of patronage
in a legal context. These letters from Fronto and Symmachus were written
two centuries apart but are in complete harmony:

 Of course there is some self-conscious classicizing at work, with, e.g., Symmachus’s letters recalling
those of his predecessors deliberately and intertextually. Pliny and Symmachus (among other
sources) are fruitfully used by C. Sogno (in this volume, ch. ) to throw light on the arrangement
of marriages among the Roman upper classes. This is an aspect of patronage which is not discussed
in the present chapter. But see pp. – concerning John Lydus.
 Symm. Ep. ..  Ibid. .. See Matthews .
 See Liebeschuetz : –.  Hopkins ; Matthews : –; C. Kelly : –.
 peter garnsey

Fronto to Claudius Severus, greeting. The custom of giving recommendations is

said to have originated in good will, everyone wanting to introduce his friends
to other friends of his and establish good relations between them. By degrees the
custom grew up of giving recommendations even on behalf of those who were
involved in public or private lawsuits to the actual judges or their assessors – and
nothing reprehensible was seen in the practice. This was not done, as I see it, in
order to tamper with the fairmindedness of the judge or to cause him to depart
from giving a true judgment. But as there was a time-honored custom in the courts
of bringing on witnesses to character, after the case had been heard, to declare what
they honestly thought of the defendant, so these letters of recommendation were
treated as fulfilling the function of testimonials. Why this historical prologue? To
prevent you from thinking I have had too little consideration for your dignitas and
auctoritas in recommending my most intimate friend Sulpicius Cornelianus, who
will very shortly plead his cause in your court. But, as I have said, I venture to praise
my friend to you, appealing to time-honored precedent. He is an industrious man
and energetic . . . . Therefore I entreat you as earnestly as I possibly can to look
with favor on this very dear friend of mine in his suit.

Symmachus to Flavius Eusignius:

Even if family and friendship should be silent and make no demands, public
pressure would be unable to stand in the way of our just desire. In fact, nothing
sits so naturally with justice as to nullify the effect of the trickery of the man
who has duped my kinswoman, a woman most honourable and praiseworthy, and
to prevent him from helping himself to the revenues of another, leaning on the
support of his provincial patrons [patronorum]. Their arbitrary behavior has forced
a lady who is shy of litigation to flee to the tribunal of the praetorian prefect. The
matter at issue is this: a few months ago, my kinswoman let out to this man for a
knockdown price some barns situated in the territory of Aquileia and unfamiliar
to her because so distant. He damaged them so much that in next to no time he
had reduced the place, it is said, to ruins. She met him several times to get him
to give up his lease, but on each occasion she was blocked by the contempt of
this man and the interventions of certain individuals. Now she needs help, and
beseeches your indulgence [beneficium] twice over: first to have this most crooked
of tenants dislodged from the place; and second, to force him, once the damages
are assessed, to pay a fair price for returning the place to its former condition. All
we are asking for is equity [aequa]. Hence I trust that you will look benignly on
the interests of my kinswoman.

 The only surprise is Fronto’s labored introduction. See Fronto, Ep. ad amicos .. Claudius Severus
was twice consul and Marcus Aurelius’s son-in-law; Symm. Ep. . (). Eusignius was praeto-
rian prefect of Italy and Illyricum. For other examples, and discussion, see Harries : –.
Symmachus’s letters of recommendation, and those of other late Roman aristocrats, include some
written on behalf of individual peasants who work for them and are represented as clients. See, e.g.,
Symm. Ep. .; .; .. This phenomenon of vertical patronage in the countryside is better
attested in the late empire than in earlier periods. See Grey .
 Roman patronage 

The attitude of our sources to rival channels of influence and alternative

methods of obtaining promotion is uniform and predictable. Symmachus
dismisses the patrons who are pulling strings on behalf of the tenant of
his kinswoman as “provincial” and their behavior as “arbitrary” (inpoten-
tia). Libanius, in a well-known oration, complains bitterly about a powerful
military man (of name and status unspecified) who has undercut Libanius’s
own operations as patron, preventing him from collecting rents from cer-
tain of his tenants and from seeking justice in the courts of the governor.
To Ammianus, the use of influence (suffragium) is quite above board if exer-
cised by the right people in the interests of the best candidates. He comes
down hard on the sale of offices and the promotion of the undeserving.
This is his judgment on Valentinian I: “He was most cautious in bestow-
ing high official positions: under his rule no money-changer governed a
province, no office was ever sold, except at the beginning of his reign, a
time when it is usual for some crimes to be committed with impunity
through reliance on the distractions of a new ruler.”
Libanius was scathing at the expense of “some fullers’ sons with houses
more glorious than palaces” who had risen to high positions in the admin-
istration through the influence of court secretaries, notarii (hypographeis).
The context is his discussion of Julian’s purge of the court, of which he
heartily approved. His real target in this section are the notarii themselves,
whom he regarded as upstarts, hired for menial jobs but wielding power in
a scandalous way in order to feather their own nests. A little later Libanius
fulminates against the imperial couriers, agentes in rebus (or curiosi), who
had mushroomed under Constantius. These “underlings” had paid hard
cash to secure the privilege of “spying” for the emperor, and proceeded to
behave like “hucksters,” leaving “a trail of pillage and looting” through the
empire.
This is not a case of an odd individual who is not “one of ours” securing
a post: whole classes of high officials are here judged by Libanius to be

 Libanius, Or. . See Liebeschuetz : –; Carrié ; Garnsey and Woolf . Other
references to so-called patrocinium: CTh ..–.
 Amm. ... One man who slipped through the net in the early days of Valentinian I was the
baker-turned-governor Terentius, whose arrival in his province was prodigiously anticipated by the
continuous braying of a donkey. See Amm. .., with Matthews : .
 Lib. Or. .. Libanius speaks of huparchoi, translated by Norman as “prefects”; he cites Dulcitius,
one-time proconsul of Asia; see Or. .; cf. also Or. .ff.
 Lib. Or. . says that the corps was reduced by Julian to seventeen but had swollen to ten thousand
by  – a “manifest exaggeration” according to Jones (a: ). How, one wonders, would
Libanius have reacted to the conversion of two agentes in rebus on reading the Life of St. Antony?
See Aug. Conf. ... On the other hand, Libanius intervened on behalf of individual notarii and
agentes in rebus with whose families he had connections. See Liebeschuetz : .
 peter garnsey

beyond the pale, their status, job, and manner of acquiring it all utterly
unacceptable. The matter is the more intriguing in as much as one of
Libanius’s charges against the agentes in rebus is that that they were people
“who had robbed their own cities of their services, having fled from the
town councils and the customary civic duties.” They were curiales, actual
or potential, in other words. And if old-style patronage was good at any-
thing, it was at securing the advancement of ambitious curiales. Old-style
patronage – or at least, old-style patrons – would seem to have slipped up
badly.
The curial class had long been a major source of recruitment into the
governing classes of Rome. The extraordinary growth of the bureaucracy –
from a few hundred paid officials under the principate to more than thirty
thousand in the late empire – required that this source be drawn upon
more heavily than ever. In fact the very size of the bureaucracy – plus
the growth of the ecclesiastical hierarchy and the regular replenishment
of the military establishment – points to the successful flight of many
curiales. “Flight” is an appropriate word because, at the same time as old
avenues of advancement were expanding and new avenues were opening
up for curiales of ambition and wealth, the same class came under intense
pressure from the central government to stay put and perform its traditional
role of supervising the extraction of taxes from the provinces (and making
up for any shortfall).
In such circumstances the system of patronage was bound to evolve.
Patronal competition had always existed, but it reached a new level in
the late empire. There were now around six thousand high officials in a
position to act as broker/patrons, thousands of short-term posts to be filled
and refilled, and a much larger pool of interested candidates. Existing
channels remained open and business was resumed after the shocks and
uncertainties of the third century. But the manning and reproduction of an
expanded bureaucracy necessitated an expansion of the patronage network
itself.
Patronage did not have it all its own way. It remained essential to the
recruitment process – for the great majority of new officials were no more
skilled than imperial administrators had ever been: “In the absence of
 Lib. Or. .. However, Libanius actively supported select individuals, including his son Cimon,
who sought to evade curial obligations. See Pack .
 Jones :  (thirty thousand); cf. MacMullen :  (thirty thousand to thirty-five thousand);
Heather : – (twenty-three thousand); cf. Garnsey and Humfress : , also for the early
empire.
 The many laws assembled in CTh . bear witness to this continuing concern of the government.
 Heather : ca. six thousand “very good jobs”; followed by Smith .
 Roman patronage 

a system of examinations, patronage in the field of appointments was

inevitable.” But an incipient bureaucratic ethos favoring orderly promo-
tion by seniority and merit was at odds with the accelerated promotions
and leapfrogging that were the hallmarks of the patronage system. In fact,
the progress of this new value system was held back, and undermined, by
the tendency of the top bureaucrats themselves to favor their own families
and friends.
Then, money increasingly came into the picture. It is relevant that
officials were paid less than their early imperial counterparts had been –
otherwise the government would have been unable to afford to employ so
many more of them – and were permitted, or expected, to make up the
deficit by charging fees for sundry services. That said, the legal evidence
shows that the attitude of governments to the sale of office was ambigu-
ous and inconsistent. The earliest law, one of Constantine, prohibits the
practice; later emperors indirectly condoned it, banned it, permitted it
in specified areas, cashed in on it, and, finally, as in the case of Justinian,
regulated it; and indeed it seems that there was no detailed list of charges
for offices until Justinian’s of . Justinian claims to have pegged charges
at levels that were lower than before, even if this reduced revenues to
individuals and to the state.
How did practitioners of the old value system cope with the new? A
flat cash payment might seem to be at odds with continuing, reciprocal
exchange. In practice the two systems might well work together. For a
buyer to be in a position to negotiate with a seller he would need to have
secured an introduction. Money and connections might well be produc-
tive in combination. Or alternatively, the two systems might work more
or less independently at different levels. John Lydus in the sixth century
owed his entry into the bureaucracy and accelerated promotion (not to
 Liebeschuetz : .
 See, e.g., CTh .., : within the corps of the agentes in rebus the hereditary principle is
sanctioned at the highest level.
 See Kolias ; Liebs ; Schuller  and ; C. Kelly ; and Marcone .
 CJ .., of ? : venale suffragium, the acquisition of the status of perfectissimus by inappropriate
individuals (slaves, curiales, ex-bakers, etc.) is condemned.
 See, e.g., CTh .., ; .., ; .; CJ .., ; Malchus . (cited in Jones a: );
Just. Nov. . The fact that the background to individual laws is usually unknown makes the task
of reconstructing the attitude(s) of governments that much more difficult. It is, of course, unclear
that any of the laws in question secured their end.
 It is interesting that costs of litigation (at any rate at the court of the provincial governor) were posted
in public places as early as the s. See Chastagnol ; C. Kelly : ff. (Timgad, North
Africa). Recent discussions of Justinian’s legislation include Bonini ; C. Kelly : –; and
Marcone  and . For comparable schedules of charges for episcopal appointments see Just.,
Nov.  ().
 peter garnsey

mention an advantageous marriage) to his fellow countryman Zoticus, but

he may be supposed to have paid for his various upgradings, as custom
or rule demanded. Finally, the two systems might be thought of as alter-
native strategies, the circumstances governing which of the two would be
employed in a particular case. A man of some station within the congre-
gation of Bishop Amphilochus of Iconium sought exemption from curial
obligations. Basil of Caesarea was called in to help. The way ahead, he
wrote to Amphilochus, was to “set about asking this favor from each of
our friends in power, either as a gift or for some moderate price, however the
Lord may help us forward.” The letter nicely illustrates the coexistence
and intersection of different styles of patronage, the capacity of patronage
to adapt itself to a world undergoing profound changes in politics and reli-
gion, and, finally, the indispensability of the institution. From “Romulus”
to Justinian patronage was a defining characteristic of what it meant to be
Roman.
 Basil, Ep. . (); Kopocek : . Patronage within the church has received short shrift in
this chapter. Much of the literature, however, is concerned with the bishop as community leader
and dispenser of charity and other benefits. See Lepelley ; Brown ; Finn ; Cooper
and Hillner ; and Soraci . Finn (: –) is careful to distinguish this role from the
private patronage that is the subject of this paper. Another topic neglected here is rural patronage.
See Garnsey and Woolf ; Grey . Patronage in family relations is explored by C. Sogno in
this volume (chap. ).
 chapter 3

Roman matchmaking
Cristiana Sogno

“From ‘Romulus’ to Justinian patronage was a defining characteristic of

what it meant to be Roman.” Far from disappearing in the late empire,
patronage remained an essential tool for the recruitment of imperial admin-
istrators and happily coexisted (and, indeed, flourished) alongside the new
system of the sale of offices. This has been argued by Peter Garnsey in
this volume with special reference to socio-political advancement and the
exercise of patronage in a legal context, and with the use in particular
of letters of recommendation. My contribution will illustrate a related,
but less studied aspect of patronage, namely, that concerning the activity
of powerful patrons in arranging socially and economically advantageous
matches on behalf of their protégés. The core evidence for my chapter is
also provided by letters. In this respect, the letters of Symmachus are a
treasure trove of first-hand information and a precious document of the
continued importance of betrothal among the Roman upper classes in late
antiquity. Furthermore, a comparison with the letters of Pliny and August-
ine respectively will show the remarkable continuity of practices and ideals
in both the arrangements of marriages and the exercise of this aspect of
patronage.
The letters that I will discuss are best understood as a subgenre of the
more familiar letter of recommendation, the epistula commendaticia, with
which they share the same rhetorical strategies. Just as in the case of the let-
ters of recommendation, great importance is attached to the moral qualities
and social standing of the prospective husband, which make him suitable

I would like to thank Jen Ebbeler, Judith Evans-Grubbs, Peter Garnsey, Zina Giannopoulou, Noel
Lenski, Scott McGill, Josiah Osgood, Karin Schlapbach, Edward Watts, Gordon Williams, and the
anonymous press reader for reading and commenting on earlier drafts of this paper. Their insightful
suggestions have helped me focus the argument and polish the exposition, but any infelicity in the
text is mine alone. I also would like to thank John Matthews, who first introduced me to the world
of late antiquity and sparked my enduring interest in the writings of Symmachus. This paper is
dedicated to him, mentor and friend.
 Garnsey p. .


 cristiana sogno

for the match, but almost invariably the recommendation calls attention
also to the moral authority and social prestige of the recommender. “If I
seem to you a recommender of some consequence” – Symmachus argues –
“evaluate the man on the basis of my assessment and deign to respond to
one, whom I am delighted to count among my dear ones, with the promise
of marriage to your daughter.”

age and contractual aspect of roman betrothals

A Roman betrothal can be defined as a contract in due form entered into by
the male heads (patresfamilias) of the families of the prospective bride and
groom respectively and was the product of extensive negotiations between
the two families. The relatively high age on average of Roman men at the
time of their first marriage (twenty-five to thirty years old) seems to indicate
that most men were legally independent (sui iuris) when they married
and, therefore, could conduct the negotiations on their own. But Roman
mothers, if alive, could exert considerable influence in the arrangements of
their sons’ marriage, as illustrated by the example of Augustine’s mother,
Monica.
The correspondence of Symmachus offers interesting insight into the
question of the age of Roman men at the time of marriage. A letter written
to Stilicho refers to the wedding of Symmachus’s son, Memmius, that
took place presumably in , not long before Symmachus’s death (ca.
). Memmius was probably only sixteen or seventeen years old when
he married Galla, the niece of his brother-in-law and granddaughter of
Nicomachus Flavianus senior. On account of Memmius’s young age, it
is safe to assume that it was his father who made both the betrothal and
marriage arrangements on his behalf. In contrast with this scenario, all

 For the ideal qualities required of a prospective husband, see Treggiari a.
 The letter is addressed to the senator Carterius on behalf of Auxentius, also a member of the
senatorial aristocracy and young friend of Symmachus (see below).
 Symm. Ep. ., quare si tibi videor non inanis adsertor, perpende hominem de testimonio meo et ei,
quem pignoribus meis libenter adnumero, dignare filiae tuae respondere coniugium.
 Evans Grubbs : .
 For the substantially lower average age of women at the time of marriage that ranged between early
(in the case of aristocratic women) and late teens, see Saller ; on the average age of men at the
time of marriage, see Shaw b.
 On the practical limits that the demographic realities of high mortality and late male marriage
placed on the extent of the father’s authority, see Saller : –.
 Augustine, Confessions . and ; see Shaw a and Shanzer .  Symm. Ep. ...
 The exact date of Memmius’s birth is unknown but can be dated speculatively between  and
. See Cecconi : – and –.
 It is certainly Memmius’s father who sends the sportula to Stilicho (see Ep. .); cf. also Ep. .,
where Symmachus complains that Florentinus had failed to inform him of the marriage of his
 Roman matchmaking 

the other letters of Symmachus that deal with betrothal arrangements were
written on behalf of the prospective grooms and addressed to the father
of the prospective bride. The fact that none of the letters mention the
fathers of the prospective husbands seems to indicate that the latter were
adult, independent males (sui iuris) who conducted the negotiations on
their own.
A quick comparison with the other two sources (Pliny’s and Augustine’s
letters) that will be discussed in this chapter reconfirms that prospective
grooms conducted the negotiations on their own only if they were sui iuris.
That was certainly not the case of the (presumably young) man mentioned
in Augustine’s letters (Epp. –, discussed below), which indicate that the
man’s father was in charge of making arrangements for his son’s marriage
and had personally sought the help of a bishop to find a suitable bride
for his son. A letter of Pliny (Ep. ., discussed in the next section) offers
some interesting insight into the question of age at the time of marriage,
but poses more questions than it answers. When the letter was written,
the prospective groom (Minicius Acilianus) was approximately thirty-five
years old, and his father was still alive (Ep. .. and ). It is reasonable to
suppose that Acilianus was actively involved in the negotiations concerning
his own marriage, but, if he was under patria potestas, he would have had
to obtain his father’s consent. According to Roman law, only a legally
independent individual (sui iuris) was competent to enter into contracts,
and only contracts completed in due form were valid. The same applies to
betrothals, which thus fall under the regulations of contract. Furthermore,
the contractual aspect of a Roman betrothal is evidenced by the custom
of drawing up the so-called tabulae nuptiales or matrimoniales – a custom
that can be found also among Christians.

matchmaking: practice and ideals

The practice of recommending the best candidate for a match was certainly
not new or peculiar to Symmachus, but finds a precedent worthy of

son Minervius or send the sportula (although Symmachus does not complain openly about it and
pretends not to care).
 Symm. Ep. ., ; .; ., , , , , .
 Sherwin-White : : “as an undistinguished praetorius he must be  to Pliny’s .”
 For fuller discussion of the role played by age in this letter, see below.
 For the question of paternal consent, see Evans Grubbs : –.
 Evans Grubbs : –; for Christian practice, at least in North Africa, see Hunter .
 Matchmakers were used to make the first discreet inquiries or could be asked to suggest suitable
candidates. Whereas in the east it was customary to rely on elderly women who seem to have
worked as professionals, in the western empire matchmaking was an “amateur activity conducted
by friends.” See Noy : –; cf. Arjava : .
 cristiana sogno

notice in Pliny’s letters. Of the three letters that are devoted to the topic
of matchmaking in the carefully edited correspondence of Pliny, the first
one (Ep. .) is especially interesting for the idealized portrayal of the
groom that it offers and for the ideals of marriage that it addresses.
Unlike Symmachus, who wrote his recommendations upon the request of
the prospective husband, Pliny wrote this letter supposedly in response
to the request of a friend (Junius Mauricus) to help him find a suitable
husband for his orphaned niece. As Pliny’s letter makes clear, a carefully
arranged marriage, based upon the equality of the moral qualities and social
standing of the partners, would ensure the continuity of the family and
produce heirs morally and socially worthy of their ancestors (Ep. ..).
And – Pliny argues – the moral worthiness of the prospective groom is
shown by his relationship with Pliny and especially by Acilianus’s desire
to be “molded and instructed” by Pliny (Ep. ..: formari a me et institui
cupit). This line of argument is perfectly consistent with the rhetoric of
the commendaticia: the commendatus is presented as morally worthy on
account of his relationship with the commendator. Even though Pliny is not
writing a letter of recommendation upon request of the prospective groom
(Minicius Acilianus), it is interesting to notice that he deliberately assumes
the role of Acilianus’s patron in the letter by stressing the (insignificant) age
difference between himself and Acilianus (est enim minor pauculis annis).
The insistence upon the small difference in age is used to emphasize
Acilianus’s appropriately deferential attitude toward Pliny and introduces
the element of inequality that is characteristic of patronage relationships,
as opposed to friendship.

 Plin. Ep. .; ., .

 See Treggiari ; cf. Treggiari a and b: –. Cf. also the still fundamental study of
Roman ideals of marriage by Williams ().
 Plin. Ep. . bears only a superficial similarity to Symmachus’s letters of recommendation: Pliny’s
endorsement of the future husband (Fuscus Salinator) follows the betrothal between Fuscus Salinator
and the daughter of Julius Servianus. One wonders whether the point of the letter is to advertise
Pliny’s connection with a family related to the emperor, given the fact that the future bride happened
to be Hadrian’s niece. For the prosopography, see Sherwin-White : –.
 The context suggests that the niece of Junius Mauricus was very young at the time that she was
embarking on her first marriage. Because of the death of her father, she was no longer under patria
potestas, and had she been above the age of puberty, she could have entered a marriage of her own
volition but would likely have required the permission of a tutor. See Treggiari a: .
 Disparity of birth between husband and wife was to be avoided, as shown by the stereotype of the
rich domineering wife, a stock character in Roman comedy and a favorite target of Latin satire.
In general, marrying someone socially inferior was deemed disreputable, and a stain attached to
children born of such marriages (see Hor. Sat. ..; Tac. Ann. ..). See Treggiari b: –.
 On this point, see Evans Grubbs : –.
 For the definition of patronage, see Saller : , and Garnsey’s comments in this volume.
 Roman matchmaking 

The importance of the ideal of moral equality between prospective

groom and bride is well exemplified by Pliny’s letter. First, the origo of
Minicius Acilianus is presented as an indirect proof of his moral character.
The fact that he came from Brixia, namely, “that Italian region of ours
that still maintains and preserves a great deal of modesty, sobriety, and
old fashioned simplicity,” indicates that Acilianus possessed those very
qualities. The expression Brixia . . . ex illa nostra Italia need not be taken
as proof that bride and groom shared the same origo. Rather, Pliny’s
point is that both families shared the same moral principles, and that
Junius Mauricus would not find anything in Acilianus’s domus that was
not as acceptable to him as anything in his own.
Next Pliny focuses on the cursus honorum of Acilianus and on his physical
appearance. The latter, he argues, should not be neglected, because a
handsome exterior is the fitting reward for the chastity of girls (Ep. ..:
debet enim hoc castitati puellarum quasi praemium dari). The use of the
word praemium in combination with the idea of bridal chastity (puellarum
castitati) is especially interesting. A law of Constantine () refers to the
praemium pudicitiae, namely the gift or gifts given to the bride before
the wedding in recognition of her virginity, but the practice of rewarding
the bride for her virginity is well attested already in the first two centuries
of the empire. Even though the practice seems to have originated in the
provinces rather than in Italy, Pliny was clearly aware of it, and, in fact,
his remark adds a (Roman?) twist to the provincial practice. Instead of the
customary gifts in gold, the physical attractiveness of the groom becomes
a more fitting recompense for a virgin bride.
One can speculate that Pliny found the custom of a monetary reward
for virginity distasteful. Interestingly, Pliny had no qualms in suggesting
the importance of the groom’s attractiveness as the basis for the attrac-
tion between future husband and wife, and his insistence upon the “rosy
countenance and high color” of Acilianus’s complexion – a clear sign of

 Plin. Ep. ..: Patria est ei Brixia, ex illa nostra Italia quae multum adhuc verecundiae frugalitatis,
atque etiam rusticitatis antiquae, retinet ac servat; cf. Plin. Ep. .., where maxima verecundia is
presented as one of the main qualities of Acilianus.
 For the link between provincial origo and old-fashioned Roman morality, cf. Tac. Ann. ., ..
 For doubts concerning the northern Italian origo of the family of the bride, see Syme : .
 Plin. Ep. ..: In summa nihil erit in domo tota, quod non tibi tamquam in tua placeat. For the
meaning of domus as encompassing the maternal and paternal side of the family, see Saller .
 The use of the plural puellarum instead of the singular puellae indicates a more generalized remark.
 CTh .. = CJ ...
 Juvenal refers to it in his notorious sixth satire (.–). For an overview of the practice with full
bibliographical references, see Osgood : –.
 Plin. ..: Est illi facies liberalis, multo sanguine multo rubore suffusa.
 cristiana sogno

health and vigor – leaves little doubt as to the nature of his bride’s reward.
In contrast with this “erotic” openness, Pliny shows some coyness in men-
tioning the wealth of the future husband’s father. Apparently Acilianus
stood to inherit a fortune, but when thinking of the people for whom he
was seeking a son-in-law, Pliny thought that he should “say nothing about
wealth.” The Junii (the bride’s family) might indeed have been “exces-
sively wealthy” to care about money, and “as philosophers” they might have
“despised money.” But, more in general, Pliny seems to have felt that any
extended talk about something as vulgar as money might offend the refined
sensibility of his addressee, who might not have shared the same origo as
Pliny and the groom, but who certainly shared the same moral values.
Both Pliny and Symmachus can be taken as good representatives of
the educated and wealthy elites of the (western) empire under the prin-
cipate and in late antiquity respectively, and the comparison between the
letters of Pliny and Symmachus pertaining to betrothal highlights some
important differences within the continuity of marital practices and ideals.
Symmachus’s letters reflect the continued importance of equality between
husband and wife, but with an emphasis on socioeconomic equality. In rec-
ommending the son of Severus, a vir honestus, to an anonymous recipient,
Symmachus observes that “the equality of the individuals (involved) and
their similarity in social standing and means” should facilitate Severus’s
request to marry his son to the daughter of the recipient.
It is not that the moral fitness and personal qualities of the husband
have ceased to be important. In fact, Symmachus duly mentions in his
recommendations the “praiseworthy lifestyle” of his friend Fulvius (Ep.
..: laudabiles eius vitae artes) and the “natural qualities” (Ep. ..: orna-
menta naturae) of his dear Auxentius. Moreover, his willingness to act as
guarantor of these young men should be evidence enough of their moral
worth. However, equal and more explicit emphasis is put on the socio-
economic advantages deriving from marriage. Symmachus is less shy than
Pliny in talking about money. In a letter of recommendation on behalf
of a certain Fulvius, a member of the senatorial aristocracy, Symmachus
 Plin. Ep. ..: Nescio an adiciam esse patri eius amplas facultates. Nam cum imaginor vos quibus
quaerimus generum, silendum de facultatibus puto.
 Sherwin-White : .
 As Syme observes, whatever their provenance, the Junii shared the same moral principles outlined
in Pliny’s letter. Syme : .
 Symm. Ep. .: Cuius rei impetrationem facilem fore personarum promittit aequatio et honestate et
possibilitate congruitas.
 Cf., e.g., Symm. Ep. ..: Vadem me tibi in omnia spondeo, quae solet parentum sollicitudo trutinare.
One can safely assume that moral character and conduct are chief among the things that anxious
parents put on a scale when evaluating the suitability of future sons-in-law.
 Roman matchmaking 

does not hesitate to point out that Fulvius is not inferior in ancestry to
the woman he wishes to marry and is “perhaps better off financially” than
she is. Wealth is considered as crucial a factor as mores in determining
the “honorableness” (honestas) and, therefore, suitability of a prospective
husband.
Greed would no doubt be unacceptable, and the story of Maximinus
related by Ammianus clearly exemplifies this point. Maximinus’s avarice
and the means he used for achieving his ends were extremely objectionable,
but his proactive attitude in securing a remunerative match for his son is
far from unique. In fact, the utilitarian view of marriage extends also to
social relationships, and “marriage continued to be viewed as a social and
political act uniting not only a man and a woman but, more to the point,
two aristocratic men and their family.” Among the great advantages
that marriage offered was the opportunity of expanding one’s circle of
connections. As Symmachus points out with an elegant praeteritio, what
made Fulvius appealing as a son-in-law besides his pedigree, considerable
wealth, and personal qualities were his friendships with the people who
mattered (ex bonorum amicitiis), because they could ensure a brilliant future
for him. The importance of a well-connected relation could not have
been lost on the recipient of the letter on behalf of Fulvius. Nicomachus
Flavianus junior, Symmachus’s own son-in-law, had much profited from the
connections of his father-in-law, who managed to save his political career
and family assets after the disgrace and suicide of his father, Nicomachus
Flavianus senior.
But what do the differences between the letters of Pliny and those of
Symmachus mean? More specifically, what does the greater emphasis placed
by Symmachus on utility and the material advantages of marriage indicate?
Although there is no simple solution, the answer to these questions should
be sought in the different nature of the correspondence of Pliny and
 Symm. Ep. ..: iuvenem nec genere minor et re fortassis uberior.
 Symm. Ep. ..: Licet noverim futuros generos moribus aestimari, attamen huic post ornamenta naturae
etiam census ad honestatem redundat.
 Amm. ... According to Ammianus, the cruel minister of Valentinian was not only bloodthirsty
but also outrageously greedy (aviditate nimia flagrans). Not content with getting hold of half of
the inheritance of the deceased Victorinus, Maximinus managed to bully Victorinus’s widow into
consenting to the marriage of his son with her stepdaughter, in order “not to miss out on the great
opportunity of profiting from the rich patrimony that was available to him.” On Victorinus and
his widow Anepsia, see Lizzi Testa : –.
 In Maximinus’s case, this moneygrubbing and effort to climb through marriage were especially
objectionable because of his lowly provincial background.
 See Salzman : –.
 Symm. Ep. ..: ex bonorum amicitiis spes secundas inter sponsalia ornamenta non numero.
 See Matthews : –; Sogno : –.
 cristiana sogno

Symmachus. Anyone who has worked on ancient letter collections knows

how arbitrary any dichotomy between “real” and “literary” or “private”
and “public” letters is. Rather than divided into rigidly distinct categories,
letters can more profitably be arranged along a spectrum from “real” (e.g.,
letters found in papyri) to “fictional” (e.g. Ovid’s Heroides). Pliny’s and
Symmachus’s letters considered here fall in between these two extremes:
they are both real letters of recommendation and literary artifacts that their
authors intended to publish. And yet the fact that the entire correspondence
of Pliny was carefully edited and, perhaps, even written with a view to
publication seems to suggest that his letter should be placed closer to the
“literary” end of the spectrum. That does not mean that Pliny’s letter is
“fictional” or that Junius Mauricus did not ask for Pliny’s help in finding
a husband for his niece. Rather, the purpose of Pliny’s letter is as much
to recommend Minicius Acilianus as it is to sketch a portrait of the ideal
husband, and Pliny seems to acknowledge this when he remarks that
he would have had to look for a long time before he found a suitable
candidate, had not the ideal one been at hand.
Notwithstanding the fact that Symmachus, too, had intended to publish
his correspondence and was probably working on that project before he
died, his letters of recommendation on behalf of prospective husbands
seem closer to the “real” end of the spectrum. These letters are brief and
matter-of-fact, because “truth does not need the help of many words” and
only favors that are hard to grant require long periphrasis. Symmachus
was clearly not interested in depicting the ideal husband but was bent on
persuading the recipient that his candidate was ideal for the match. In

 As Nicole Méthy has recently argued in her meticulous study of Pliny’s correspondence, Pliny’s
letters “present or depict characters who incarnate to different extents an ideal man.” See Méthy
: .
 For the similarities in theme and tone between the depiction of marital mores in Pliny’s letters and
in Plutarch’s Advice to the Bride and Groom, see Evans Grubbs : , –.
 Plin. Ep. .., Qui (i.e., iuvenis ex quo nasci nepotes Aruleno Rustico deceat) quidem diu quaerendus
fuisset, nisi paratus et quasi provisus esset Minicius Acilianus.
 Traces of Symmachus’s editorial activity might be detected in the first seven books of the correspon-
dence (Sogno : –), and Symmachus might have published his first book of letters during his
lifetime, as Michele Salzman suggests in her forthcoming commentary on book  of Symmachus’s
correspondence. It is likely that Symmachus’s son Memmius published the correspondence of his
father in the immediate aftermath of his death; however, the complete disarray of books  and 
suggests that these two books were published later by someone who had access to the family archive
(cf. Roda : –).
 It is perhaps not by accident that most of the letters about matchmaking survive in book , one of
the “messy” books that Symmachus did not edit.
 Symm. Ep. ..: Possem de eo copiosius loqui, sed non vult veritas verbis iuvari.
 Symm. Ep. ..: Praefato opus est, si ardua postulentur; pronis ac facilibus admoveri ambitum non
oportet.
 Roman matchmaking 

real life parents tended to be as anxious about the moral character of the
prospective husband as they were concerned about his finances, and that
is the information that the letters convey. The change that we perceive
between Pliny’s and Symmachus’s letters about matchmaking seems due
less to a transformation in the views about and attitude toward marriage
than to a shift in the focus and purpose of the letter writers.

matchmakers as patrons
Another important difference between the letters of Symmachus and Pliny
might explain their different nature. As was mentioned previously in this
essay, both Symmachus and Pliny wrote in order to recommend a suitable
husband. But, whereas Pliny wrote in response to a request on the part of
the bride’s family (Ep. .), Symmachus wrote upon request of either the
groom himself (Ep. ., .) or the father of a groom who was not sui iuris
(Ep. .). On the whole, Symmachus’s letters offer a more aggressive and
competitive picture of matchmaking, especially when compared with the
idealized picture provided by Pliny’s letters. Symmachus the matchmaker
was a powerful patron to whom prospective husbands applied in order to
conclude a match. In this capacity he acted just as he would have if he
were called upon to introduce a petitioner to an important man, obtain
a favorable verdict in court, or arrange promotion to a particular post.
In fact, a promotion and an advantageous match were not only equally
desirable but interchangeable, as shown by the case of Athanasius, which
will be discussed below.
Such a dynamic might explain the eagerness with which Symmachus
pursued the cause of the people he recommended. “The successful conclu-
sion of this match,” argued Symmachus, “is for me so high a priority that
I will treat as a weighty gift your care in obliging the desire of that young
man.” The success of a recommendation was a clear sign of the power
and prestige of a recommender, and, in the case of a successful match,
 The similarity of the situation described respectively in Pliny’s Ep. . and Symmachus’s Ep. .
(discussed below) is only superficial and should not blind us to the profound differences in the
purpose of the two letters. Both letters are addressed to prospective fathers-in-law (Fuscus Salinator
in Pliny’s letter and an anonymous recipient in Symmachus’s), and they convey the authors’ warmest
congratulations on the choice of the husband. But, whereas Pliny’s letter aims at advertising his
connection with the powerful Fuscus Salinator, Symmachus’s aims at expediting his friend Nicias’s
marriage.
 All of these activities are well illustrated in Garnsey’s contribution.  Symm. Ep. ..
 Symm. Ep. .: Mihi autem tanta est perficiendae coniunctionis huius antiquitas, ut in gravi dono
habiturus sim, si illius votis cura vestra profuerit.
 See Roda : –.
 cristiana sogno

the recommender could count on the gratitude of both the husband and
his father-in-law: “Believe me: when what I desire comes to fruition, you
as the recipient of my request will be grateful to me for my intervention
no less than the person I recommended.” Of course, Symmachus’s remark
was meant to reassure the future father-in-law that he would not regret his
choice and that the match would be mutually satisfactory. But at the same
time the remark reveals what was in it for Symmachus himself.
In order to affect the desired outcome, Symmachus did not hesitate to
remind the addressee about friendship as one more reason to grant the favor
that he was requesting, nor was he above dangling his friendship in front
of the eyes of the recipient of the recommendation as one of the welcome
outcomes of the match. These tactics are especially evident in a letter
written by Symmachus on behalf of his friend, the philosopher Nicias (cum
primis philosophorum numerandus). The letter begins with the expression
of Symmachus’s delight upon discovering that the (anonymous) recipient
had promised his daughter in marriage to Nicias. In fact, the news of the
engagement was so welcome that it made Symmachus desire to establish
a friendship with Nicias’s future father-in-law. What follows, however,
underscores the condescension of the great Roman senator in honoring
the recipient with his friendship. The letter lays emphasis exclusively on
the advantages that will accrue to the father of the bride as a result of the
match: the addressee is praised for his good judgment only because he chose
Nicias as his future son-in-law. Because of this sensible choice all the “good
people” will now want to pursue a friendship with him. Symmachus’s
patronizing attitude reaches its peak toward the end, when he reveals the

 Symm. Ep. ..: Credas velim, cum vota processerint, interventui meo non minorem te gratiam, qui
rogaris, habiturum quam istum quem commendo.
 Nothing else is known about Nicias, who is not otherwise attested (see PLRE I, s.v. “Nicias”).
In all likelihood, the philosopher Nicias was a teacher who taught the children of the senatorial
aristocracy at Rome and, thanks to his profession, was able to forge connections with power-
ful aristocratic families. Other such cases are documented by the letters of Symmachus; see Ep.
.., written by Symmachus on behalf of the philosopher Priscianus (Priscianus frater meus cum
primis philosophorum litteratura et honestate censendus senatu auctore salarii emolumenta consequitur),
and Rel. , concerning the co-optation of the philosopher Celsus into the Roman senate (Celsus
ortus Archetimo patre . . . iuventuti nostrae magisterium bonarum artium pollicetur nullum quaestum
professionis adfectans atque ideo dignus in amplissimum ordinem cooptari).
 Symm. Ep. ..: Desiderium mihi amicitiae tuae frater meus Nicias cum primis philosophorum
numerandus incussit, postquam nobis per epistulam fecit indicium, quod ei filiam tuam fida pactione
desponderis.
 See, in contrast, the deferential tone of letters written by Symmachus with the intent of establishing
a friendship with his social betters, as in the case of Ep. ..
 Symm. Ep. ..: quae res (i.e., fida pactio) testimonio est, sincerum tibi inesse iudicium et propterea
familiaritatem tui ultro omnes bonos debere sectari. adserit enim probabilem mentem talis electio.
 Roman matchmaking 

real purpose of the letter. Using his influence as the stick and “the favor of
many and the prestige of the alliance” as the carrot, Symmachus writes to
spur the addressee to marry off his daughter to Nicias as soon as possible.
To be sure, Symmachus approached matchmaking with the same attitude
with which he approached the writing of commendaticiae. As a letter to
Nicomachus Flavianus senior makes clear, Symmachus was inclined to
accept all requests for letters of recommendation, but he expected the
recipient of the recommendation to make the appropriate distinctions
among the commendati on the basis of the tone of his letters. The same
seems true of the letters written on behalf of prospective husbands. In
recommending the son of Severus, Symmachus made clear to the recipient
(Justus) that he could not refuse a reasonable request, especially after one
that had been repeated so many times. But the lukewarm tone of the
letter, especially when contrasted with the “vehemence” of other letters,
is an indication that Symmachus was performing his duty, but was not
particularly invested in it. Symmachus’s “coolness” is not necessarily a
consequence of the lower social standing of the people involved. Nothing
is known either about Severus or about Justus, but the fact that Severus is
qualified simply as an “honorable man” (vir honestus) seems to indicate that
neither he nor Justus was a member of the Roman senatorial aristocracy.
And yet Symmachus did not reserve his warmest endorsement exclusively
for members of the senatorial aristocracy, such as Fulvius or Auxentius.
In fact, he seemed equally keen on finding a wife for the philosopher Nicias.

 Symm. Ep. ..: quare optimae voluntati tuae calcar admoveo teque hortor ut quamprimum lau-
datissimum generum diis iuvantibus sortiaris; cuius accessio tibi et conciliabit gratiam plurimorum et
decus adfinitatis adiciet.
 See also Symm. Ep. ., written by Symmachus to encourage the (anonymous) recipient to expedite
the marriage between Symmachus’s friend Titianus and the daughter of the addressee. This kind
of request might have been prompted by a desire to ensure that the marriage took place. For the
breaking up of engagements, see Symm. Ep. ., ., ., and commentary below.
 Symm. Ep. .: Multi a me conciliationem litterariam poposcerunt, se quorundam meritis, aliorum
precibus parem gratiam dedi. Nec tamen vereor, ne tibi civium ignota distinctio sit iudicium de his
commune recolenti.
 Cf. Symm. Ep. .: Voto ac desiderio suo me adiutorem Severus vir honestus adscivit; cuius cum
petitionem probabilem iudicassem, totiens flagitatas litteras negare non debui. The mention in the
letter of Severus’s repeated nagging (totiens flagitans) makes Symmachus’s initial reluctance to assist
all too clear!
 Had Justus been a senator, Symmachus would not have regarded the request of a match as probabilis
on account of considerations of social equality discussed above.
 Nothing else is known about Fulvius, but the fact that he was hoping to marry the sister of Gabinius
Barbarus Pompeianus (the unfortunate urban prefect who was later torn to pieces during a riot
of the Roman plebs) and vaunted a family as good as hers indicates that he was a member of the
senatorial aristocracy. See PLRE II, s.v. “Pompeianus” ().
 For the career of Auxentius’s father-in-law Carterius, see PLRE I, s.v. “Carterius” ().
 cristiana sogno

As for the question of Symmachus’s success as a matchmaker, there

is at least one positive piece of evidence that a match that he arranged
came to fruition: it is a letter, in which Symmachus expressed his gratitude
to Carterius for trusting his assessment of Auxentius and for accepting
the young man as his future son-in-law. The betrothal of Auxentius to
Carterius’s daughter is seen as strengthening the ties of friendship between
Symmachus and Carterius.

senators and bishops

Even though Augustine was apparently reluctant to involve himself in the
arrangement of marriages, a small cluster of his letters, written after
he became bishop of Hippo, offers an interesting parallel to Symmachus’s
activity insofar as it presents the Christian bishop Benenatus acting as a
matchmaker. Bishop Benenatus was operating on behalf of the son of
Rusticus, probably a member of the local aristocracy, though not yet a
Christian. The tangled picture that emerges from Augustine’s letters can
be summarized as follows. The girl to whom Rusticus wished to marry
his son was a ward of the church. At the time she was apparently still
too young to marry but old enough to speak and express her desire to
become a consecrated virgin. Her father was dead, whereas her mother
was still alive, but for reasons that are not explained in the letter she was
momentarily out of the picture. The girl had a maternal aunt married to

 Cf. Symm. Ep. .: Gratum habeo de testimonio meo non esse dubitatum receptumque esse in familiam
tuam Auxentium filium nostrum lectissimum iuvenem in meo aere duco. Quo nomine amicitias nostra
in artiora fidei iura convenit.
 Possidius, Vita Augustini, .–. Augustine’s reticence is attributed to his fear “lest that spouses
quarrel and curse the one who had brought them together.” Cf. Hunter : .
 Aug. Epp. –.
 Aug. Ep.  (addressed to his fellow bishop Benenatus): Et quoniam audivi quod de negotio illo
(i.e., the marrying of a ward of the church with a paganus) transigere cogitas, si verum est (quod
miror, si verum est), nosti quemadmodum debeas episcopali paternitate catholicae Ecclesiae providere;
ut non cum quolibet id agas, si tamen, ut dixi, verum est quod audivi, sed potius cum domo catholica;
cf.  (addressed to Rusticus, the father of the prospective husband): tradi a nobis christianam nisi
christiano non posse; nihil tamen mihi tale de filio tuo, quem adhuc paganum audio, promittere voluisti.
 See PLRE, s.v. “Rusticus” ().  For the meaning of the word, see O’Donnell .
 Aug. Ep. : Puella de qua mihi scripsit Sanctitas tua, in ea voluntate est, ut si aetas ei iam matura
esset, nulli in nuptiis conveniret. In ea vero aetate est, ut si voluntatem nubendi haberet, nulli adhuc
dari vel promitti deberet. Augustine’s words seem to imply that the girl is not yet twelve years old.
For the age of girls at marriage, see Arjava :  n.  with bibliographical references.
 Aug. Ep. : Fortassis enim quae nunc non apparet, apparebit et mater, cuius voluntatem in tradenda
filia, omnibus, ut arbitror, natura praeponit. Augustine underscores the natural prerogative of the
mother in choosing a husband for her daughter when the father is no longer alive. The situation
described is puzzling: Where was the mother? Did she run away and abandon the girl, or did she
 Roman matchmaking 

a certain Felix, also a local notable, whom Augustine had contacted and
informed about Rusticus’s and Benenatus’s plan.
What emerges clearly from this group of letters is the struggle between
the two bishops, Augustine and Benenatus, with regard to the girl’s fate.
Augustine did not state his preference that the girl remain unmarried,
although he conceded that she was too young to make that decision.
More important for the purpose of this essay is Benenatus’s role as
matchmaker. Benenatus’s case offers a clear example of a bishop appro-
priating a role that had traditionally been played by members of the local
aristocracy. What is especially interesting is his disregard for the question
of religious difference in contrast with Augustine’s firm opposition to it.
Benenatus’s attitude toward matchmaking and marriage is consistent with
the traditional Roman attitude. Even though we do not have his letters,
Benenatus’s position parallels that of Symmachus, and the two letters of
Augustine addressed to him suggest that he pursued the match with the
same determination and aggressiveness displayed by Symmachus.
It is tempting to see in a law promulgated in  (CTh ..) a reaction
against the aggressiveness of matchmakers. Although the law addressed
the remarrying of (young) widows in particular, it reconfirmed the author-
ity of the father (patria potestas) or of family members (propinqui), if the
father was dead, against the undue influence of intermediaries (sequestres
et interpretes) in the arrangement of a new marriage. The law targets
simply put the girl under the care of the church when she remarried? Did Augustine intend to find
the mother?
 See PLRE, s.v. “Felix” ().
 Aug. Ep. : utrum autem nuptura sit, etsi illud quod in ore habet (i.e., her desire not to marry) magis
optamus, nunc tamen ignoramus; quia in his annis est, ut et quod se dicit velle esse sanctimonialem,
iocus sit potius garrientis, quam sponsio profitentis.
 For the role of bishops as patrons, see Rapp : –.
 For the question of mixed marriages and that of Christian objections to marriage with pagans, see
Cameron : –; and Salzman : . One wonders whether in this context Augustine’s
opposition to marriage on account of religious difference is not a pretext to prevent the girl from
ever marrying, which Augustine considers as the preferable outcome.
 Aug. Ep.  seems to have been written in response to Benenatus’s reply to Aug. Ep. . Benenatus
must have confirmed to Augustine that indeed the rumors that he (Benenatus) was in charge
of arranging the marriage between Rusticus’s son and the girl were true, and he probably asked
Augustine to consent to the marriage, a request that prompted Augustine’s long list of reasons why
he could not.
 The law is mentioned by Arjava (: ), who concludes that “the lawgivers mentioned match-
makers perhaps with some reserve, but the activity itself was not illegal.”
 For commentary on this law, see Evans Grubbs : –. “Sexual morality was apparently of
serious concern to Valentinian” (Lenski a: ), and this law should be read in the context of a
larger effort on the part of Valentinian to clean up sexual improprieties. For Valentinian’s attitude
toward magic and sexual mores, see Lizzi Testa : –.
 Cf. CTh ..: cessent itaque sequestres atque interpretes, taciti nuntii renuntiique corrupti. nuptias
nobiles nemo redimat, nemo sollicitet, sed publice consulatur affinitas, adhibeatur frequentia procerum.
 cristiana sogno

specifically “secret dealings” (taciti nuntii renuntiique corrupti), presumably

between the prospective bride and the intermediaries of the groom. Inter-
estingly, the letter of Augustine to Benenatus hints at the displeasure of
the girl’s maternal uncle for not having been informed about the marriage
plans involving his niece. But one might wonder what the girl’s uncle
thought of Augustine’s intervention.
What Symmachus’s and Augustine’s letters seem to show is that mar-
riageable girls (with property and connections) were a hot commodity.
Young men who desired to advance socially relied on the patronage of
well-connected friends in order to secure an advantageous match and to
bring it to fruition.

breaking the bond of betrothal

Given the number of people involved, the effort spent, and the interests
of all the parties concerned, it is reasonable to suppose that betrothals
were not entered upon lightly and could not be broken easily. Even before
Constantine’s laws, when according to imperial legislation engagements
could be broken by either party without legal repercussions, people did
not break betrothals for trivial reasons. The insouciant attitude toward
betrothal displayed by late republican dynasts and by the Julio-Claudians
should not be taken as typical, even among the urban upper classes.
Repeated rejection could damage the reputation of a man, given the fact
that sponsi were often rejected on moral grounds. As for women, the
consequences of a broken engagement were even more serious. If rejected
on moral grounds, a woman could be prosecuted under the Augustan law of
adultery (Dig. ...). And, even if she escaped prosecution, a woman’s
reputation could be damaged severely on account of a broken engagement.
Apuleius’s Apology offers a chilling example of the malicious rumors that
could arise when a young woman was jilted by her fiancé. In keeping
with the best Roman courtroom tradition, Apuleius had no qualms about
trashing the reputation of his opponent’s daughter by insinuating that
the girl had been circulated among young men in the hope of gaining a

 Aug. Ep. : sed iure amicitiae non importune doluit quod eis nihil inde sit scriptum.
 As Hoffer “half-seriously” suggests, in the early empire “if upper-class children were exposed, it
was the male infants,” because daughters allowed one to make “useful alliances while avoiding the
political threat of building one’s own family dynasty” (Hoffer : ).
 On the role of patrons in matchmaking, see Noy : .
 Treggiari b: –, esp. .  Evans Grubbs : –.
 Apul. Apol. ; cf. Evans-Grubbs : .
 Roman matchmaking 

husband and was rejected by the youth to whom she had been betrothed
“after he had his fill” (post satietatem).
As for the post-Constantinian world, late Roman legislation indicates
that “emperors considered the marriage pact to be a binding contract.”
The same attitude is reflected also in the rulings of local church councils
both in the east and in the west that tried to prevent not only the dissolution
of marriages but also the breaking of the betrothal bond. The tightening
of the rules concerning betrothal agreements seems to have originated
with Constantine. By stating that the party responsible for breaking the
engagement had to return all prenuptial gifts, Constantine’s legislation
must have made the prospect of a breakup even less appealing. And yet,
in all likelihood, the drafting of such a law stemmed from the need to
respond to (frequent) inquiries from either citizens or officials who had
been called to judge lawsuits concerning broken betrothals. This tension is
nicely reflected in Symmachus’s correspondence, which, on the one hand,
shows that people might indeed wish to break an engagement (and did so)
but, on the other hand, highlights the social pressure that was brought to
bear against the breakup.
It is worth noticing that the three letters concerning such cases (Ep. .
and .; Ep. .) show Symmachus once again acting on behalf of the
jilted fiancés, whose interests he wished to protect. A short letter on behalf of
the senator Athanasius illustrates the predicament of Symmachus’s protégé,
who was denied marriage with his fiancée notwithstanding a written agree-
ment and the exchange of gifts. The letter is addressed to Patruinus,
an influential member of the palatine administration and close friend of
Stilicho. The relationship between Symmachus, Athanasius, and Patrui-
nus illustrates the transition from a society based on face-to-face interaction
to a society where relationships were established and maintained, for the
most part, through epistolary exchange. As a second letter of Symmachus
suggests, Athanasius had been introduced to Symmachus by a commendati-
cia of Patruinus. After his arranged marriage fell through, Symmachus
recommended Athanasius back to Patruinus. What happened next is a

 For an illuminating commentary on this episode, see Osgood : –.

 Evans-Grubbs : –, and – on Constantine’s laws.
 See Symm. Ep. .: pactae ei [i.e., Athanasius, v.c.] et nobis arbitris vinctae obligatione pignorum
nuptiae denegantur; for the gift exchange see Roda : , . On the arrhae sponsaliciae, see
Evans Grubbs : –.
 PLRE II, s.v. “Patruinus”; cf. Callu :  n. .
 Symm. Ep. .: Athanasium bonis artibus clarum securius amare coepi, postquam de eo iudicium
tuum conperi.
 cristiana sogno

clear example of the importance of marriage as a means for social advance-

ment. Even though Patruinus was unable to do much about Athanasius’s
first engagement, he apparently secured another and greater match, and
Symmachus politely requested Patruinus to do his best to ensure that
this second opportunity would come to fruition. The language used by
Symmachus in the letter to Patruinus is ambivalent enough that one could
think that Patruinus had promised Athanasius a promotion, as Callu points
out. Evidently, a brilliant match offered the same advantages of career
advancement.
The case of his great friend (amicissimo meo) Herculius, probably an
advocate and a member of the senatorial aristocracy, is equally interesting,
but more nuanced. Symmachus had not been involved in the planning
of the engagement of Herculius but had clearly been called upon to put
some pressure on Julianus, the recalcitrant father-in-law. The delay in
the celebration of the actual wedding had given Herculius cause to fear
that Julianus intended to call the whole thing off, and Herculius did not
waste any time in finding help to prevent a breakup. He seems to have
found a very efficient supporter in Symmachus. Not only did Symmachus
write to Julianus, suggesting that a change of mind on his part would
be an ugly breach of “senatorial loyalty,” but, in case the moral stain
turned out to be an insufficient deterrent, Symmachus asked Nicomachus
Flavianus junior, his own son-in-law, to intervene (Ep. .). Moreover,
the letter to Nicomachus Flavianus junior mentions a third party, a certain
Valentinus (probably also a relation of Symmachus), who had advised
Julianus to follow the course of action that Symmachus wished in regard
to the engagement. Taken together, the two letters offer a neat example
 Symm. Ep. .: Illa pars mihi pronior dictu est, ut ames in eo spem quam dedisti addasquae inchoato
beneficio perfectionem. Quam si aliquis eventus eluserit, effice, oro te, maiora praestando, ut illi prioris
uoti prosit amissio.
 Callu : .
 Symm. Ep. . indicates that he was a clarissimus vir, and that he practiced law at Rome (ea
modestia, ea in foro Romano dignitas); cf. PLRE II, s.v. “Herculius” ().
 The case is mentioned briefly by both Evans Grubbs (: , ) and Arjava (: –).
 The name of the father-in-law is mentioned in another letter addressed to Nicomachus Flavianus
junior (Ep. .). Nothing is known about him apart from the fact that he must have been a
Roman senator too; cf. PLRE I, s.v. “Julianus” (), and Marcone : .
 Symm. Ep. ..: Nolo igitur existimes culpabiliter ad me esse perlatum quod pactas filiae tuae nuptias
cum Herculio amicissimo meo haesitatione suspendas, sed ceteris quos matrimonii de te inpetrati vel
adiutores scit fuisse vel testes me quoque voluit admonuere, ut suspensa diu vota novi hortatoris inpulsus
acceleret.
 Symm. Ep. ..: Maneat igitur, oro te, stabilitas promissorum, nec senatoriam fidem voluntatis
mutatio devenustet.
 See Symm. Ep. . (addressed to Nicomachus Flavianus junior): filius meus Valentinus adseruit se
quoque consultum litteris Iuliani eadem quae volumus de nuptiarum pactione suasisse.
 Roman matchmaking 

of the social pressure brought to bear upon the prospective father-in-law in

order to prevent the breaking of an engagement. A whole social network
could be put into motion in order to protect the interests of the groom
to be.

conclusion
The correspondence of Symmachus offers a very lively picture of marital
strategies and practices in late antiquity and illustrates the great vitality
and importance of betrothals among the Roman upper classes. When
compared with Pliny, Symmachus’s letters seem remarkably practical and
display an eminently utilitarian view of marriage. Such a view is in sharp
contrast with the idealized view of matchmaking and marriage that Pliny
puts forth. But this difference might be more a consequence of the different
focuses of the authors and the different purposes of their letter collections
than a reflection of any actual changes in matchmaking practices or the
attitudes toward marriage. Symmachus’s letters reconfirm beyond any
doubt the importance of marriage for the formation of family alliances.
The correspondence depicts in great detail the struggle of members of the
senatorial aristocracies to find suitable brides for their protégés. In such a
competitive marital market, the help of a well-established aristocrat with a
wide network of connections was crucial. Matchmaking seems to have been
just another aspect of patronage, as attractive to aristocratic matchmakers
as it was to Christian bishops for the opportunity that it afforded them to
expand their social network and call in favors.

 Different kinds of sources (even Christian treatises on marriage) show a remarkable continuity
with regard to Roman ideals of marriage. See Evans Grubbs : –.
 chapter 4

Constantine the lawgiver

Jill Harries

Nearly sixty years before Constantine was proclaimed at York on  July

, Philip the Arab celebrated the conclusion of Rome’s first millennium.
Constantine, therefore, was the product of Rome’s eleventh century, the
first fifty years of which had been dominated by military emperors. Con-
stantine himself claimed descent from one, Claudius II Gothicus. The
soldier-emperors’ priority had been the security of the empire, and their
focus had been the frontiers, far from Rome. Like them, Constantine sel-
dom visited the ancient capital after , returning only for his decennalia
in  and vicennalia in . Much of the rest of his reign was spent on the
move. Constantine’s role as a military leader in charge of a mobile court,
which, by his own account, was more like a camp, also had an impact on
his legislation: his palatini were granted the soldiers’ legal right to hold
property independently as peculium castrense, because, despite being schol-
arly civilians, they shared the hardships of the military life – dust, toil, and
long marches.
Among the long-suffering palatini were, perhaps, the emperor’s jurists.
The sidelining of the ancient capital, with its long tradition of legal edu-
cation and learned interpretation of statutes and custom, may have been
one element in the decline of the independent jurist. Herennius Modesti-
nus, the pupil of Ulpian of Tyre, had flourished under the Gordians (he
is last heard of in ), and his successors were few and obscure: Arcadius
Charisius, the latest jurist cited by the Digest, may have been known to
Constantine but is a shadowy figure, as also are Gregorius and Hermo-
genianus, compilers of the Diocletianic collections of rescripts in  and
–, and the native of North Africa known as Paulus, who composed
the Sententiae, probably late in the reign of Diocletian. But the distance,
physically and intellectually, of emperors from the cultural environment
in which jurisprudence had flourished did not result in the demise of the

 Pan. Lat. ()..–.  CTh ..,  May .


 jill harries

discipline. Continuity was maintained noiselessly over the decades by the,

to us, nameless bureaucrats whose job it was to manage court controver-
sies and legal conundrums brought to their – or rather the emperor’s –
attention. Although we do not know for certain which officials were
entrusted with the task of drafting imperial legislation, it is probable that
edicts were the responsibility of the magister memoriae (who would later
act as the quaestor’s deputy), while letters were drafted by the magistri
epistularum. The legal texts were promulgated throughout the empire in
the form of edicts and letters as being the emperor’s own words and would,
in practice, have been a combination of what the emperor had decided
and what the officials deemed compatible with their understanding of the
Roman legal tradition. In such a process, tensions were inherent: whatever
the legal administrators thought, the emperor had the power to disregard
precedent, create his own rules, and even perhaps impose his own style.
The focus of this chapter will be for the most part on aspects of testamen-
tary legislation. While the Christian influences on Constantine have been
extensively discussed, it is becoming increasingly clear that the emperor
was basically a traditionalist, a legislator who on the whole worked within
the established juristic tradition. However, it was not in the interest of his
Christian propagandists to draw attention to this. Constantine, qua first
Christian emperor, was a prisoner of the image-makers, both in his own
time and posthumously. The full significance of his conversion could be
appreciated only in retrospect, and his “Christian” achievement became
subject to distortion, not only by such opponents as Eunapius, followed
by Zosimus, but also by his natural supporters. Prominent among the lat-
ter was Eusebius, whose Life of Constantine (VC) was put together soon
after the emperor’s death, and whose account of Constantine’s legislation
has proved influential down to the present day. But he was not alone
in exploiting the image of Constantine the legislator. The distortions to
the record introduced by Antiochus Chuzon and his fellow editors of the
Theodosian Code were less blatant but nonetheless significant.

constantine and the theodosian code

The compilers of the Theodosian Code did not set out deliberately to per-
petuate or redefine the image or the legacy of Constantine. Nevertheless,
 Humfress : –.
 E.g., Barnes : –. For analysis of Constantine in the context of the general morality of the
fourth (or ab urbe condita eleventh) century, see Evans Grubbs .
 For its compilation in detail and problems of its text, Matthews  is the standard work. For a
briefer study, see Honoré ; and for controversy on the sourcing of the contents, Sirks  and
Matthews .
 Constantine the lawgiver 

this was one effect of their project, and it applied in three ways: to the code
itself; to the concept of the so-called lex generalis, a fifth-century category
of legislative activity imposed on fourth-century material; and to the edit-
ing of some individual entries. On the code, a choice was made to begin
the time span covered by the code with the legislation, in the forms of
edicts and letters, but not rescripts, of the famous (inclitus) Constantine.
Although this was consistent with the notion that the Theodosian project
was a continuation of the codes (of rescripts) by Gregorius and Hermoge-
nian in the s, the authors of the enabling constitution of March 
were careful not to say this. Instead, the new code would be compiled “in
the likeness of” (ad similitudinem) the Diocletianic codes, but it would not
be, officially, a continuation. The use of the term “likeness” also got round
the awkward facts that rescripts formed the bulk of the Diocletianic codes
but had been excluded by Theodosius; that Gregorius’s and Hermogenian’s
arrangement of their material was different; and that the Codex of Her-
mogenian had itself been continued after its original redaction in , with
the addition of later imperial constitutions.
While the implications of the decision to start with Constantine for
the “Christian” character of the code are debatable, the effects of the
decision were twofold. One was to give the legislation of Constantine
priority under every titulus relevant to his concerns: his was the constitution
that lawyers consulting the code would see first. The other, till recently
“hidden from history,” was to enable the unacknowledged assimilation
into the code of laws that emanated first from Galerius, Licinius, and
even one from Maximinus Daia on exempting the urban plebs from the
capitatio. All this reinforced Constantine’s special position as a legislator.
However, because, unlike Eusebius, the editors of the Theodosian Code
worked within the secular Roman legal tradition, the laws of Constantine
were preserved within traditional structures, those of the Praetorian Edict
(books –), and criminal and administrative law. Only in book  were
the compilers obliged to innovate, to create a new framework for the laws
of Constantine and his successors on right religion. For fifteen books out of
sixteen, therefore, Constantine is, largely, categorized as a secular legislator,
even when his laws explicitly referred to matters Christian.
In the interests of brevity and clarity, the editors of the collected
constitutions removed superfluous verbiage, tidied up the grammar, and
moved sections of laws around to fit under the relevant tituli. This editing

 CTh ...pr. March .  Turpin ; Matthews : .

 Mitchell ; Corcoran .  Harries ; Matthews : –.
 CTh .., December .  Matthews : –.
 jill harries

process created a misleading appearance of timelessness and legislative

uniformity; the collected laws were a statement of “what the law was”
for the benefit of scholars and litigants in the fifth century, although in
the interests of scholarship, some obsolete legislation was preserved. But
the definition of “law” offered by implication by the codifiers is itself
problematic. In what sense were these excerpted edicts and letters “law”?
Fergus Millar has observed that “it is evident that the vast majority [sc. of
laws] do not contribute to the principles of Roman law itself, but are in
essence administrative measures, of greater or lesser degrees of generality,
often laying down penalties for non-observance.” While this must, of
course, be recognized, there are two caveats. First, ancient jurists themselves
increasingly accepted that “administrative law” counted as “law.” This
recognition, in turn, generated learned little treatises on the officia of the
praetorian prefects and other officials by Ulpian and others. Second, it
is quite hard to find anyone in antiquity who wrote or spoke about the
“principles of law,” apart from orators, who used law for emotive effect,
and authors of philosophical treatises; juristic writings, like imperial legal
rescripts, which the jurists both wrote for emperors and then exploited in
their own writings, were about specific cases. It is not a coincidence that
the word responsum or its cognates was used for both jurists’ responses to
consultation and imperial rulings.
The second challenge posed by the Theodosian Code to the reputation
of Constantine the legislator was its dependence on the lex generalis and the
extent to which “generality” could be inferred from responses to specific
problems. That imperial legislation was in general “reactive” is in danger of
becoming an unquestioned orthodoxy, regardless of the very clear statement
in the imperial oratio to the Roman Senate in November , which gave
equal weight, on the one hand, to the “spontaneus motus,” which inspired
emperors to legislate, and, on the other, to external promptings from
petitions (precatio), reports (relatio), and legal cases (lis mota). True, even
Theodosius II was unlikely to advertise imperial passivity, but even were
he personally more reactive than innovative, this would not apply to all
emperors equally. As Millar commented of the tetrarchy, “the edicts of this
period reflect a quite new degree of innovation and intended impact on
the population.” Indeed, the act of codification itself was an assertion of
central control. As Roger Rees has argued, the codifications of Gregorius
and Hermogenian can be contextualized in the assertive legislative character

 Honoré : –.  Millar : .  CJ ...

 Millar : –.  Rees : .
 Constantine the lawgiver 

of the tetrarchy as a whole in the s, asserting its control not only over
the courts of the empire but also the past history of law itself. Constantine
did not choose to celebrate Diocletian in any of his extant laws, but they
had left a model of innovation and assertive, “spontaneous” lawmaking to
which he could, if he needed, conveniently turn.
In the relatively ordered world of fifth-century Constantinople, time
and attention could be devoted to what was, and was not, a law of general
application. It may well have been, as S. Corcoran has argued, that in the
early fourth century, too, “the division in the material should come to
be, not that between letters and edicts, but between the letters and edicts
that form part of a general enactment and those letters which are rescripts
in reply to officials.” But that question was not asked in those terms,
as far as is known, in the retinue of Constantine. For Constantine’s legal
administrators, the most recent handbook of imperial law was the codi-
fication of rescripts by Gregorius and Hermogenian. Although in theory
applicable only to the case for which they were issued, in practice rescripts
were often reused in parallel or analogous cases, as their function was to
state a legal principle or issue to be resolved, not to determine facts. Even
if Constantine had preferred to distance himself from a project so closely
associated with Diocletian, the two codes would have reinforced, for him,
the “generality” in practice of rescript law. Conversely, the compilers of the
Theodosian Code conceded some flexibility in their idea of “generality” as
applied to Constantine: what he said in a public hearing to two women, or
to an assembly of soldiers, was counted by Antiochus Chuzon’s team as
“general law.” The decision to extract a general principle from a specific law
was that of the code compilers; it did not (necessarily) reflect the intention
of the early fourth-century legislator.
A consequence of the arrangement of the code, which permitted the
inclusion of separate snippets of single laws under different headings, was
that the actual specificity of an alleged lex generalis could be obscured. One
example is a Constantinian “law” on the citation of jurists, which may
have been edited in order to provide a precedent for an initiative in 

 The one apparent exception, CTh .., can be ascribed to Maximinus Daia.
 Assertive or innovative lawmaking does not, of course, preclude the presence of traditional or
“reactive” content in the resultant legislation. We must distinguish clearly between reactivity in the
process of generation of laws (through relatio, suggestio, etc.) and reactive content, derived from the
legal tradition.
 Corcoran a: .
 CTh .. (Constantine gives his ruling on the fate of unlawfully confiscated property in Latin,
although Agrippina and Codia are Greek speakers); CTh .. (confirming privileges of veterans);
Corcoran : –.
 jill harries

to control the citation of jurists in courts. CTh . is a section about the
writings of the jurists and which were to be approved or not. CTh ..
of  September  records Constantine’s ruling that the notae of Ulpian
and Paulus on Papinian were to be “abolished” (aboleri), meaning that they
should not be valid for citation in court. Second in the section, and also
Constantinian, was an endorsement, in superlatives, of Paulus’s Sententiae
for its clarity (plenissima luce), style (perfectissima elocutione), and quality
of legal reasoning (iustissima iuris ratione). Third was the extract from the
oratio to the Roman Senate of ,  which restricted the number of jurists
to be cited in court to five, and those cited by them, subject to a check
of the manuscripts. Moreover, Constantine was credited with anticipating
Theodosius II’s impatience with the long-winded and tedious disputes of
jurists. The emperor refers to the desirability of ending the “perpetuas
prudentium contentiones,” the everlasting disputes of jurists: as Caroline
Humfress has suggested, this could apply to the emperor’s irritation with
prudentes on his own staff. But set in its new context of CTh . as a whole,
Constantine’s “laws” would be read as providing precedent both for the
“law of citations” from , itself an extract from a long and detailed legal
manifesto addressed to the Roman Senate, and for the aims of Theodosius’s
code project.
In fact, the comprehensive abolition of the notae of Ulpian and Paulus
may never have happened. The measure was triggered by arguments as
to the status of a son-in-power of a father punished by deportation, who
is then pardoned and returns with his civil rights restored. The specific
question related to a will made by the filiusfamilias, and Constantine ruled
that the son should return to the potestas of his father, whose status and
property were also restored to him, but that, assuming the son was legally
an adult, the measures he had taken in his father’s absence were to stand.
In so doing, Constantine agreed with the opinion of Papinian on the
matter “setting aside the notae of Ulpian and Paulus” (remotis Ulpiani et
Pauli notis). This law and CTh .. are dated two weeks apart; given the
delays in the production of the multiple copies required by the system,
this may be the same law or generated by the same legal discussion. In
other words, the authors of the notae found themselves in the unfortunate
position of disagreeing with the emperor (and Papinian) on one legal point;
for this, their entire writings were apparently rendered inoperative, because
“they preferred not so much to correct (Papinian) as to malign him.” But,

 CTh ...  CTh ...  Humfress :  n. .

 CTh ..,  September ; cf. Matthews : .
 Constantine the lawgiver 

despite the language of “abolition,” it is not clear that Constantine had any
intention of banning the notae from all cases. The ban is the creation of
the code excerptors, not of Constantine; the editors, in this case, had also
become the lawmakers.

constantine, eusebius, and the jurists

Eusebius’s representation of Constantine, the Christian lawgiver, has long
dominated the tradition. His emphasis was understandable, as his aim in
the VC was to provide a full portrait of a new kind of ruler, and good
laws were a traditional part of the activity of the good ruler. Moreover,
many of Constantine’s laws really were Christian in application – such as
those concerning privileges and exemptions for the clergy, manumission in
church, or some kind of privileges for episcopal jurisdiction. Eusebius’s
inclusion of Constantine’s Christian letters to the provincials in his bio-
graphical construction of the new Christian pius princeps as lawgiver was
legitimate, because they both affirmed Constantine’s personal faith and,
being also an imperial enactment, were an encouragement to others to do
likewise.
Legislation, its rhetorical justification, and the personal beliefs of the law-
giver are thus deliberately and inextricably intertwined. Eusebius’s account
of Constantine’s personal addresses on Christianity follows immediately
on his account of Constantine’s legislation that removed the penalties on
celibacy and allowed simplification of the wording of wills; they are
treated as virtually part of the same activity – and, as we shall see, gave the
impression that Constantine’s laws were more Christian than they in fact
were. While emphasizing Constantine’s personal authorship of his spon-
taneous addresses to his inevitably admiring courtiers, Eusebius adds the
reasoning and the rhetoric that justified the changes from the old, unfair
laws; only those who offended deliberately should be punished, and this
category did not include those who could not have children or had chosen
abstinence for reasons of philosophy or religion. Yet it is not clear whether
this gloss should be ascribed to Eusebius or to the emperor. Both may at
times have sounded remarkably similar; Eusebius depicts Constantine, the
Christian preacher, breaking off in a sermon from a disquisition on provi-
dence and judgment to castigate thieves, frauds, and profiteers so forcefully

 CTh ..– (privileges of clergy); CTh . (manumission in church); CTh .. (a corrupted
text, on episcopalis audientia).
 VC .–.  VC .–.  VC .–, cf. CTh ...  VC ..–, .
 jill harries

that some courtiers bowed their heads in shame. Given the ferocity of
Constantine’s laws on such wicked actions as aiding abduction or denying
access to justice, perhaps fear rather than shame was the true motive for
not meeting the emperor’s eye.
The rhetoric of belief was already a part of how government worked,
and emperors could expect practical consequences to follow. When, in
, Maximinus Daia indicated his distaste for Christians (but did not
directly order persecution), sycophantic cities responded by turning on
their Christian populations. Conversely, as the idea of Constantine’s
Christianity spread among the populations of the east, at least one com-
munity, Orcistus in Phrygia, saw its chance to further its case for becoming
an independent city by advertising to the emperor that it had all the tradi-
tional qualifications – and was Christian as well. Eusebius, who had earlier
observed Maximinus’s skill in furthering persecution without appearing to
endorse it, celebrated Constantine’s grant of city status, because of their
Christianity, to Maiuma in Palestine (renamed Constantia) and to Con-
stantine in Phoenicia, and he hinted at great crowds of converts but was
unable to produce further examples; in fact, despite the self-assertion of
its Christian element, Maiuma retained a strong pagan character and a
controversial festival till late in the fourth century.
The use of rhetoric to convey “law” through the publicly disseminated
and posted edict or epistula carried the further risk that, irrespective of con-
tent, its language would obscure its relationship to the tradition represented
by the jurists, whose preferred style avoided the techniques of persuasion
dear to emperors. Down to the end of the Severan dynasty in , the juris-
tic tradition had, on the whole, avoided the use of rhetorical language, thus
rendering itself both specialist and boring in the eyes of some readers, then
and now. In Alan Watson’s words, “they had a style of interpretation that was
inward-looking and not too geared to social engineering.” For jurispru-
dence deliberately to distance itself from the public culture of rhetoric and
the rewards that followed from it required justification. The answer lay in
the alleged purity of legal reason, described by Ulpian following Celsus
as the “true philosophy,” and an implied disdain for considerations of mere
utilitas, although it was always expected and part of the jurist’s social role
 VC ...
 CTh .. (): the mouth of a nurse who connives at abduction/elopement is to be stopped with
molten lead; CTh ..– (both , from same law), holding governors and apparitores to account
for ensuring open access to justice.
 Mitchell .  MAMA ..  VC ..  VC ...
 CTh .. (Maiuma permitted, ); .. (Maiuma abolished, ).  Watson : .
 Ibid.
 Constantine the lawgiver 

that his learned opinions would be fed to advocates for use in court. These
considerations could have put ancient jurisprudence, with its intricate and
self-referential traditions, on a potential collision course with the priorities
of practical men such as Constantine.
In the world of imperial lawmaking, however, the jurist and the ora-
tor were less distant than might appear. All men with sufficient means
had received the same education with, first, the grammaticus and then the
rhetor. All, then, were schooled in language and vocabulary and rhetorical
technique. Part of the education system entailed the practice of controver-
siae, speaking on one side or the other in different kinds of arguments,
including legal cases. Thus all schoolboys received some legal training
to some degree, although not all would go on to be legal specialists. By
the fourth century, legal training at Berytus (Beirut) was highly developed
and the tradition of legal education at Rome through public seminars and
discussions probably continued unbroken from the early empire through
the third century. It was thus possible for Constantine from  onward, if
not earlier, to have access to legal specialists based at Rome, and from  to
the graduates of Berytus, whose expertise could support the draftsmen of
the emperor’s legal letters – provided they could endure the alleged rigors
of the emperor’s military-style court.
The extent to which he made use of them as heads of the secretariats,
and the juristic tradition represented by them, is another matter. In a reign
spanning more than thirty years, there would have been a fast turnover of
legal advisers of different levels of intellectual attainment or opportunities of
access to the emperor. Willingly or not, the writers of the laws subscribed to
a preference shared by many emperors for finding solutions that “worked.”
However, as we shall see, they also had a general understanding of legal
practice, perhaps superior to that of Eusebius in some respects. And in
one area, the law on the making of wills, concern for which goes back to
the jurists of the late Roman Republic and beyond, the jurist-bureaucrats
(unlike Eusebius) knew exactly what they were doing.

last wishes
Arguments from silence are more than usually risky when using the Theo-
dosian Code as a check on the veracity of ancient accounts of imperial law,
as some two-thirds of the code is missing. It is not possible, therefore, to
argue that, because Eusebius’s version of Constantine’s legal reforms is not

 Crook : –.  Liebs : –.

 jill harries

supported by the extant constitutions in the Theodosian Code, Eusebius

was mistaken: “lack of authenticity does not follow from its [a constitu-
tion’s] absence from the Theodosian Code.” As we shall see, there is no
support in the Theodosian (or Justinianic) Code for the impression cre-
ated by Eusebius that Constantine enacted a general simplification of the
language in which wills were expressed; yet that in itself does not disprove
the existence of such a ruling. Rather than apply a dubious argumentum
ex silentio, an alternative approach is to identify the legal discourse within
which the lawmakers of Constantine appear to have operated and to deter-
mine whether there was scope within it for the radical reform implied by
Eusebius’s account.
The case for believing Eusebius is prima facie a strong one. He was there
or thereabouts, he had access to laws and public documents, he was inter-
ested in such texts, and he knew things that we do not, which validated his
presentation of Constantine the lawgiver; so how can we be certain than
Eusebius was wrong? I will argue that Constantine’s extant legislation on
the wording of wills, incomplete though it may be, consistently addresses
questions of wording and validity, for which there were already precedents
in the works of jurists and in imperial legislation – including grants of
testamentary privileges to soldiers – relevant to the question of “simplifi-
cation.” Even so, we cannot prove Eusebius wrong: the dying Constantine
could have, in theory, produced some blanket enactment lost to posterity,
which broke completely with previous legal discourse on wills. But we can
at least argue that his version of Constantine’s policy is inconsistent with
the legal conventions on the subject, to which, based on all the available
evidence, Constantine seems to have been content to adhere.
Firstly, Eusebius’s presentational techniques themselves may prompt
skepticism. After devoting much of book  of the Life of Constantine
to that emperor’s regulation of the Church and his laws against pagans,
Eusebius’s relatively brief account of Constantine as a general legislator
focuses on a few select enactments, which he endowed with an appro-
priately Christian gloss. One was the removal of penalties for celibacy set
out in the Augustan lex Papia; despite Eusebius’s Christian reading of
this measure, it is now agreed that the main beneficiaries would have been
the Roman aristocracy in general, rather than Christians in particular.
Moreover, contra the impression created by Eusebius, Constantine did not
repeal all the lex Papia Poppaea, which is assumed decades later to be

 Matthews : .  VC .–.

 Recorded also at CTh ...  For references, Averil Cameron and Hall : –.
 Constantine the lawgiver 

still in force. The same propensity to generalize (and thus exaggerate)

is shown in Eusebius’s statement that gladiatorial games were abolished:
CTh .., which orders that criminals condemned to become gladiators
should instead be sentenced to the mines, is a far more limited measure,
and later constitutions under the same heading show no knowledge of the
reform. As Eusebius may, in theory, have used an alternative lost source,
all that can be said is that corroboration of Eusebius is not preserved either
by the CTh or the CJ; rather, the CJ itself generalizes by omitting the
section on the mines, recording for posterity Constantine’s “law” that
“bloody shows displease . . . and therefore we entirely forbid the existence
of gladiators.” As happened with the unlucky notae of Ulpian and Paulus
discussed above, the editors became the legislators.
Eusebius represented the situation under the “ancient laws” on wills as
unfair to the dying, who were obliged to remember precise legal formulae
and terminology, adding that the requirement that language be accurate also
led to ingenious attempts to circumvent the wishes of the deceased. As we
shall see, linguistic accuracy was far less important than Eusebius implies.
Moreover, in line with rhetorical practice, Eusebius exploited the “ancient
law” or vetus ius, as his arguments required: while Licinius was vilified for
overturning ancient law, Constantine’s legislation was praised for doing
precisely the same – because the “ancient law” was bad. His insistence, not
explicitly but by implication, that the provision had a Christian purpose
encouraged the editors of the Oxford Vita Constantini to assume that this
“is to be read as intended to prevent the circumvention of donations to the
Church.”
In Eusebius’s account, Constantine’s reform allowed the dying person –
any dying person – to express his (or her) wishes in simple, plain, everyday
language in an ordinary document or even without writing at all, pro-
vided, as Eusebius carefully states, that this was done in the presence of
reliable witnesses who were capable of remembering precisely the instruc-
tions entrusted to them. Eusebius is clear that the permission applied to
any kind of testator on the point of death. On witnesses, he has direct cor-
roboration from the code: CTh .., dated probably to , stipulates the
requirement that seven or five witnesses be present to validate either a will
or a codicil. The extract from the law has no mention of simplification and
is phrased in negative terms: without the witnesses, the codicil or will lacks
validity. How far this was effective in practice cannot be known; evidence
 CTh .. (); CTh ... ().  CJ ..; cf. Matthews :  n. .
 VC ...  Eusebius, HE .. = VC ..–; Corcoran a: –.
 Cameron and Hall : .  Eusebius, VC ...
 jill harries

from the Judaean desert in the second century suggests that the number
of witnesses for various kinds of contract could range from two to seven,
and, as we shall see, many people made wills without recourse to expert
help, including, presumably, the formality of witnesses.
A central text for establishing the limitations on Constantine’s thinking
is CTh .., which I will discuss in detail. It was addressed to Lucer(ius),
or Locrius, Verinus, vicarius of Africa from late  to mid- and prefect
of the city of Rome between September  and June . Verinus, whose
name Locrius is uncommon, may have been the son of two individuals
buried in the Christian cemetery at Clusium and, perhaps, therefore was
a Christian, too. The recipient of a poem from the Elder Symmachus,
Verinus had perhaps fought in Armenia under Galerius in ; in that
case, he and Constantine would have been acquainted for a long time. As
vicarius, he received at least three constitutions from Constantine on the
administration of the criminal law, listed separately but probably part of the
same law as all were received and/or posted at Carthage in March . These
order, first, the reintroduction of the culleus, or sack, as the punishment
for parricide; second, a list of penalties on a sliding scale depending on
status for the crime of counterfeiting; and, third, instructions to seek
out the authors of defamatory writings and punish them, even if some of
their allegations were proved true. If part of the same enactment, these
may represent the epistolary version of a general edict on the criminal law,
rather than anything specific to the recipient. More important for the role
of imperial initiative (spontaneus motus) in the generation of laws is that,
while one part of the law may have been a response to a specific question,
the enactment as a whole – albeit one posted in serial form, if the dates are
right – seems to represent a Constantinian initiative to clarify the workings
of the criminal law as a whole.
More relevant, because it relates to the law on wills, was Constantine’s
instruction, perhaps in response to a question raised by Verinus, that half-
brothers with the same mother could not launch a lawsuit complaining
of unduteous will on the part of the paterfamilias; the constitution uses
jurist-speak in referring to the “assistance of the praetor” (praetoris auxilio),
meaning resort to the Praetorian Edict. As prefect of the city, Verinus was
the recipient of a number of instructions from Constantine on matters
ranging from the liabilities of debtors to municipalities to the age of legal
 Meyer : –; : –.  PLRE I, s.v. “Locrius Verinus” (), pp. –.
 Symmachus, Ep. ...  Barnes : –.
 CTh .., issued in November , received at Carthage on  March .
 CTh ..,  March .  CTh ..,  March .  CTh ...  CTh ...
 Constantine the lawgiver 

capability and the monetary equivalents to be charged by suarii for pigs

in Italy, which were to be fixed by detailed enquiries into the local market
rates. Such concerns illustrate the realities of the life of any late Roman
Emperor: pigs were often a more pressing concern than were reflections on
the Divinity.
The constitution on wills addressed to Verinus was listed by the Code
compilers under the heading De familiae erciscundae. This, as the terminol-
ogy indicates, was a very old institution, going back to the Twelve Tables,
and it concerned the division of the estate of a familia among the sui here-
des, on the death of the paterfamilias. If the option that it should be held
in common by all the sui heredes (heir-executors) was rejected, the heirs
could proceed to a division by agreement, perhaps involving a sell-off of
all or part of the holdings to allow each to receive what was his or her due.
Cicero’s correspondence with Atticus recounts a number of share-outs in
which he or Atticus was involved as a joint heres; in these cases, as he was
not related to his co-heirs, Cicero would have been empowered to act by
the will of the deceased owner and could perhaps expect little financial
return. But it was also possible – and probably quite common – for a
testator to apportion unequal shares among the sui heredes, a measure that
might not always find favor with those disadvantaged by the will.
The allocation of CTh .. to the titulus is justified by the opening
statement that, when an estate requires to be divided (cum dividundae res
erunt), there should be no doubt but that, ideally, an agreement should be
reached by the sui heredes without any resort to litigation (consensum sine
ulla controversia commodandum). The main topic, however, is to the extent
to which precise legal language is required of the testator. As the relevant
first provision has eight separate clauses, I will split the text (but not the
sentence):
() Quod vero ad huiusmodi spectat scripturas, in quibus talibus defunctorum fuisse
mens invenitur, ut de testamento intellegatur tantummodo cogitatum . . .
As regards writings in wills of such a kind as are those in which intention of the
deceased persons is found to have been such that he is understood only to have
thought about his will . . .
etsi repugnare ius videatur, huiusmodi quoque conscriptiones, inter suos dumtaxat
heredes valere oportet . . .

 CTh .., posted at Rome on  May.  CTh ...

 CTh ...  See Att...; ..; .a.; ..; ..; and ...
 Pharr’s translation ends a sentence here, which makes for more comprehensible English but loses
the connection of “etsi . . . videatur” with what follows.
 jill harries

. . . even though it may appear to be incompatible with strict legal usage, yet
writings of this kind too should have validity, but with reference only to the sui
heredes . . .
. . . quemadmodum valent scripturae simpliciter incoatae, quas nulla solemnitatis
adminicula defendunt, solis nixae radicibus voluntatis.
. . . on the same principle as that by which writings merely begun are valid,
although they have no protection afforded by legal formalities, being rooted only
in the intention [sc. of the deceased].
() Licet enim sub testamenti vocabulo coeptae, cum perfectae non sint neque appellari
aliter ullo modo evanuisse videntur . . .
For although (writings) begun under the name of a will and which cannot be
called anything else, when they are not completed are held to be null and void . . .
. . . tamen dispositiones ultimae coloratam iuris imaginem referentes iustius in se legum
proclivem favorem debent provocare.
. . . nonetheless the final statements of intent, which convey a color and semblance
of law, ought with greater justice to make appeal to a favorable reading of the laws.
This law is far more limited in scope than is the Eusebian version of
Constantine’s reform. Very carefully, in a succession of qualifying clauses,
the concession with regard to an incomplete will (“[valere] oportet”; and
“debent [provocare]” are the main verbs) is hedged about with conditions.
It applies to the sui heredes only, and only if the sui heredes (that is, those who
are in potestate of the testator and would inherit in the event of intestacy)
are the beneficiaries of what can be inferred as the testator’s intention. This
is to apply, if the wording of the will contains the “color and semblance
of law” and is compatible with existing laws; if those conditions are met,
the children and grandchildren should receive the shares allotted to them.
In other words, if the will was left unfinished or failed to conform to strict
legal requirements, it would stand – provided that it benefited those who
would have been the heirs if no will at all had been made. If this is the law
to which Eusebius refers – and the reference to the semblance of legality
in the wording suggests that it is – the reform was far more limited in
scope than Eusebius implies. Moreover, given its restricted application to
sui heredes, this law at least did not apply to bequests to the Church.
But CTh .. was not the only Constantinian text to deal with the
simplification of the wording of wills. Its general approach, to simplify
requirements in specific situations, is paralleled by a law that survives in
the CJ concerning the right words to use when instituting heredes in a will.
 CTh ....  CJ ...
 Constantine the lawgiver 

Naming the heir-executors was perhaps the single most important function
of a testament, and testators were obliged to follow certain conventions
in both naming and not-naming (i.e., disinheriting) those who in the event
of intestacy might inherit anyway. The law is headed “Constantine to the
People” but is dated ; if the date is correct, this is a law of the three sons
but is none the less significant for that as, if it postdates Constantine, it
establishes that Constantine did not reform wills in their entirety.
Again, the first part of the text is worth citing, not only for the ques-
tionable beauties of its style but for the nature of its self-justification:
Quoniam indignum est ob inanem observationem irritas fieri tabulas et iudicia mor-
tuorum, placuit ademptis his, quorum imaginarius usus est, institutioni heredis ver-
borum non esse necessariam observantiam, utrum imperativis et directis verbis fiat,
an inflexa. Nec enim interest, si dicatur “heredem facio” vel “instituo” vel “volo” vel
“mando” vel “cupio” vel “esto” vel “erit” (etc.)
As it is unworthy that because of an empty adherence to convention wills should
be made invalid along with the judgments of the deceased, it has been decided
that, setting aside these terms, of which the usage is illusory, for the institution of
the heir, there is no necessity for the observance of a set form of words, whether it
be delivered in direct speech or the imperative mode, or indirectly. For it makes
no difference if it is said “I make as heir” or “I institute (as heir)” or “I wish (as
heir)” or “I mandate” or “I desire” or “let X be” or “X will be” (etc).
Despite its apparent generality, the “simplification” relates to a clearly
defined aspect of will-making and, stripped of the language of moral indig-
nation, the ruling conforms to legal convention. Gaius, in the second
century, had specified the formulations for institution of heirs that were
most appropriate, but he had never said that testators who adopted a dif-
ferent form of words would have their wills invalidated. Moreover, the
importance of intention over defective phraseology had been recognized
in a rescript of Antoninus Pius, as quoted by Ulpian, which ruled that a
will that made the wishes of the testator with regard to his heirs clear was
valid even though the phrase “heres esto” was omitted altogether. In this
law, despite the ornate rhetoric, Constantine or his successors had done no
more than Antoninus Pius and the jurists before them.

wills and the legal tradition

How did Constantine’s preference for testamentary simplification, albeit
in restricted contexts, fit more broadly with the secular, legal, and juristic
 Gaius, Institutes ..  Gaius, Institutes ..  D. .....
 jill harries

tradition? Disputes over wills were a staple of juristic practice from the
first century BCE onward, and much ingenuity was expended in advising
on, or even inventing, problems that might arise. Moreover, Constantine’s
motivation was more complex than Eusebius suggests: his dislike of unnec-
essary complication was a product both of his own military background
and soldierly ethos and of the legal education of his administrators. The
two had come together in the law referred to above: civilian palatini were
entitled to the peculium castrense of the soldier because (allegedly) they
were obliged to undergo the rigors of camp life in Constantine’s migratory
court.
Constantine won an empire through a series of military campaigns
extending over two decades; he would know from experience that the
rules governing the making of wills by soldiers were less rigorous than
those imposed on the rest of the population. His forerunner, Trajan, had
issued instructions to provincial governors allowing the wills of soldiers to
stand, regardless of how they were drafted, because of their “simple minded
ignorance.” Later, in a rescript, Trajan clarified his position: a will can
be properly executed even if it is not put down in writing, provided that
witnesses were summoned and the testator made a clear oral statement
about who was to be his heir and which slave was to be manumitted. As
Constantine was later also to insist, it was the presence of the witnesses
and the formal declaration of wishes that were crucial to the validity of
the will. A casual remark to a colleague, Trajan said, would not count.
Thus Constantine’s emphasis on the presence of witnesses for all wills and
codicils is in line with the procedure insisted on by the tradition even for
informal (but valid) soldiers’ wills.
The juristic tradition on wills in general also lent Constantine’s attempts
at simplification some support. Jurists were interested in wills because their
clients were concerned with their inheritances – if they were instituted (or
not) as heredes (heirs and executors) – with bequests and legacies, or with
trusts (fideicommissa), which could also be established by a will. Not all wills
would have been drawn up incorrectly and then been set aside as a result: for
a will to fail, there had to be a legal challenge from claimants who believed
their rights had been ignored. A will, however phrased, that honored the
rights of those who would have inherited anyway in the event of intestacy
and contained no contentious legacies was unlikely to be challenged. So we
should not imagine an empire full of Roman citizens obsessing, as Eusebius

 Cited by Ulpian, On the Edict Book  = D. ..–.

 Florentinus, Institutes  = D. ...
 Constantine the lawgiver 

would have it, with the niceties of legal jargon, thanks to the complexities
of the “ancient laws.”
Nonetheless, challenges to wills could be expected to be relatively fre-
quent among the elite, especially when large sums of money were at stake,
and the jurists devoted many volumes to asking and answering questions of
how wording should be interpreted. Concern with wills began early, with
republican jurisprudence that created a framework for what came later and
is largely known from citations in the works of jurists living under the
Roman empire. Wills and their wording were of interest to the late repub-
lican jurist and friend of Cicero, Servius Sulpicius Rufus. On the subject
of how a bequest of “furniture” should be interpreted, Servius stated that
the intention (sententia) of the testator should be understood in terms of
common linguistic usage. This was based on two principles. One was that
intention was paramount and that disputes over text should be directed
toward ascertaining what the intention was. Secondly, where interpreta-
tion of language was required, the guide should be what was commonly
understood by a particular word or term. In other words, jurists or litigants
were not encouraged to invent specialist meanings for the use of language
in wills, although clearly there could be much dispute about what mean-
ings were “commonly understood.” Thirdly, as we have seen, while it was
essential that a testator instituted heirs who would carry out the provisions
of the will and also stood to benefit, nevertheless, if the writer got it wrong,
a lenient view was taken.
A second principle was that mistakes that were peripheral to the main
meaning should not be allowed to affect the statement of the deceased’s
wishes as a whole; “superflua non noceant” – the superfluous should not
create a disadvantage. Early jurists were, in fact, tolerant of the vagaries
of human nature and accepted that people made mistakes: Servius, for
example, said that it did not matter if civil litigants disagreed over the
nomen or praenomen of their judge, provided they were agreed as to who
it was. Sensibly, the pupil of Servius, P. Alfenus Rufus, suffect consul
under the second triumvirate in  BCE, argued that anything that was
unintelligible should be ignored, as if not written down at all, while the rest
should stand on its own. The same principle applied to the “impossible
condition,” such as an imaginary testator’s insistence on the survival of
a nonexistent daughter, which was also to be struck out. However, the
nonfulfillment of a condition of an inheritance or bequest – such as the
 D. ....  Cf. CTh .., of March , upholding Constantine’s views on witnesses.
 D. .., Servius as reported by Pomponius, On Sabinus .
 Alfenus, Digesta , De Legatis = D. ...  D. ...
 jill harries

premature death of a beneficiary required to climb the Capitol before

receiving his share – would mean that the legacy would fall.
There was, therefore, some scope for mistakes in matters of detail –
provided the disputants could agree about what was “superflua.” But early
jurisprudence also provided precedent for creative responses to errors of
detail central to the meaning of a will. Alfenus, in many ways one of the
most approachable of the jurists, imagined (or knew of ) a will in which
the testator had made a condition that his heredes should erect for him
a monument or tomb on the lines of the one belonging to P. Septimius
Demetrius in the Via Salaria (a long and ancient road running from the
Tiber mouth inland to Rome and beyond). If they did not, they would
incur a heavy penalty. After the testator’s death, the heirs looked in vain on
the Via Salaria for a monument to Demetrius but found instead another
appropriate structure, to P. Septimius Damas; they suspected that this was
the monument intended in the will, but they were not sure. Their ques-
tions to Alfenus, therefore, were what kind of monument should they
build and, if they did not, because the Damas case did not apply, would
they incur the penalty? Alfenus replied that, if they were sure that the
Damas tomb was the one intended, even though the name had been given
incorrectly in the written will (tamestsi in scriptura mendum esset), they
should follow the intention of the deceased. If, however, they were not
sure, they would not incur the fine, but they should build him a suitably
dignified memorial regardless. Here the error, although only in one word,
is central to the meaning of one provision of the will. Alfenus’s solution
is to resort, again, to intention, and if the intention cannot be ascer-
tained, the heirs are expected to act in accordance with common sense and
pietas.
Both the precedent of the soldier’s will and elements in an admittedly
complex juristic tradition provide a context for Constantine’s simplification
of testamentary formalities. It was already accepted that some mistakes
were permissible and that ascertaining the intention of the testator was all-
important. However, it would be a mistake to push the absence of formality
present in the tradition too far. In the latter part of the second century,
the jurist Q. Cervidius Scaevola, whose opinions on wills were cited by the
legal draftsmen of Arcadius, imagined the outcome of a will, which also
contained a fideicommissum, written by a man who had decided to ignore
formalities altogether:

 D. ...  D. ...  D. ...

 Constantine the lawgiver 

Lucius Titius hoc meum testimonium scripsi sine ullo iuris perito, rationem animi
mei potius secutus quam nimiam et miseram diligentiam; et si minus aliquid legitime
minusve perite fecero, pro iure legitimo haberi debet hominis sani voluntas.
I, Lucius Titius, have here written my will without the assistance of any expert in
the law, preferring to follow the dictates of my own reasoning rather than some
excessive and depressing exercise in pedantry; if I should have written anything
not sufficiently in conformity to law or expert drafting, my wishes should be held
to stand, being those of a man of sound mind, as if in accordance with the law.
“Lucius Titius” is the author of, in my view, an imaginary will; this
allowed Scaevola greater scope for adapting its language to his requirements.
“Titius,” clearly, does not like jurists: their efforts are dismissed as mala
diligentia. Scaevola’s testator, therefore, is an extreme case, a creation of
that jurist’s imagination, who makes no effort at all to get the law right.
Instead, he has based his rejection of the legal experts on the primacy of
intention, acknowledged in juristic discourse, and a statement that he is
not mad and is, therefore, legally competent.
In this imaginary case, “Titius” ignores all the rules of how to draft a
will, even the most basic. But, as the most famous of Roman testators, M.
Grunnius Corocotta, the author of the Testamentum Porcelli, illustrates,
the phraseology of wills was virtually part of folklore: anyone – even a
piglet – could manage the basics. Scaevola, therefore, did not refer to
anything specialist but to the conventions that any testator would be
expected to follow. The consequences of the irresponsibility of “Titius”
were predictable. Scaevola’s next sentence shows the sui heredes, who would
have inherited in the event of there being no will, in court suing for
bonorum possessio, on the grounds that “Titius” had died intestate; the
specific question, peripheral to the present purpose, was whether the trust
could be petitioned for as well.
The moral, as might be expected from a legal consultant, was that it was
better to observe the formalities. Intention and sanity could not protect
a controversial will from legal challenge if the wording was found to be
incompatible with the legal conventions. But the world of Scaevola and
the jurists was itself far removed from the reality of the experience of most
testators. Of the total population of the Roman empire perhaps one-tenth
were qualified to draw up a will in due form, and of those not all would
have done so. Of those who sought outside help, many would have used,
not jurists, but scribes, who knew how to write (on the whole) but were
not versed in legal technicalities; others might have dictated their wishes to
 Contra Champlin : .  On Grunnius as testator, see Champlin .
 jill harries

family members. Edward Champlin’s surveys of real wills found in papyri

and elsewhere suggest that most people did not employ jurists or legal
experts of any description when issuing their “final judgments” – but
that they did understand, in very general terms, what they were doing and
expected their wishes to be honored.

conclusion
Issues of right vocabulary and the institution of heirs is perhaps dry stuff,
even when the economic importance of inheritances is conceded. But
the four different approaches to the same topic outlined above reveal a
cultural world of competing, even contradictory, realities. First, and unac-
knowledged by the elite approaches of lawyers, emperors, and even church
historians, is the reality of Roman legal experience. Wills mattered in law,
but most people could not afford specialists, even if such were available, so
they just made do. Second, Eusebius praises Constantine’s simplification
of testamentary law, allowing flexibility to all testators, as, by implication,
an attempt to help Christians; yet, as we have seen, Eusebius – and his later
readers – may well have overplayed its Christian context and motivation.
Third, we have the perspective of Constantine the legislator: as argued
above, the emperor’s extant laws cannot be read as a general attempt to
simplify wills. Instead, they address questions on specific aspects of the
process, questions also addressed by the fourth element, the juristic tradi-
tion. In offering a restatement of policy on the division of property by will
among sui heredes and the wording of the institution of heredes by will,
Constantine imposed his own moralizing, even opaque, rhetoric, but he
made no substantial departure from precedent. Indeed, Constantine, his
lawyers, and Eusebius all adhered to a central principle, that intention was
all-important. So when Constantine laid a similar emphasis on the impor-
tance of the intentions of those whose wills left assets to the Church, this
was not a special favor to Christianity but merely a tidying-up operation,
enabling the application of a long-held legal principle to a new kind of
testator.
 Champlin : –.
 This does not imply that even country villagers were ignorant of Roman law; cf. the second-century
archive of Babatha, from the Roman province of Arabia (on guardianship, Cotton ).
 CTh ...
 chapter 5

Constantine answers the veterans

Serena Connolly

Moments of high drama are not normally the stuff of law codes. But the
compilers of Theodosius’s Code have preserved for us a dramatic encounter
between the emperor Constantine and his veterans that reveals insights into
the development of both imperial hearings and the recording of them in
the early years of the fourth century, a century often characterized by drama
and ceremony. As I hope to show, the exchange demonstrates important
continuities in the basic structure of hearings from earlier in the principate
and also reflects the increasing use of organized acclamation at emperors’
public appearances. Yet the form of the exchange is not duplicated in
our extant evidence, and records of later imperial hearings did not follow
its model precisely; rather they demonstrate that while the conduct of
hearings may have undergone gradual developments, records of them did
not develop in parallel. Constantine’s text was a one-off: an extensive edited
account of a hearing that was preserved in toto in a legal code. It is possible
that its extensiveness and detail were never replicated because this text
was preserved not just for its legal content but also for its formalities: in
bestowing benefits on veterans who had just fought for him in the year in
which he was elevated to Augustus, Constantine was using a legal hearing
to demonstrate both his commitment to the military and also the style in
which he would govern as emperor.
In a general’s camp at the city of the Velvocorians, a group of veterans –
many of whom had probably fought for Constantine – complained of the
ill-treatment they received despite many years of loyal service. This was
by no means the first time that a group of veterans had complained to a
general or an emperor, and the general outline of Constantine’s response is
equally well precedented: he stated that he wished his veterans enjoyment
in life after their labors and then bestowed privileges on them. But looking
at the exchange more closely, it is possible to see that while core aspects of
the exchange and this situation are familiar from earlier exchanges between
emperors and their subjects, both military men and civilians, there are

 serena connolly

some novel components that mark out this text as typically late Roman
and reveal important changes in imperial ceremonial in the fourth century
that have otherwise gone unrecorded in our changing evidence. In locating
the traditional and the novel in this exchange, I hope to demonstrate that
our record of it reflects an important period of change in the imperial office
as it continued responding to the traditional demands subjects placed upon
it, while also implementing a new style of imperial presentation.
The meeting was recorded and partly preserved in the Theodosian Code
(hereafter CTh) .. and later in the Justinianic Code (CJ) ... On
the basis of these texts, it is possible to reconstruct the following:
Imperator Constantinus Augustus cum introisset principia et salutatus esset a praefectis
et tribunis et viris eminentissimis acclamatum est: “Auguste Constantine, dii te nobis
servent: vestra salus nostra salus: vere dicimus, iurati dicimus.”
Adunati veterani exclamaverunt: “Constantine Auguste, quo nos veteranos factos,
si nullam indulgentiam habemus?”
Constantinus A. dixit: “magis magisque conveteranis meis beatitudinem augere
debeo quam minuere.”
Victorinus veteranus dixit: “muneribus et oneribus universis locis conveniri non
sinamur.”
Constantinus A. dixit: “apertius indica: quae sunt maxime munera, quae vos
contumaciter gravant?”
Universi veterani dixerunt: “ipse perspicis scilicet.”
Constantinus A. dixit: “iam nunc munificentia mea omnibus veteranis id esse
concessum perspicuum sit, ne quis eorum in nullo munere civili neque in operibus
publicis conveniatur neque in nulla collatione neque a magistratibus neque vectigal-
ibus. In quibuscumque nundinis interfuerint, nulla ad venditionem proponenda dare
debebunt. Publicani quoque, ut solent agentibus super compellere, ab isdem veteranis
amoveantur. Fisco nostro quoque eadem epistula interdiximus, ut nullum omnino ex
his inquietaret: sed liceat eis emere et vendere, optimis negotiis pecuniam tractare et
mercimonia agitare, ut integra beneficia eorum sub saeculi nostri otio et pace proferan-
tur et eorum senectus quiete post labores suos perenniter perfruatur. Sed etiam nullo
munere civili, id est corporali sive personali, vel de portorio onere eos adfici concedimus.
Filios quoque eorum defendant decertationes quae in patris persona fuerunt, quosque
optamus florescere sollicitius, ne si contumaces secundum eosdem veteranos comprobari
 In order to recreate as accurately as is possible the original wording of the text, it is necessary
to combine the wording of the CTh and CJ texts: the former has been abbreviated, the latter
interpolated. The text above adheres mostly to that of the CJ, but where words or phrases from the
CTh are included, they are indicated as such, while the wording of the CJ is also pointed out. I am
using Mommsen’s edition of the CTh and the stereotype edition of the CJ.
 CJ: deus te nobis servet.  CTh: magnificentia.
 quiete post labores suos perenniter perfruantur, which is found in the CTh before fisco nostro, seems to
be an abbreviated form of eorum senectus quiete post labores perfruatur, which the CJ retains but puts
after pace proferantur. I add suos perenniter to the CJ text.
 et eorum . . . adfici concedimus: CTh om.
 Constantine answers the veterans 

potuerint decimentur his sententiis, cum praesidali officio adiungentur probabilius ius-
sione mea. Curabunt ergo stationarii milites cuiusque loci cohortis et parentes eorum
desperationem et ad sanctimoniam conspectus mei sine ulla deliberatione remittere, ut
sint salvi, cum semel has consecuntur poenas indulgentiae.” 
D. k. Mart. in civitate Velovocorum Constantino A. vi et Constantino C. Conss.
Once the emperor Constantine Augustus had entered the principia and had been
greeted by the prefects and tribunes and most eminent men, there was an accla-
mation: “Augustus Constantine, may the gods preserve you for us: our salvation
is your salvation: we speak truly, we have given our pledge and speak.”
The assembled exclaim: “Constantine Augustus, why were we made veterans,
if we have no special favor from you?”
Constantine Augustus said: “I ought more and more to increase rather than
reduce the happiness for my fellow veterans.”
The veteran Victorinus said: “Do not allow us to be weighed down by offices
and burdens in all places.”
Constantine Augustus said: “Tell me more openly: what are especially the offices
which persist in oppressing you?”
All the veterans said: “Of course, you yourself see clearly.”
Constantine Augustus said: “Now then let it be made known that by my
munificence it has been granted to all veterans that none shall be subject to any
civic service or public duties or any imperial gratuity or magistracies or taxes. At
whichever markets they attend, they will not be expected to pay the market tax
on what they wish to sell. Tax-collectors too, as they tend to apply more pressure
than do our agents, will be kept from those same veterans. After their labors, let
them enjoy rest all the year. We have also made known to the treasury in the same
letter, that they should not disturb them at all: but permit them to buy and sell,
in the best occupations to do business and trade, so that they may enjoy all their
benefits under their retirement and the peace of our age. But moreover, we do
not allow them to be affected by any service to their municipality, be it as a civic
service or by tax payment, or by the customs tax. Let the decisive contests, which
their fathers fought, provide protection too for their sons, whom we with most
care desire to flourish, and if they can be shown according to those same veterans
to have resolve, let them not be cut down by these decisions, since they will be
handed over to the provincial governors’ offices more fitly by this, my command.
Therefore, let the soldiers of the rural police of the cohort of each district and their
parents be concerned with desperate boldness and see to it that they restore them
to the sanctity of my presence without my needing to deliberate the matter, so that
they may be freed from their difficulties, when they once obey these conditions of
my bestowal of favor.”
Given on  March in the city of the Velovocori in the consulship of Constantine
the Augustus for the sixth time, and Constantine Caesar.

 Filios quoque . . . poenas indulgentiae: CJ om.

 serena connolly

Those who have discussed this text have noted its strangeness but have
not explored it at length. That strangeness may derive from the combina-
tion of the nature of the hearing, the account of it, and the editing of that
account after the exchange had taken place.
The text is a highly stylized and artificial account of a meeting and
conversation, the editing of which has reconfigured a messy, real-life
encounter into a slick administrative procedure. The original encounter
most likely contained more complex initial exchanges, presentation of evi-
dence, detailed negotiations, and consultation by the emperor with his
advisers before he gave a response, which formed the basis of the extant
version. It is possible to see where spoken material was excised or never
written down by looking for those points in the exchange where there is
divergence from the natural and expected exchange of utterances. It is in
the original exchange, which I shall attempt to recreate in outline, that
any innovations in the hearing and the account can be pinpointed. By
comparison of this text with similar examples predating and postdating
the reign of Constantine, I shall also pick out continuities in the text from
earlier periods that persevered later into the fourth century and beyond.
The encounter took place, as is stated at the end of the text, in the
civitas of the Velovocori on  March . Though the date is striking – it
marks the beginning of the Roman administrative year – this is perhaps just
coincidental. More striking, however, is the fact that the year and location
do not correlate. Constantine seems to have been near Serdica in the spring
of , while the most convincing identification of the Velovocori is as the
Bellovacori of Gaul. Three of the Panegyrici Latini place Constantine in
Gaul in . While setting the text in that year would require altering the
consular year at the end of the entry from Constantino A. vi et Constantino
C. Conss to Maximiano A. ix et Constantino C. in the west or Maximiano
A. ix et Maximino C. in Rome, the text is introduced with the phrase idem
 The constitution is “peculiar,” according to Pharr (: ), who goes on to describe its form
as “most unusual, dramatic and vivid, thoroughly characteristic of the Emperor Constantine with
his unconventional type of mind.” He provides the most extensive treatment of the text. The
constitution is “remarkable,” according to Campbell (: ). It is also discussed by Millar (:
), who notes that Constantine has surrounded himself with military officials in a military setting;
and Mitchell (: ), who sees Constantine’s dispensation in connection with a tax break offered
by Licinius to his troops (on which see below).
 The identification was made by Barnes (:  n. ) and is followed by Corcoran (: –),
though Corcoran () keeps the date of  CE and describes Barnes’s conjecture (see below) of
 CE as highly speculative. Alternative identifications of the people of Velovocori are the Vellavi and
the Veliocasses. Both groups are Gallic, and the second is probably another name for the Bellovacori.
On the whereabouts of Constantine during his reign, see Barnes (: –).
 Pan. Lat. ()..; ()..ff.; ()..ff. The dating of these references to  CE comes from
Barnes :  and : .
 Constantine answers the veterans 

Aug., which simply refers back to Constantine as author of the previous

entry; the possibly incorrect consular date of  CE may have been added
subsequently. A date of  CE presents the problem that no entry in
the Theodosian Code is dated as early as this, but it solves a couple of
others: the dispensation to veterans would follow naturally from military
action that took place in the region the previous year, and use of the plural
in the phrase dii te servent is also explained (see below on this phrase).
Constantine is addressed as “Auguste” in the text, though he was invested
as Augustus later in . There are several potential explanations: the
investiture may have simply been a formal recognition of the title he had
already been using that year, or perhaps the title was added by archivists or
editors after the fact.
If Constantine did indeed meet the veterans in Gaul, then we have a
nice example of the regional concerns that David Potter in this volume has
identified as being so important in the later Roman empire. In the years
before  Roman security in the region had been fragile, but the same
panegyrics that locate Constantine in Gaul in  also record his victories
against the Franks. It is likely, therefore, that the veterans Constantine
encountered at the city of the Velovocori had fought for him recently (or
perhaps in earlier unrecorded periods of unrest).
Given the regional specificity of the text, it seems that Constantine’s
dispensation was at first directed only to the veterans of the Gallic cam-
paigns and thus had specific legal validity. Some of the other Constantinian
entries in the Theodosian Code title ., de veteranis – those addressed
to a provincial governor (CTh ..), a prefect of the city (..), and a
praetorian prefect (..) – were specific only to their recipients’ spheres
of jurisdiction; CTh .., however, was not specifically directed in this
way and does not contain language suggesting that the veterans in Gaul
were in a special situation and that only they could have been helped by
its provisions. It is therefore likely either that the dispensation was directed
to all veterans in mind at the time that Constantine delivered it at the

 Alternatively, the text could be dated to  CE; Constantine was based at Heraclea in February and
March of that year according to CTh .., .. and .. (Barnes : –). Consular dating
for this year is Constantino A. vii et Constantio C. Conss., which requires very little alteration of the
dating formula as it currently stands. Moreover, Velovocori may also be a misreading of Vovorum
Minor or Vicianum, both of which are within several days’ journey of Heraclea. Nevertheless, this
suggestion has found little support. Dates of  and  are also possible, though Constantine can
be securely placed in Gaul only in  (Barnes : ).
 For a description of the frequent clashes between the Franks and Romans in the earliest part of
Constantine’s rise to power, see Potter : –.
 serena connolly

camp or that any specific provisions that pertained to Gaul were subse-
quently removed. It may also be the case that the dispensation as it was
composed in  CE was legally specific but acquired de facto generality
through the empire. Though there is no proof for this particular case, Jill
Harries points out in this volume the importance of the phenomenon of
de facto generality in this period; the de iure generality she also discusses
was certainly acquired by the dispensation when it was included in the
Theodosian Code.
Veterans in Gaul would have been familiar with acclamation from mili-
tary life and perhaps civilian life also: Charlotte Roueché has found exam-
ples of acclamation in the west as early as the first century CE. Constan-
tine may have exploited both this familiarity and the regional concerns of
veterans for political purposes and ordered the meeting in Gaul to be chore-
ographed so as to produce a mutually beneficial outcome. But more than
that, the encounter was carefully recorded and moreover edited, perhaps
to benefit veterans elsewhere, but more certainly to bolster Constantine’s
reputation as an emperor who looked after those who fought for him.
This supposition is supported by the fact that in  Licinius offered a tax
break to his soldiers in the Balkans. Depending on its date, Constantine’s
dispensation may have been a catalyst for or a response to that of Licinius.
Perhaps the emperors were in competition for the continued loyalty of
Roman soldiers and veterans.
The text opens with a description of the setting at the principia, in which
the general’s tent was to be found, along with those of other officers and
commanders. It was here, in the administrative heart of the camp, that
generals gave out orders to their officers. This was also the focus point
for discipline, “where military court was held, complaints of the soldiers
heard, and judgments rendered,” and where imperial legislation relevant
to the military was kept in an archive. It was a logical place for a group of
veterans to be heard, and it was logical, too, for the veterans to choose it,
since the military setting might focus the emperor’s mind on the army and
its active participants sway him in their favor. But as will become clear,

 This editing may have taken place between  CE and the entry’s inclusion in the Theodosian
Code.
 Roueché : .  AE , .
 Gabriele Wesch-Klein (: ) points out that CTh .. provided simply a confirmation of a
dispensation granted earlier by Diocletian (CIL  p.  no.  = FIRA   l.ff.).
 EDRL, s.v. “principia.”  Pharr :  on the basis of D . and CJ ..
 See Campbell (: –) on soldiers choosing military courts. The veterans could presumably
also have approached the provincial governor or the emperor in a civil setting, if he was available.
 Constantine answers the veterans 

the veterans were to find themselves quickly subsumed into the ceremony
of the occasion and carefully managed by its choreographers.
Waiting at the principia, along with the throng of veterans ready to
voice their demands, were prefects, tribunes, and eminentissimi, who were
probably praetorian prefects. Far from this being an exclusively military
occasion, the veterans now found themselves joined by civilian officials,
including those who had accompanied Constantine from the imperial
court. The text mentions that assembled officials greeted the emperor, but
it does not mention that the veterans did so also, and perhaps they did
not join in. The phrase salutatus esset has a special meaning in a military
context. Generals could be hailed (salutatus) as emperors by their soldiers,
signifying their support in a competition for the imperial office; generals
(including emperors) were also hailed (salutatus, but also acclamatus) after
victories. When emperors began to be hailed each year, the military
salutatio became simply a ceremonial confirmation of the army’s support
(and it is possible that  March was an apt day for this). (Salutatus esset
would refer simply also to a greeting by the civilian officials.) Whether it
had been arranged or simply prompted by Constantine’s men, the greeting
by the officials made clear to the veterans that they must follow whatever
protocol or ceremony was in place. Moreover, the fact that Constantine
chose to walk into the principia – and is recorded as doing so – is probably
significant. It may have signaled that all present were waiting for and
therefore dependent upon his presence. With the salutatio, the veterans
had secondary significance behind the officials since their complaint was
not part of the initial ceremonial protocol.
We can imagine Constantine ascending a tribunal similar to that
depicted on the columns of Trajan and Marcus Aurelius. The command-
ing general of the camp would probably have used a tribunal on a regular
basis; its use by Constantine demonstrated clearly that the military camp

 This suggestion is based on an inscription found at Dmeir, which is discussed later in this
paper. The rank of eminentissimus should refer to the praetorian prefects (on whose identities,
see Corcoran : ). Alternatively, this word could be a later interpolation, which could
refer also to the lower-ranking praefectus vigilum, though it is difficult to justify their pres-
ence in a military camp, unless they were touring with the emperor. Contrast with the Dmeir
inscription: cum salutatus a praefectis praetorio eminentissimis viris, item amicis et principibus offi-
ciorum. The fact that the opening of this inscription is in Latin but the text of the actual
exchange (including what was said by the emperor) is in Greek shows that the opening is simply
rubric.
 Aldrete (: –) supplies examples of acclamatio in this context from Varro, Ling. . and
Ov. Tr. ...
 Campbell : –.
 serena connolly

had now come under the direct sway of the emperor. The tribunal served
a twofold purpose: it was a ceremonial prop that served to set Constantine
literally above and apart from everyone else, and it was also the traditional
place from which individuals with power and authority heard presentations
and passed judgment.
Once Constantine had taken his place, acclamations followed: Auguste
Constantine, dii te nobis servent: vestra salus nostra salus: vere dicimus, iurati
dicimus. The fourth-century Historia Augusta asserts that Commodus was
acclaimed (anachronistically) with di te servent, thus marking this phrase
out as typical of the period. The sentiment vestra salus nostra salus: vere
dicimus, iurati dicimus is bold: the speakers acknowledge that if their leader
is safe, so are they; by extension, if he is in danger, so are they, perhaps
because they have failed to ensure his security. This sentiment, as the sub-
sequent word iurati suggests, is expressed as an oath, which binds speakers
and hearer in a protective relationship that is theoretically mutual. The
sentiment is found in earlier imperial oaths and can be traced back even
to the Near Eastern and classical Greek worlds, long before the Roman
period, and the opening phrase finds parallels in acclamations in the early
principate. It is unclear whether vestra salus nostra salus: vere dicimus,
iurati dicimus is part of the acclamation voiced by all those present (except
Constantine) or only by the veterans, who were about to make a presen-
tation to the emperor. It seems likely that the acclamation was led by the
officials and that the veterans joined in. We can be sure that the veterans
said it, and in doing so they vouched for the veracity not only of what they
were about to say, but also what they had said, i.e., Auguste Constantine,
dii te nobis servent: vestra salus nostra salus. Whether the veterans wished to
or not, through the collective power of acclamation, they had just sworn
that they would never harm the emperor and that all they were about to
say was true. The significance of the acclamation here is that, rather than
 The importance of the emperor to the soldiers would have been underscored by the probable
presence of an imperial portrait somewhere in the camp, which Stäcker believes would have had a
cultic role (: chap. , especially pp. –).
 For images of tribunals, see Schäfer : –. These images depict magistrates but may be
transferable to imperial audiences and judgments. The Anaglypha Traiani depict a woman before
Trajan, perhaps on trial or perhaps petitioning him. The emperor is seated on a tribunal, with his
officials behind him, as Constantine would be later – though we do not know whether he was
sitting or standing. There is a military adlocutio on Trajan’s Column, section .
 Au. Cass. .–. On the anachronistic acclamations of the HA, see Matthews : –. The
expression was also used by the Arval Brethren in the late second and early third centuries, as
evidenced by CIL .. The phrase appears earlier, as far back as Plautus’s Aulularia (l.), but
is not addressed to emperors until the third century.
 On early Roman imperial oaths of loyalty, see Connolly ; on use of the opening phrase in the
principate, see Aldrete : –.
 Constantine answers the veterans 

being simply something soldiers do or that people do at the games, it is

now part of a regular administrative/judicial protocol, and this acclama-
tion is presumably being voiced by everyone. It therefore unifies all those
who speak it – Constantine’s officials, any army officers present, and the
veterans – in support of the emperor and his authority. With the salutatio
and acclamation, Constantine’s officials have guaranteed that the veterans
have been subsumed into a ceremony over which they have no control,
that any complaint they make must be true, and that they must acknowl-
edge the emperor as their ruler. The circumstances in which the veterans
could present their complaints were carefully circumscribed. According to
Roueché, acclamation could be part of the petitioning process, and so
the veterans might have expected that they would acclaim the emperor and
perhaps present acclamations as part of the process. But they found that
the acclamations were not entirely expressed in their words, and that the
process as a whole was controlled by officials.
While examples of senatorial acclamation can be traced back to the reign
of Trajan, military acclamation could perhaps have a longer history, having
developed from soldiers’ hailing of the emperor and their rallying cries
before battle. Mass acclamation by civilians can also be traced back to the
start of the principate: Vespasian was acclaimed by the local people as kÅrie
Ka±sar OÉespasian¼v e³v swtŸr kaª eÉerg”thv (or something close to
it – the extent of each line of text is not secure) as he entered Alexandria in
triumph, and even earlier Germanicus was acclaimed as kÅrie on his arrival
in Alexandria.
The acclamation of Constantine is apt for its time. The practice of
“call-and-response,” which Gregory Aldrete has identified in the first three
centuries of the principate, in which imperial addresses were interrupted
by (positive) acclamation and which is exemplified by the Germanicus
text, is absent. The disappearance of the to-and-fro from this text of
Constantine is perhaps the result of greater formality at such proceedings.
Roueché describes how acclamations given later to officials in the fourth
and fifth centuries CE began with general honors and were followed by
more specific acclamations, and this is what we find in the Constantinian
text also. But Roueché also finds typical the repetition of acclamations and
the mixing of acclamations of request with declarations of loyalty. The

 Roueché : .

 On senatorial acclamation, see Matthews : –. A useful collection and discussion of the early
evidence for acclamation is provided by Aldrete .
 Vespasian: Acta Alex. B = P.Fouad . Germanicus: P.Oxy. XXV .
 Aldrete : .  Roueché : .
 serena connolly

Constantinian text seems to be simpler, and this is probably the result of

time and context: the more elaborate examples that Roueché finds belong
to later, court contexts. The Constantinian text, then, is situated in time
and therefore also in content between the examples of Aldrete and Roueché.
The text of the encounter, as Roueché points out, reflects strikingly
a phenomenon that began a couple of centuries earlier: the increased
recording of acclamations. This hearing may have been recorded both
because of a fashion for such records and perhaps also because scribes
were now used to recording erstwhile regular official processes (such as
the granting of an indulgentia) in a new, more stylized and ceremonial
way. The Constantinian text also reflects two competing urges: that of
Constantine for acclamation to express more accurately popular thinking
and for acclamation to be institutionalized. We can see the latter urge
beginning to win out in CTh ... Later, acclamation, which was to
be “absorbed into ritual and standardized” in the sixth century, was
increasingly used in ecclesiastical contexts; a precursor to the later, more
elaborate style is found at the end of the long fourth century in the Gesta
Senatus.
The phrase dii te servent preserved in the Theodosian Code, which uses
the plural, is striking, even more so when we see that Justinian’s compilers
edited the phrase into the singular. Given that the plural is the reading
of the earlier text and given the historical and religious context of the
Theodosian Code, it is the lectio difficilior. If the text dates to  CE, the
phrase is not surprising. But if a date of  CE or later is correct (and
this is by no means certain), the text provides evidence of the incomplete
Christianization of the imperial court and army. It is also possible that
this acclamation was so formulaic that by the reign of Theodosius it had
still not been modernized to keep pace with religious changes.
The hearing opens and the veterans have their chance to speak – and
they do not simply speak, but rather cry out (exclamaverunt). This is no
rhetorical term; rather, the veterans were presented even in the official
version of the meeting as venting their frustrations. Their cry began
Constantine Auguste, a brief address. One might have expected something
longer and more elaborate and obsequious, more fitting to the period.
 The text’s simplicity may also be the result of subsequent editing, though it is unlikely that
Constantine’s officials would have excised such declarations.
 Roueché : .  Ibid.: .
 Of the phrase’s significance in  CE, Lane Fox (: ) says, “The scene reminds us, more
than any other, that the majority of Constantine’s army and high command were still pagan in 
and greeted him in their own pagan way.”
 According to Quintilian (Inst. ..; ..; ..), exclamatio refers to feigned surprise or outrage.
 Constantine answers the veterans 

They then followed with quo nos veteranos factos, si nullam indulgentiam
habemus? Indulgentia is a technical term – an “imperial grant of special
favor,” in Clyde Pharr’s translation – that may not have come naturally to
the veterans, and besides, this seems a lengthy cry to be uttered en masse.
It is therefore possible that the veterans let loose a stream of complaints,
which a scribe later summarized. Alternatively, this was an acclamation,
as exclamaverunt may indicate and the homoeoteleuton of the expression
suggests. I think it most likely that among the confused shouts of the
veterans, some voiced this acclamation, which was then recorded by the
scribe.
If the veterans at this point had begun their presentation in an emotional
state, their behavior was well precedented. Tacitus tells us that the veterans
Germanicus encountered were so disordered – a reflection of their mutinous
feeling – that they had to be grouped into their cohorts. Again according
to Tacitus, when Drusus faced another recalcitrant group, he was met with
a roar, but the men soon were so overwhelmed in the presence of imperial
blood that they fell silent. The outburst of Constantine’s veterans may
have been not simply the result of bad temper but perhaps rather of their
disorder. The soldiers’ disorder before Germanicus and Drusus made for a
good story for Tacitus. It did not, however, suit a formal imperial hearing, so
Constantine’s officials may have tried to manage the proceedings; moreover,
a disordered outburst did not suit an imperial constitution, so the scribe
may have been careful to edit what was said.
At this stage we can imagine that order was called by the local senior
officials or Constantine’s accompanying officials. In the hush that followed
came Constantine’s response: magis magisque conveteranis meis beatitudinem
augere debeo quam minuere. The veterans may have been puzzled at this
statement, since it did not answer their question. Rather, it sounded like a
standard phrase, a platitude that functioned as a conciliatory gesture. The
function of this section was to express Constantine’s agreement to hear the
veterans’ grievance, and Constantine’s response was therefore more of an
enabling statement than a genuine reply. Constantine surely did say this or

 The translation is from Pharr : .

 This opening of the statement is strange: it is not the statement of grievances or plea for help that we
might have expected; it is instead a very open question. I suggest, however, that we should consider
its function, which is to request a hearing.
 Given that acclamation of someone as a new emperor was termed conclamatio (Roueché :
), it is possible that other compounds of the verb clamare could have signified the giving of
acclamations. On the language and rhythm of acclamations, see Aldrete : – passim. See
also Aldrete : , on acclamation used to express complaints and petitions.
 Campbell : –.
 serena connolly

something similar, but only after a pause, in which he was given the text
or prompted by his officials. Perhaps this was a statement he often gave at
hearings, with only slight changes to the wording depending on who was
approaching him. The repetition of magis was an effective opening that
guaranteed nothing of substance might be missed as Constantine began
to speak; he also addressed the men as “fellow veterans of mine,” but in
typically late Roman ornate style he used the abstract noun beatitudo instead
of a simpler phrase containing a related adjective or verb.
Constantine’s officials probably stepped forward once more to call upon
a representative of the veterans to make their presentation. Victorinus
delivered the complaint: muneribus et oneribus universis locis conveniri non
sinamur. It is a curiously short statement and is perhaps a summary of the
complaint composed by the scribe. This notion is supported by the fact that
the phrase muneribus et oneribus appears also in CTh .. and ... Or
perhaps Victorinus was tongue-tied, unsure of what to do or what to say,
because Constantine now said to him, apertius indica: quae sunt maxime
munera, quae vos contumaciter gravant? All the veterans – and the entirety
of them is emphasized (Universi veterani dixerunt) – responded ipse perspicis
scilicet, which could be rendered, “Surely you can see for yourself.” The
tone of the statement, encapsulated in scilicet, is of exasperation, despair,
and anger. Given that the veterans and soldiers who were addressed by
Drusus and Germanicus interrupted and jeered their imperial visitors, it
is conceivable that Constantine’s veterans were equally forthright in their
complaints. The veterans probably shouted out many different things,
with one or a few showing their battle scars or (purposefully) bedraggled
appearance and shouting, “Surely you can see for yourself.” Another way
to think about this statement is to imagine them gesturing or pointing to
Victorinus, who held a written text of their complaints for presentation.
At this point the officials probably intervened again so that Victorinus
and perhaps a few others might be brought to one side to present their
complaints in writing or orally while the emperor listened. The other
veterans stood by.

 Campbell :  on Tac. Ann. . and ., describing meetings with Drusus and Germanicus
respectively.
 It is possible that ipse perspicis scilicet was also an acclamation, since acclamations were not always
positive, as Roueché points out (: ). She also reminds us that Constantine strove to ensure
that acclamations about him and others should be genuine expressions of feeling and not the product
of manipulating claques (). But the phrase is introduced in the text with the verb dixerunt, rather
than with a compound of clamare. While this might not necessarily be an acclamation, it is certainly
a genuine expression.
 Constantine answers the veterans 

There was most likely a temporal gap between the veterans’ last statement
and Constantine’s lengthy response. Simon Corcoran, on the basis of
P.Fouad . and P.Yale inv. , believes that the veterans had already
petitioned the emperor and that the court had had a few days to prepare the
official response, which Constantine then delivered at the principia. It is
also possible, however, that the gap is shorter than this. If the emperor was
on the move, which is likely, his court would not have waited at a military
location to receive the veterans’ complaints and to deliver the response
there. In addition, if the court had prepared a written answer after the
complaint had been handed in, one wonders why a ceremonial meeting
should have taken place at all on  March – and we certainly should not
see the entire account as a fiction. If a written petition had been handed in,
it was only a brief time, perhaps a few hours, earlier. It is more likely that
all Constantine and his officials were given in advance was an oral précis
of what the veterans would argue.
The officials, perhaps with the input of Constantine, certainly with his
approval, put together an answer for the men. It is split into two parts. The
first runs from iam nunc to vectigalibus and summarizes what follows. The
rest of the answer supplies details. In the first part, Constantine expresses
his symbolic agreement with the veterans’ complaints. The language of
the opening clause was carefully chosen. Id esse concessum (it has been
conceded) is a formal statement of Constantine’s bestowal of privileges,
which is made possible by his munificentia, the act of bestowal, a term that
can be understood as a synonym for indulgentia. The receipt of privilege by
all the veterans included in the petition is stressed, an aspect of the bestowal
that will be made more effective by publication of the answer, which is
probably alluded to in perspicuum sit (be it known). This phrase echoes the
veterans’ ipse perspicis – they asked him to look at their complaint and he
now makes sure that all may look at his response to it. The second part of
the answer is more technical and legalistic. In fact, the answer is so detailed
that the veterans’ complaint did not need to be included. Constantine
addresses three separate issues, marked off by the repetition of quoque:
the veterans’ right to trade without impediment, the expectation that they
should be exempt from civic services, and the protection afforded to their
sons.
The emperor’s words are recorded not just in this text and kept for the
archived acta; they were probably also transmitted to the veterans in some
form of document and to the treasury in a letter, as we learn from the
 Corcoran : .
 serena connolly

phrase fisco nostro quoque eadem epistula interdiximus. Pharr believes that
the epistula Constantine mentions is the edict that we are reading, though
it is unlikely the treasury would have received anything except this final
part of the text, Constantine’s answer.
What follows is a group of sentiments found elsewhere: Constantine
wants the veterans to be allowed to buy and sell and trade, ut integra
beneficia eorum sub saeculi nostri otio et pace proferantur et eorum senectus
quiete post labores suos perenniter perfruatur. Constantine’s wish that his
veterans enjoy the benefits bestowed by his reign draws on a sentiment
found in many other places: it can be found in an earlier dispensation
granted by Constantine and Licinius to the military, and further back in
time it was voiced by subjects to their emperors; for example, in a well-
known petition to Gordian III, the Skaptopareni ask the emperor to stop
soldiers’ and local officials’ depredations so that they may enjoy the blessings
of the time. Indeed, it is recommended for inclusion by Menander Rhetor,
which suggests that the sentiment was a commonplace. Now Constantine
uses it, as emperors did in the past, too – for example, Diocletian used it
in his preamble to the Edict on Maximum Prices. It is a topos that is used
by both parties: by emperors to advertise their power over their subjects’
circumstances and by subjects to express their acknowledgment of imperial
power and encourage the emperor to help them. Its function here is as a
structural filler, an element that could be inserted by officials who were
asked to produce texts on the spot.
The text closes with the abbreviation “D,” followed by the consular
date. It is possible that “D” could stand for die; alternatively, the letter
could stand for an administrative procedure – data is a possibility, which
John Matthews believes was inserted wrongly since the text as it stands
would have been not necessarily given to anyone but, rather, included in
the imperial archives. Moreover, constitutions that record the proceed-
ings of audiences are regularly considered edicts and accompanied by the
abbreviation acc., short for accepta.
No reaction from the veterans is recorded. From the beginning of their
audience with the emperor, they had been carefully managed by Constan-
tine’s officials, and the editing of the present text was the final stage in the
operation. Any negative reaction may have been edited out. Admittedly
the veterans’ original list of complaints does not survive, but the privileges
Constantine gave were significant, and so the veterans, with nothing left
 It is found in Plin. Ep. ., Men. Rhet.  and in AE () no.  (on which see below). See
also Hauken : – for this phrase in petitions.
 Matthews : .
 Constantine answers the veterans 

to ask, probably left the camp contented. But even if some of the men still
had complaints, the authors of this text had given the emperor the last
word, and their complaints were forever silenced – a most effective way
of presenting Constantine as both a sympathetic respondent to unhappy
veterans and a powerful ruler in the new style of the later Roman empire.
From the reign of Augustus, emperors bestowed privileges on their
veterans. These privileges concerned mostly financial matters, such as
exemption from tribute, taxes, and compulsory public duties, and bestowal
of political privileges, such as extended voting rights for veterans and
citizenship for their families. These grants of privileges contain shared
language that can be found also in the Constantinian text. For example,
Marcus Aurelius is recorded at a praetorian camp responding to his veterans,
presumably as they complained about their privileges. Concerned that
his veterans faced difficulties marrying, he promised to “seduce them with
gifts” by bestowing on fathers-in-law the privileges that came with being
father of a veteran. He called his men veterani nostri, though according to
Suetonius, Augustus had warned his sons and stepsons against using the
possessive when addressing troops. Constantine would likewise call his
veterans conveterani. This second-century text contains other vocabulary
familiar from the Constantinian text, such as the verbs sollicitare and frui. A
rescript of Diocletian concerning veterans contains significant terms such as
indulgentiae nostrae, fidam, devotionem, and militum nostrorum, all of which
appear in the same or similar form in the Constantinian text. Finally, an
exemplum sacrarum litterarum of  CE concerning soldiers’ and veterans’
privileges exhibits striking linguistic similarities to our text. For example,
indulgentia and beneficium, technical words for a privilege, are included,
as are expressions centering on labor and the enjoyment (fructus/frui) that
should follow it.
Post-Constantinian texts in the Theodosian Code, however, exhibit a
change in tone. While CTh .. and .., dating to the reigns of
Constantius and of Valentinian and Valens respectively, bestow or reaffirm
privileges, the remaining entries contain descriptions of punishments for
individuals, soldiers or civilians, who abuse them. While the code is, of

 See, for example, BGU II , an edict of Octavian, and also Chrest.Wilck. , col. , dating to the
reign of Domitian. See also a letter of Constantine, AE () no. , which is discussed below.
 The text is preserved as Fragmenta Vaticana  in FIRA ii, p. . Campbell (: ) translates
sollicitabimus as “we will seduce with gifts.”
 Suet. Aug. ., in which Suetonius reports that Augustus told them to address soldiers as milites,
not commilitones, which he regarded as ambitiosius.
 AE () no.  = ILS  = FIRA i .
 serena connolly

course, selective, it may be possible to see here a fissure in the relationship

between emperors and the army.
The following texts were entered into the imperial commentarii and
show, as does our text, an emperor, together with his officials, giving an
oral response to a group (or individual) and the resulting response being
edited to suit the conventions of written imperial responses. Yet there are
differences between these and the Constantinian text.
A text dating to the reign of Augustus is among our earliest evidence
of the procedure of imperial hearings. P.Oxy. XXV  contains on the
recto an account of (probably) Germanicus’s arrival at Alexandria and
the rapturous welcome from the people. The text, which may have come
from the Acta Alexandrinorum, is of interest for several reasons. Firstly, the
editor believes that Germanicus’s speech was improvised, in part because
his reaction to the effusive Alexandrians was emotional. Germanicus’s
tendency toward emotional, even hysterical outbursts is well illustrated by
Tacitus’s account of his dealings with the Pannonian mutineers. Secondly,
the editor believes that the text is not a verbatim account and points to
expressions of audience support that do not seem to be motivated by
Germanicus’s words.
If, as seems likely, this text is a copy from an official account, then it is
possible to see both similarities to and differences from the Constantinian
text. Firstly, neither text is a verbatim record of the emperor or imperial
family member’s words. Secondly, not only are the texts not verbatim but
in both cases, the audience is presented as delivering uniform expressions
(a difficult thing to expect any crowd to do). There is, then, a precedent for
the Constantinian text’s presentation of an audience’s words in a stylized
manner.
The differences between the texts are also instructive. Firstly, Germani-
cus may be speaking extempore, but even if this is not the case, he does not
use the highly artificial style of Constantine. Secondly, Germanicus tries to
keep control of his audience, as happened in Constantine’s exchange; but
in the later text, the emperor is not presented as exerting control over the
veterans, probably because his officials were doing this unrecorded. Thirdly,
while admittedly the earlier text is incomplete, the extant papyrus demon-
strates clearly that it does not begin in medias res. It therefore lacks the

 The editor also points to the repetition of the phrase präton m”n, though I am unconvinced
that Germanicus’s use of it three times constitutes “monotonous repetition” (Lobel, Turner, and
Winnington-Ingram : ). It is possible, however, that the other instances of repetition, for
example, of –gÜ pemfqeªv Ëp¼ toÓ patr»v, could support his point.
 These expressions of enthusiastic support are found in lines –, , and .
 Constantine answers the veterans 

elaborate protocol that begins the Constantinian text and includes a record
of salutation and acclamations. Finally, the audience in the Germanicus
text voices standard expressions of praise, which were probably inserted by
an editor. In the Constantinian text, however, the editors have attributed
expressions that are reflections of sentiment but are more ad rem.
In short, this early text demonstrates that some aspects of the Con-
stantinian text show continuity from the earlier principate: editing of the
audience’s words – which we have seen in other recorded hearings and
audiences of the pre-Constantinian period – and attempting to manage
that audience. What is new in the Constantinian text is that the emperor
does not voice his own words but delivers carefully rehearsed pronounce-
ments. Earlier records for the acta show emperors delivering some probably
prepared statements, but these are accompanied by later edited statements
that provide additional explanation. The Constantinian text, in the form
in which it survives today, may contain the emperor’s words as he spoke
them on the day, with only minimal subsequent editing. Another differ-
ence is that Constantine is greeted in a highly ceremonial fashion, unlike
Germanicus, who seems, according to the editors, to have been surprised
by the Alexandrians’ reaction to his arrival. This was not a stage-managed
event. Emperors or members of the imperial family could expect vociferous
and sometimes frank receptions from and exchanges with their subjects.
More similar to the Constantinian text generically is P.Oxy. XLII ,
which records Septimius Severus in Alexandria in a court sitting in judg-
ment with his amici and the officials he had summoned, who were probably
men in the local government. The emperor called in envoys from a delega-
tion of Egyptians and listened to a certain Dionysius’s presentation, which
is recorded in the text. His response followed, at which point unfortunately
the text breaks off. Given its Roman style of dating, this text is most likely a
Greek copy of a Latin original, which presumably belonged to the imperial
commentarii. The protocol is similar to that of the Constantinian text in
that it describes those present at the hearing. Yet the Severan text lacks
the ceremonial aspects of the Constantinian – the titles of the officials, the
salutation, and the acclamation.
 These additional steps in the hearing procedure may have taken place, but the fact that they were
not noted suggests that they were believed neither to be an integral part of the hearing procedure
nor to have importance for underlining the emperor’s role in the proceedings. The editor of P.Oxy.
XLII  notes: “Later examples are more suspect, because their framework is more elaborate and
their speeches more polished; they have even been thought to be entirely fictional.” It is tempting
to believe, given the increasing elaboration of the protocol, that this was so, though there seems to
me insufficient evidence to prove it. Inclusion of the phrase metì Šlla before Dionysius begins his
address indicates that this is not a verbatim account. P.Mich. IX  (= SB XIV ), one of the
 serena connolly

An inscription from Dmeir, Syria, which provides an account of a hear-

ing in consilio before Caracalla, also contains a praescriptum of interest.
The emperor was greeted by his praetorian prefects, who are described as
most eminent, and also by his “friends” and officials. The wording of this
section (cum salutatus a praefectis praetorio eminentissimis viris, item amicis
et principibus officiorum sedisset in auditorio) is remarkably similar to that
in the Constantinian text. There is no chance of post factum editing, since
the text is preserved on an inscription, not in a law code. Also interesting
is the fact that the praescriptum is in Latin, while the actual exchange that
follows is in Greek. From these two facts may be made the suggestion that
the opening is a standard element of acta that had persevered for well over a
century, despite the supposed changes from the somewhat ad hoc consilium
to the more formal consistorium that began either with Diocletian or with
Constantine. But the text that follows shows Caracalla engaging actively
in the role of judge, responding to each point made by the lawyers. The
text of Constantine, on the other hand, is far more ceremonial and, even
when read and unpacked so as to recreate the situation, it seems to reveal
an encounter in which Constantine himself never responds directly and
immediately to what the veterans said – he responds to their utterances
with procedural and ceremonial facilitating statements.
There is a similar opening also in another text from the reign of Caracalla:
Cum salutatus (C insert esset LR) ab Oclatinio Advento et Opellio Macrino praefectis
praetorio clarissimis viris, item amicis et principalibus officiorum et utriusque ordinis
viris et processisset, oblatus est (v.p.e.o. esset scr.) ei Iulianus Licinianus ab Aelio
Ulpiano tunc legato in insulam deportatus, Antoninus Augustus dixit: Restituo te in
integrum provinciae tuae. Et adiecit: Ut autem scias, quid sit in integrum: honoribus
et ordini tuo et omnibus ceteris.
When Antoninus Augustus had been greeted by Oclatinius Adventus and Opellius
Macrinus, who were praetorian prefects and illustrious men, and by his friends
and the most senior of his officials and men of both ranks and had proceeded, and
when Iulianus Licinianus was presented to him, a man who had been exiled to an
island by the then legate Aelius Ulpianus, the emperor said: “I restore you to your
rights.” And he added: “So that you know what is meant by ‘your rights,’ it is to
your offices and rank and everything else.”

apokrimata from Severus and Caracalla’s visit to Egypt in  CE, contains a similar phrase: meqì
™tera, which also suggests that the text preserves only part of the hearing. Perhaps when several
parties in succession approached the emperor for a hearing, the second and later parties found metì
Šlla or meqì ™tera on their copy of the proceedings standing in for the hearing of any earlier party.
 The inscription was first published by Roussel and de Visscher (–). See now Oliver . The
praescriptum is found in lines –.
 CJ ...
 Constantine answers the veterans 

Also of interest in this text is the editing: it contains neither the officials’
presentation of Iulianus Licinianus nor his defense, and the first wording in
the text is Caracalla’s pronouncement Restituo te in integrum provinciae tuae.
It seems likely that something similar to the exchange recorded between
the emperor and the veterans in the Constantinian text before he gave
his response also took place between Caracalla and Iulianus Licinianus,
but it has been omitted from the acta, as happens in most other examples
of this type of text (see below). The Constantinian text is among the few
exceptions and is therefore unusual for the amount of prejudgment material
that is recorded in it. The phrase Ut autem scias . . . et omnibus ceteris, which
seems strangely disconnected from what comes before, was most likely
added either during the audience or immediately afterward. Whichever
is the case (and on the basis of the Constantinian text, the former seems
likely), this additional phrase was surely included in order to make clear
to Iulianus Licinianus what was meant by Caracalla’s first pronouncement.
Caracalla’s first pronouncement enabled; this second provided definition.
In the following text, Diocletian also listens to an exchange between
lawyers:
Imperatores Diocletianus, Maximianus. Pars id. Febr. Inductus Firmino et Apollinario
et ceteris principalibus Antiochensium adstantibus. Sabinus dixit: . . . Diocl.: Certis
dignitatibus data a nobis indulgentia est munerum civilium et personalium, id est
his, qui aut ex protectoribus sunt aut ex praepositis. ii ergo ad munera personalia aut
civilia non vocabuntur.
Emperors Diocletian and Maximianus. Part of the acta,  February. He was led in
as Firminus and Apollinarius and the other senior officials of the Antiochenes stood
by. Sabinus said: . . . Diocletian: “Immunity from civic and personal obligations
has been granted by Us to certain persons of distinction, that is to those who are
former members of the corps d’elite or chamberlains. These men, therefore, will
not be called to personal or civic obligations.”
This text, part of the acta from Antioch as the word pars suggests, provides
evidence in the word adstantibus of the change from the consilium to the
consistorium. Yet the structure of the imperial answer has not changed. A
certain Sabinus, for himself or on behalf of others, probably asked Dio-
cletian for exemption from certain munera, which the emperor bestowed.

 A second stage of editing came probably later and it is revealed by tunc: Aelius Ulpianus was at that
time a legate who had been responsible for deporting Licinianus. Copying of this text must have
happened soon after the event, while Ulpianus was alive but no longer legate, and a conscientious
court official wished to make sure his audience knew of Ulpianus’s change in office (for better or
worse).
 CJ ...
 serena connolly

Diocletian’s answer, like those of Caracalla, came in two parts: an initial

statement that enabled the exemption and a subsequent statement that
added further detail. The Constantinian text, while much longer than
those just discussed, follows a similar format. An opening protocol sets the
scene and, though it is not stated, identifies the text as belonging to the
acta. The recording of the actual exchange is not new – the texts of
Caracalla and Diocletian also record the to-and-fro between emperors and
their subjects – but the artificiality seems to be an innovation. Moreover,
Constantine’s response can also be divided into two basic elements, as was
discussed above.
Some later texts preserved from the acta do not preserve the proto-
cols and ceremony we find in the Constantinian text. For example, CTh
..pr. opens simply with Pars actorum habitorum in consistorio Gratiani
Augusti, and CTh .. with Pars actorum habitorum aput imperatorem
Iulianum Augustum Mamertino et Nevitta Conss. x kal. April. Constantinop-
oli in consistorio: adstante Iovio viro clarissimo quaestore, Anatolio magistro
officiorum, Felice comite sacrarum largitionum. Et cetera. Finally, CTh ..
begins, Pars actorum habitorum in consistorio aput imperatores Gratianum,
Valentinianum et Theodosium cons. Syagri et Eucheri die iii kal. Iul. Con-
stantinopoli. In consistorio.
Two factors may explain the brevity of these introductions: first, they
were produced at regular meetings of the consistory at imperial residences,
not in the provinces, where local dignitaries would also be present. Second,
while the acta may originally have recorded hearings more fully in the
mold of the Constantinian text, only part of them has been preserved and
circulated, as was the practice during and before the fourth century. For
example, in CTh .. there is no exchange, but the emperor simply
announces episcopus nec honore nec legibus ad testimonium flagitatur. The
phrase idem dixit gives notice of a second part of the response: episcopum
ad testimonium dicendum admitti non decet, nam et persona dehonoratur
et dignitas sacerdotis excepta confunditur. This time, the purpose of the
additional statement in this text is to provide justification, rather than
detail for the first enabling statement. Likewise, in CJ .., dating to
the reign of Theodosius I, the initial response is in omni cessione bonorum

 The response was produced in three forms: one was a pragmatica sanctio, an abbreviated text sent
to the treasury, perhaps containing simply the emperor’s response (the opening sections of our text
would have been unnecessary for the treasury to enforce the emperor’s decision), which was referred
to as an epistula in our text; another was an adnotatio, which was likewise abbreviated, given to the
veterans; and lastly there is the text we have before us, the text written down and preserved apud
acta.
 Constantine answers the veterans 

ex qualibet causa facienda scrupulositate priorum legum explosa professio sola

quaerenda est, while the second is briefer: in omni cessione sufficit voluntatis
sola professio. The second part of the response is not a clarification but,
rather, a restatement in more legally professional and accurate words.
This necessarily brief overview of texts that record imperial encounters or
hearings at which pronouncements were made or dispensations were given
suggests the following developments: first, these events may have become
more formal and controlled; and, second, the textual records of these events
show signs of editing. These may have been more heavily edited over time,
but they did not become more elaborate. Indeed, detailed protocols seem
to have become less common.
How do the Constantinian text and the encounter it records fit into
these developments? Considering the nature of the hearing recorded in it,
CTh .. stands on the cusp between the principate and the later Roman
empire: before it came hearings before emperors and public exchanges
between emperors and their subjects that were characterized by less formal,
more free-form reactions and exchanges and greater and more genuine
verbal to-and-fro. The encounter between Constantine and his veterans was
more stage-managed and was not just a hearing, but now a ceremony, and
was styled by a scribe even more so. Yet the form of the text that records it is
an anomaly, but an insightful anomaly, nonetheless, and while it expresses
so well the changed nature of the imperial office from the start of the fourth
century, texts that date to later in the fourth century look similar to those
of earlier periods. Given the nature of that text and the increase in court
ceremony so brilliantly described by Ammianus Marcellinus, the hearings
from which the later texts were produced may have had all the ceremony
of CTh .., but it was simply not recorded. The Constantinian text
demonstrates that into the fourth century and beyond the protocol that
accompanied legal hearings may have grown more elaborate, yet the style
of legal texts that recorded them did not develop in parallel.
Jill Harries in this volume observes that during the long fourth century
“the editors [of the various codes] became the legislators.” Indeed, they
were not only legislators, but in the case of the Constantinian text, whose

 Interestingly, the entry from this text as it is preserved in CTh .. is shorter: Imp. Theodosius A.
dixit: in omni cessione professio sola quaerenda (requirenda B) est. Idem dixit: in omni cessione sufficit
voluntatis sola professio. Given that the compilers of Justinian’s Code went back to the original acta
for their text (and indeed there are very many occasions on which the compilers eschewed the
Theodosian text of a constitution in favor of an earlier version), we see that the texts recorded in
the CTh were not exact copies of the records of proceedings, which themselves were not verbatim
accounts.
 serena connolly

opening section (everything up to and including the emperor’s dispensa-

tion) they chose to retain, they played a crucial role in determining how
emperors were viewed. Though its form is anomalous, it is a programmatic
text for Constantine’s reign: it continues much of the form that preceded
and responds to political realities, but the text is also imbued with ritual,
and it is fascinating to see the resulting struggle to superimpose a new
formality on top of the existing mode of interaction. While the form of
this text is not found repeated in a law code, the conduct of the hear-
ing, and its emphasis on acclamation and carefully controlled interaction,
was to continue through the long fourth century. Finally, much effort was
spent on the encounter and the recording of it, which surely reflects the
importance of the military to Constantine, a fact that Harries identifies at
the start of her essay. This is a text that set the tone for a new reign and a
new period.
 part ii

Biography and panegyrics

 chapter 6

Three generations of Christian

philosophical biography
Edward Watts

The famous Symeon, the great wonder of the world, is known by all
the subjects of the Roman Empire and has also been heard of by the
Persians, the Medes, the Ethiopians; and the rapid spread of his fame
as far as the nomadic Scythians has taught his love of labor and his
philosophy. (Theodoret, Religious History, .)
The monastic saints, who excite only the contempt and pity of a
philosopher, were respected, and almost adored, by the prince and
people. Successive crowds of pilgrims from Gaul and India saluted
the divine pillar of Simeon; the tribes of Saracens disputed in arms the
honour of his benediction, the queens of Arabia and Persia gratefully
confessed his supernatural virtue; and the angelic Hermit was con-
sulted by the younger Theodosius in the most important concerns of
the church and state.
(Edward Gibbon, Decline and Fall of the Roman Empire, .)

Both of these passages refer to the same person, the famous Syrian ascetic
Simeon Stylites. The first, the opening sentence of Theodoret’s biography
of Simeon, marks him as a powerful and famous philosopher whose physical
discipline defined his philosophical achievements. The second, Gibbon’s
description of Simeon and ascetics like him in his Decline and Fall of the
Roman Empire, places the saint in a category of men “who excite only the
contempt and pity of a philosopher” and whose labors can be dismissed as
perverse manifestations of misplaced social and religious priorities. These
categorizations of the same man highlight two fundamentally different

An earlier version of this paper was delivered at Brown University. I thank the audience there for
its feedback. A debt of gratitude is also owed to Joseph Pucci, Susan Harvey, Cristiana Sogno,
Scott McGill, and Samuel Rubenson for their comments and suggestions. This paper would not
be possible without the extreme indulgence of John Matthews, who had the vision (or at least the
forbearance) to recognize ancient biography as an acceptable field for an oral examination.
 For Theodoret’s Simeon the most important study remains that of Ashbrook Harvey , esp.
–. See, as well, Price : xiv; Leroy-Mohlinghen and Canivet –, vol. : –; and
Urbainczyk : –.


 edward watts

views of what philosophy was in late antiquity. Writing in the early s,
Theodoret thought that the later Roman world had produced a new breed
of philosopher who championed the cultural attitudes of a new and excit-
ing Christian future. His work sought to celebrate these figures for the
cultural and spiritual leadership they provided and for “that highest philos-
ophy” that they practiced. Though Gibbon wrote nearly , years after
Theodoret, he accurately describes an attitude held by many late antique
thinkers. Like Gibbon, pagan philosophers like Eunapius and Damascius
scorned the “impious men in black raiment” whose actions brought dis-
aster to the culture of the classical world and whose presence symbolized
its descent into a new Typhonian age.
This confusion over the nature of “philosophy” suggests a fundamental
discontinuity between the world of the tetrarchy and the Theodosian age.
In the late third century, an almost uniformly pagan class of philosophers
carried on the cultural and intellectual traditions of their classical forebears.
By the Theodosian period, however, authors had stretched the category of
philosopher to include not only Platonists and their ilk but the solitary
ascetics of the Egyptian desert and a Syrian rustic perched atop a pillar.
To some observers, both ancient and modern, Theodoret’s effort to rec-
ognize eccentric Syrian Christians as philosophers could seem emblematic
of a sort of cultural debasement through which Christians progressively
marginalized fixtures of classical society in order to replace them with
Christian facsimiles that bore the same names. Fortunately, late antique
cultural history does not follow so starkly defined a path. Theodoret’s use
of “philosophy” to describe Simeon’s ascetic practices may seem like an
ironic nod to an outdated cultural iconography, but if one resists the urge
to dismiss this identification outright, Theodoret’s work instead suggests
that the meaning of philosophy became much more flexible and the cat-
egory of philosopher much more expansive as Christianity’s influence on
late antique culture increased.
The fundamental cultural continuity joining the tetrarchy with the
Theodosian age has been demonstrated in many different ways, most
notably in studies examining grammatical and rhetorical instruction and

 Although now a part of Theodoret’s Religious History, the Life of Symeon was originally published
separately, sometime between  and . For discussion of the merits of the two dates, see
Leroy-Mohlinghen ; Leroy-Mohlinghen and Canivet –, vol. : –; Price : xiv.
 Theodoret, RH .; cf. prologue : “We describe a life which is a lesson in philosophy and aspires
to the way of life in the heavens.”
 Eunapius, Vit. Soph. .
 On the new Typhonian age (framed somewhat differently), see Damacius, Vit. Is.  A.
 Three generations of Christian philosophical biography 

practice. At the same time, scholars hesitate to see the Christian adap-
tation of philosophical culture in the same way as, say, the Christian
use of rhetoric. While no one has particular difficulty placing a Christian
rhetorician like Procopius of Gaza or John Chrysostom in the same cultural
category as Libanius or Himerius, no modern scholar would group the so-
called Christian philosophers Antony and Simeon Stylites with Iamblichus
and Proclus. And yet numerous fourth- and fifth-century sources argue
that we should make precisely this equation.
This paper examines the development of philosophical biography, the
type of literature in which this argument most commonly appears. It shows
that, as Christianity’s interaction with philosophy evolved, so too did the
ways in which Christians engaged with this particular sort of philosoph-
ical literature to argue for new, more expansive definitions of philosophy
and philosophers. Such a study requires a couple of important qualifiers,
however. First, philosophical biography is a contrived literary genre – if
one can even call it a distinct genre at all. In antiquity, there were nei-
ther established rules for the composition of a philosophical biography nor
guidelines that its author needed to follow. While they could be different
in form, these texts shared similar functions and rhetorical approaches.
Philosophical biographies may seem like narratives, but they were primar-
ily argumentative texts designed to convince readers of the power of a
particular thinker and his brand of thought. They do this by profiling his
life and demonstrating how his actions illustrated his teaching. His deeds
then prove the power of his ideas. Whatever their idiosyncrasies, these
texts share a common rhetorical approach that uses a narrative to present
a philosophical (or quasi-philosophical) argument.
This leads to the second qualification. In the past two decades, there has
been a growing consensus that the focus of these sorts of texts shifted in the

 Among the most notable studies making this case are Kaster  and Kennedy .
 For the creation of a sort of hybridized Christian and classical rhetoric, the discussions of the
so-called “Third Sophistic” are useful. On this see Quiroga , as well as the important volume
of Amato, Roduit, and Steinrück .
 Intriguingly, few resist including earlier Christian “philosophers” like Justin and Origen in this
category. This has more to do with their relatively conventional behavior and training than it does
with their process of self-definition. For the Apologists and the manner in which they engaged
with both philosophical notions of communal definition and arguments for pagan philosophy’s
dependence upon Judeo-Christian thought, see Boys-Stone : –.
 Cox : –; echoed by Rubenson : .
 This study focuses upon individual philosophical biographies, but a similar function can be seen in
collective biographies as well. On this logic in the collective biography of Eunapius, for example,
see Watts . On collective biographies and their logic of inclusion see more broadly Cox Miller
.
 edward watts

second or third century CE. From that point forward, authors began to
concentrate on demonstrating how their subject was godlike or even divine,
instead of focusing primarily upon his philosophical significance. This
development is thought to transform these texts from literary weapons
deployed in scholastic rivalries into works that argue for specific cultic
practices. However, the late antique emphasis upon the spiritual signifi-
cance of philosophers need not indicate a shift in the fundamental function
of these texts. Divine inspiration had formed a part of the argumentation
used in philosophical biographies since their beginning in the classical
period. Divine lineage was long a feature of Pythagorean biographical
traditions; it seems to have become part of the biographical portrait of
Empedocles in the fourth century BCE (if not before), and it may even
have been claimed by Heracleides of Pontus during his lifetime. Divine
inspiration famously played a role in both Plato’s and Xenophon’s versions
of Socrates. Indeed, even Plato’s successor, Speusippus, once argued that his
uncle Plato was the offspring of Apollo, going so far as to describe him as
the product of a virgin birth. While this tendency was more pronounced
in late antiquity, classical and Hellenistic biographers did not hesitate to
advertise the divine lineage of the philosophers they profiled. These divine
links were welcome, in part, because they helped to persuade an audience
of the power of a philosopher’s teaching.
With these caveats noted, we can begin to think about the origins of
these texts. Their roots stretch at least to the coalescence of the Platonic
Academy around Xenocrates in the fourth century BCE. Xenocrates, a
middling philosopher, won control of the Academy from brilliant thinkers

 Cox : .  Ibid.; Rubenson : .  Cox : .
 A fact that Cox acknowledges and interrogates at some length (: –). Nevertheless, she sees
a fundamental shift in late antiquity that is tied to a new sort of function for these texts. If one
considers the desired rhetorical effect of such ideas, however, this need not represent a new addition
but simply an adaptation of an established rhetoric to a new social and cultural context.
 Iamblichus, Pyth. ; Porphyry, Vit. Pyth.  (note, however, that Porphyry prefers more “rational”
explanations). Most of these later sources likely draw upon Aristoxenus’s fourth-century-BCE profile
of Pythagoras, which may itself draw upon earlier traditions.
 Diogenes Laertius . = Heracleides of Pontus, fr. . Note here Tiede : –; and Cox :
.
 Various traditions describing Heraclides’s attempts to have himself recognized as a blessed figure
are found in Diog. Laert. .–; cf. Heraclides, fr. a,  (Wehrli). The reliability of this can be
questioned, however (e.g., Wehrli : –; Dillon :  n. ).
 “Speusippus in the work ‘Plato’s Funeral Feast’ and Clearchus in ‘The Encomium of Plato’ and
Anaxilaides in the second book of his ‘On the Philosophers’ say that there was a story in Athens
that, when Perictione was in the bloom of her youth, Ariston tried to overpower her but did not
succeed. Stopping the assault, he saw the face of Apollo, for which reason he guarded the purity of
his wife until a child was born.” (Diogenes Laertius, .)
 For this process see Watts .
 Three generations of Christian philosophical biography 

like Aristotle and Heraclides of Pontus because his ability to live modestly
and follow his own ethical doctrines distinguished him from rivals and
compensated for his limitations as a thinker. Academic propaganda then
began to feature stories illustrating the great personal virtue of the Academic
leadership and other tales describing the redemptive power of Academic
teaching. Rivals responded by attacking the biographical traditions at
the core of Academic propaganda. They attacked the modest intellect of
Xenocrates, the questionable behavior of his successor Polemo, and the
personal characteristics of Plato and Socrates. As early as the fourth cen-
tury BCE, it was understood that a philosopher’s deeds could be marshaled
to either reinforce or undermine the power of his teaching.
Because of their ability to advertise the benefits of a particular brand of
teaching, philosophical biographies proved to be extremely effective teach-
ing tools. As the Hellenistic world gave way to the Roman empire, biogra-
phies became an essential part of the prolegomena given to new students
of philosophy. So, for example, a biography of Plato began the course of
study outlined by the anonymous Prolegomena to Platonic Philosophy. Late
antique authors also used biographies to introduce readers to authorized
collections of a philosopher’s works and convince them of the importance
of these texts. The reason is not difficult to understand. Learning about
the life of a philosopher gave immediate meaning to his ideas and showed
a new student the wonderful things that could be achieved through dili-
gent study and active membership in a philosophical community. These
texts, both instructive and protreptic, seduced the student and fired him or
 Philodemus, Hist. Acad. VI–VII = Isnardi Parente, Senocrate, fr. .– = Tarán, Speusippus, Test.
.–. Note as well on this passage the discussion of Gaiser : –; and Dillon : –
on the voting procedures in the Academy.
 Watts : –.
 On Xenocrates’s clumsiness see Diog. Laert. . and Plutarch, Coniug. praecept. . = Isnardi
Parente, Senocrate, fr. . On the charges of stupidity, see Plutarch, De recta ratione audiendi .
e = Isnardi Parente, Senocrate, fr. . For discussion see Isnardi Parente : –. For criticism
of Polemo, see Diogenes Laertius’s account of his early life (Diog. Laert. .). Note as well the
account of Philodemus, Hist. Acad. IV–XIII, and the discussion of Dillon :–.
 Aristoxenus, fr.  (Wehrli). Cf. Riginos :  n. . For discussion, see Watts : –.
 Aristoxenus, fr. a–b (Wehrli) = Cyril, Contra Julianum VI = Theodoret, Graec. affect. curatio
XII . Cf. Aristoxenus, fr.  = Athenaeus .d and Aristoxenus, fr.  = Plutarch. Aristides .
For discussion, see Watts : .
 For another way in which the presentation of individual lives could be used as a mode of attack, see
as well the contribution of Susanna Elm in this volume, chap. .
 For introductory discussions see Mansfeld  and, less directly, Mansfeld .
 Westerink : chaps. –.
 The best example of this is Porphyry’s Life of Plotinus. Note that the full title of the text is On the
Life of Plotinus and the Arrangement of His Works. For discussion of the importance of this feature,
see Edwards : . A similar function can be supposed for Marinus’s Life of Proclus, a text that
similarly came to be attached to Proclan manuscripts.
 edward watts

her with an enthusiasm for the difficult, time-consuming, and expensive

intellectual undertaking that philosophical study had become.
The most persuasive philosophical biographies share a similar narrative
structure. They begin with the birth and early childhood of the philoso-
pher. They then catalog his (always quite extensive) education and early
career, often dwelling on his various teachers and the exotic places he stud-
ied; next they show how, after the philosopher completed his education, he
then applied his learning in the real world, with invariably extraordinary
results. Only after a thinker’s importance has been demonstrated by his
educational pedigree and his early deeds does the biographer begin any pre-
sentation of his actual ideas. Incidentally, this is when most of the variety
in philosophical biographies begins to appear. Finally, the biography often
concludes by describing the philosopher’s death in a way that reaffirms
the inspired nature of this teaching. Its entire narrative worked to make
readers predisposed to respond positively to the teachings of a particular
school or thinker.
Iamblichus’s On the Pythagorean Life, an early-fourth-century text writ-
ten as a sort of philosophical primer, shows the power of this sort of
argumentation. In this work, Iamblichus presents Pythagoras and his phi-
losophy as divinely inspired. It begins with an invocation to this effect.
The text then describes Pythagoras’s birth to a notable Samian family
and alludes to the commonly-held belief that Pythagoras was actually the
child of Apollo. Pythagoras’s father supervised his initial education, taking
him to a range of teachers. While still a teenager, Pythagoras’s fame as a
wise man spread beyond his native city and the greatest Greek thinkers of
his time – including Anaximander and Thales – welcomed him as their
student. After learning all that they could offer, Pythagoras sought out
Egyptian, Phoenician, and Babylonian sages in order to develop an ecu-
menical and all-encompassing wisdom. Iamblichus then describes the

 In the past, much attention has been paid to possible structural similarities between biographies,
going back to the monumental (though somewhat problematic) work of Leo . Another impor-
tant, though at times equally problematic, contribution is that of Stuart . Note in this context
the important criticisms of Cox : –. I do not here aim to propose a new structural model;
I only hope to note basic similarities in the way that personal details were organized in works that
drew upon the rhetoric of philosophical biographies.
 The importance of this can be seen in the rhetorical gymnastics that Porphyry and Marinus feel the
need to perform in order to explain away the inconveniently unpleasant deaths of their mentors.
 “All right minded people, embarking on any study of philosophy, invoke a god. This is especially
fitting for the philosophy which takes its name from the divine Pythagoras (a title well deserved)
since it was originally handed down from the gods . . . and after the gods, we shall take as our guide
the founder and father of this divine philosophy” (Iamblichus, Pyth. , trans. Clark).
 Iamblichus, Pyth. –.
 Three generations of Christian philosophical biography 

effect of Pythagorean ideas on his fellow Samians and, in the work’s sev-
enth chapter, the ways in which Pythagorean teaching liberated a number of
occupied Italian cities and led them to excellent government. The eighth
chapter describes how Pythagoras’s powerful gift of foresight, a product of
his unparalleled wisdom, convinced contemporaries to accept Pythagorean
teaching. Iamblichus recounts an incident in which Pythagoras told a group
of fishermen the exact number of fish they were pulling out of the water
before the net broke the surface. Amazed at this, the fishermen and the rest
of the city then listened eagerly as Pythagoras taught.
This anecdote represents a point of transition in Iamblichus’s narrative.
After describing this wonder, Iamblichus finally allows his audience to learn
something about what Pythagoras actually taught. The buildup has been
deliberate. It has allowed Iamblichus to define Pythagoras as an inspired
figure who has learned all that there is to know of Greek, Egyptian, Phoeni-
cian, and Babylonian wisdom. His ideas can resolve disputes, produce effec-
tive government where it had been lacking, and even allow an initiate to
develop a superhuman perceptive ability. At this point, Iamblichus expects
that his audience has become convinced of the importance of Pythagorean
teaching. They are now prepared to learn its substance.
The rest of Iamblichus’s text mixes Pythagorean doctrines and illustrative
anecdotes in such a way that the deeds of the sage (and some of his followers)
reinforce the power of his teaching. This focus holds all the way until his
description of the deaths of Pythagoras’s students, many of whom died
as martyrs for his philosophy. The most notable example concerns a
group of Pythagoreans who were ambushed by a detachment of Syracusan
hoplites. They had nearly outrun the soldiers when they reached a bean
field. This would, of course, not be a real problem for a normal person – but
these were Pythagoreans and Pythagoras had taught his followers never to
touch beans. Remembering their master’s injunction, the Pythagoreans
turned and offered themselves up as willing victims.
The deaths of these Pythagoreans beside a bean field had profound
significance within their community, but it probably just seemed silly
to most outsiders. This shows an important limitation to philosophical
biographies. Their arguments can be extremely compelling, but only if an
audience accepts two basic points. The audience has to believe it possible
that the philosopher being profiled could have found true wisdom. It also
has to accept that this wisdom could produce the sorts of wonders the text
describes. If a reader accepts that Pythagoras was an inspired teacher, he

 Ibid. .  Ibid. , .  Ibid. .  E.g., Diogenes Laertius, .–.
 edward watts

would be touched by the sincere devotion of those of his followers who

accepted martyrdom in defense of his teaching. If a reader does not believe
this, the Pythagorean bean-field martyrs are no more than peculiarities.
This is where the Christian reaction to pagan philosophical biographies
becomes so interesting. One would expect that Christians would have little
use for such texts. Christian authors could mine the lives of philosophers
for gems of intellectual history, but one would not assume their interest
in the larger arguments that these texts made. Indeed, many late antique
authors treated pagan philosophical biography as Marcian of Gaza did
pagan poetry. Aeneas of Gaza strip-mined Philostratus’s Life of Apollonius
of Tyana for anthropological material about India. And the unlikely
bedfellows Cyril of Alexandria and Theodoret both raided Aristoxenus
for scurrilous portraits of Socrates. But the Christian reaction to these
texts was much more complicated than this borrowing suggests. Instead of
turning away from philosophical biography as an obsolete cultural relic,
Christian authors sought to transform its argumentation into something
that set out models for a Christianized late antique philosophy.
One can see this process at work in three important texts: Eusebius’s Life
of Origen, Athanasius’s Life of Antony, and Augustine’s Confessions. Together,
these works form a chronological chain linking the tetrarchy and the
Theodosian period. Each of these is often seen as a revolutionary work that
charted new, Christian philosophical paths. When looked at more closely,
however, they reveal themselves to be the products of complicated cultural
negotiations that accompanied the growth of Christian social power and
self-confidence. Their revolutionary force lay less in their complete newness
than in their compelling reinvention of classical modes of expression.
Although not without idiosyncrasies, Eusebius’s Origen is the most
straightforward of the three. The biography makes up much of the sixth
book of Eusebius’s Ecclesiastical History and traces the life of the Christian
teacher Origen from his childhood through to his martyrdom. Once one
moves beyond his decision to embed this biography in a larger work of
Christian history, Eusebius’s dependence on the structure and basic rhetoric
of philosophical biography becomes immediately apparent. He begins with
 Bishop Marcian of Gaza was said by his encomiast Choricius to “cull from poetry whatever was
useful while smiling at the myths” (Choricius, Laud. Marc. ...–).
 Th. .. The passage he alludes to is Philostratus, Vita Apollonii ..
 E.g., Aristoxenus, fr. a–b (Wehrli)= Cyril, Contra Julianum VI; Theodoret, Graec. affect. curatio
XII .
 For a more general treatment of the relationship between ancient biography and Christian hagiog-
raphy, see van Uytfanghe .
 Easily the best and most comprehensive treatment of this text is Cox : –.
 Three generations of Christian philosophical biography 

Origen in “swaddling clothes” trying to offer himself up for martyrdom

before being pulled back by his mother. His education included “training
in the Divine Scriptures from childhood . . . for his father, besides giving
him the usual liberal education, had made them a matter of no secondary
importance.” Throughout his early life Origen balanced traditional Hel-
lenic paideia with Scripture, and in late adolescence he began offering ele-
ments of a classical education (probably grammar) to students. Eusebius
says that Origen also took it upon himself to give instruction in Christianity
and, despite his youth, converted some learned pagans. At eighteen, he
was judged talented enough to preside over the Alexandrian Catechetical
School. After describing Origen’s background and education, Eusebius
then turned to the effect that Origen had on others. Drawing upon the
most profound illustration of virtue available to a Christian author, he
listed those students of Origen who had become martyrs.
From this point, Eusebius’s attention wanders away from his portrait of
Origen and he returns only intermittently to him, but one incident in his
presentation of the mature Origen particularly stands out. Eusebius found
it necessary to dispose of the claim made by the philosopher Porphyry
that Origen was born of pagan parents, educated in classical literature,
and, ultimately, chose to turn away from this learning in order to embrace
Christianity. It is only after dismissing this criticism – and the distinc-
tion between Christianity and philosophy that it implied – that Eusebius
describes the nature of Origen’s thought. Eusebius does this largely by
cataloging his writings, a method that he had used effectively and to a
much greater extent in his Apology for Origen. The work concludes with
two significant narrative strands. First, Eusebius shows Origen intervening
in two theological disputes in ways that bring the conflicting parties into
agreement. He then describes Origen’s martyrdom, the crowning glory
of his life of Christian philosophy.
Eusebius marked Origen as a superior philosopher whose learning sur-
passed others because he added the true wisdom of Christianity to a
 Eusebius, HE ..–.  Ibid. ...
 Ibid. ... That he taught grammar seems implied by HE ...  Ibid. ..–.
 Ibid. ... For a survey of opinions about the nature of the Catechetical School, see Boulluec ;
Scholten : –; van den Broeck ; and van den Hoek .
 Ibid. ..
 Ibid. .. For discussion of this passage see Cox : –; Watts : –.
 Indeed, chs. – of this work consist essentially of presentations of Origen’s ideas through
reference to his written work.
 Eusebius, HE ., . For this sort of intervention, note especially the treatment of Galvão-Sobrinho
.
 HE .; ..
 edward watts

conventional philosophical training. His adaptation of the rhetoric of pagan

philosophical biography to a Christian subject helped him to underline this
point, but Eusebius also apparently sensed that the conventions of such
a text could prove an inexact fit for Origen. Indeed, Eusebius’s desire to
sketch out a profile of Christian–philosophical hybridity explains why he
goes to such pains to describe Origen’s Hellenic learning and spends so
much time arguing against Porphyry’s claim that Origen had turned away
from conventional paideia. For Eusebius’s model to work, Origen needed
to be an adept of both Christianity and conventional philosophy.
The next generation of Christians found Eusebius’s hopeful portrait of
cultural hybridity less desirable. Not only had Constantine ushered Chris-
tianity self-confidently onto the Roman political stage, but, in the minds of
thinkers like Athanasius of Alexandria, the cultural wedding of Christianity
and Hellenism had produced the unruly child of Arianism, a movement
that threatened to tear the Christian empire in two. This new world gave
birth to Athanasius’s Life of Antony, a work that described, in narrative
form, the education, conduct, and lifestyle of Antony, the prototypical
Egyptian solitary ascetic. In keeping with Athanasius’s own theological
and political purposes, his Antony was an ascetic who deferred to bishops
and participated in the fight against Arianism. The most important ele-
ment of the portrait, though, was Athanasius’s identification of Antony as
a Christian philosopher who demonstrated the irrelevance of conventional
intellectual approaches to an exemplary Christian life.
Like earlier textual relatives, the Life relies upon the interplay of word
and deed to convincingly portray Antony as a teacher of a philosophically
viable brand of wisdom. So, in the same way that Iamblichus used anec-
dotes to validate the training of Pythagoras, Athanasius includes similar
validating details about Antony’s ascetic education. This was a particularly
important literary strategy in Athanasius’s text because Antony’s “educa-
tion” was unique for a philosopher. Whereas conventional philosophers
(both Christian and pagan) had extensive literary training, Antony, we are
told, “could not bear to learn letters.” Instead, he embarked upon his

 For “Hellenism” as a cultural and religious identifier in late antiquity, see, for example, the discus-
sions of Bowersock a; Lyman ; and Elm : –.
 For Arianism in this light see Williams ; Brakke : –.
 For a different perspective on the interaction of Platonic philosophy and Christian asceticism, see
Siniossoglou : –.
 As Samuel Rubenson has recently argued, Athanasius even seems to have modeled his narrative on a
biography of Pythagoras (). This idea has its roots in the twentieth-century studies of Priessnig
( and ). For discussion of the approach of Priessnig, see Cox : –.
 Vit. Ant. .
 Three generations of Christian philosophical biography 

search for Christian wisdom when he heard words from Scripture read
aloud in church. He advanced in learning not at the foot of an erudite
mentor but by watching the actions of older ascetics who dwelled on the
outskirts of his town. After learning all that he could from one of these
men, Antony would move on to another. After “gathering the attributes of
each in himself,” Antony then set out to develop his asceticism by living
on his own. Later, he moved to the desert fringes and, eventually, into
the desert itself.
As recounted by Athanasius, Antony’s specific path could not have been
more different from that followed by previous generations of wise men, even
Christian wise men like Origen. But his education had the same basic stages.
Antony, like Pythagoras, developed a type of ecumenical wisdom that drew
upon the teaching of many masters. And, like Iamblichus, Athanasius is
clear that Antony’s training was effective. When Antony emerged in public
after completing his solitary ascetic formation, “the Lord worked through
him” to heal the sick, purge demons, reconcile those hostile to one another,
and even make a crocodile-infested canal safe to wade across. It is only
then, after these proofs of Antony’s wisdom are given, that Athanasius
begins to lay out Antony’s doctrines on the acquisition and practice of
ascetic wisdom.
After this initial introduction to Antony’s philosophy, Athanasius’s
account becomes a series of vignettes designed to show the exceptional
abilities that this new sort of wisdom had given Antony. Among the most
memorable of these are three extended exchanges that Antony had with
pagan philosophers. These exchanges underpin Athanasius’s identifica-
tion of Antony as a philosopher with a superior sort of wisdom. In the
conversations that the text presents, philosophers were drawn to Antony
because they thought that, as a simple-minded Christian, Antony would
be easy to defeat in an argument. When they actually spoke with him,
however, Antony outwitted the philosophers with the simplicity of his
responses. In the third conversation, Athanasius has Antony turn back the
attack of the Hellenic philosophers and then shift the discussion to one
evaluating the relative power of Hellenic and Christian philosophy. Based
upon faith in an all-powerful God, Christian philosophy overwhelms the
syllogisms of philosophers and shows its great power through healings and
conversions that Hellenes cannot match. Athanasius takes pains to show
that, even when the encounter took place on terms set by conventional

 Ibid. , . On this apparent reference to Deuteronomy ., see Rousseau : .
 Vit. Ant. –.  Ibid. .  Ibid. –.  Ibid. –.  Ibid. –.
 edward watts

philosophers, Antony’s philosophy of asceticism was shown to be superior

to that practiced by Hellenic thinkers.
In the end, the educated reader of the Life of Antony was to take away a
revolutionary message. The Christian asceticism of Antony had rendered
traditional philosophy, and those teachers trained in it, irrelevant to the
search for true wisdom. This was powerful rhetoric, but it obscures some
of the significance of Athanasius’s project. Despite his attempts to present
Antony as a better alternative to the traditional philosopher, it is equally
clear that Athanasius wanted his readers to understand that Antony was
a philosopher devoted to a new type of philosophy that derived from the
purity of Scripture and clarity of faith.
Understanding the Life of Antony as a philosophical biography helps
to explain some of the great resonance that the work had in the decades
following its publication. Like other philosophical biographies, the Life
served as a prolegomenon to the study of a particular type of philosophy,
and (ostensibly at least) it seems to have been pitched to prospective stu-
dents of ascetic philosophy in much the same way that beginning Platonists
absorbed lives of Plato. This aim becomes clear in Athanasius’s prologue,
when he directs the text to monks abroad in order to convince them to
follow the ascetic example of Antony. We can also see the text performing
this function beyond the confines of an ascetic environment. Augustine, for
example, famously describes two imperial officials who decided to aban-
don their careers at court to pursue an ascetic life after reading the Life
of Antony. If, as its rhetoric suggests, we read the Life of Antony as a
Christian philosophical biography this sort of reader response begins to
make more sense. Augustine’s imperial officials had responded to the Life
exactly as they should – and exactly as previous generations of philosophy
students had responded to similar texts. The post-Constantinian age gave
Christians the confidence to abandon the Eusebian model of elite cultural
hybridity and chart a purely Christian philosophical path, but the ancient
rhetoric of philosophical biography remained one of the most potent tools
available to set novices along this new course.

 “Since you asked of me about the way of life of the blessed Antony . . . with the result that you
may bring yourselves to imitate him, I very eagerly accepted your request . . . And I know that you,
when you have heard, in addition to your amazement at the man, you will wish also to emulate
his determination. Indeed, for monks, the life of Antony is a sufficient pattern for asceticism” (Vit.
Ant. prologue).
 These men were described by Augustine’s friend Ponticianus (Conf. .). They came across the Life
of Antony at Trier while waiting for the emperor to return from the games and decided to leave
behind their lives at court.
 Three generations of Christian philosophical biography 

Athanasius’s Antony charts this new cultural direction, but Augustine’s

Confessions serves as perhaps late antiquity’s most complicated and most
interesting adaptation of philosophical-biographical rhetoric. As is well
known, the Confessions chronicle Augustine’s journey away from and back
to God. The Confessions is not primarily a narrative of Augustine’s life, but
its narrative elements truly distinguish the text. They represent a natural
point of entry for readers who, by design, would have to return to the
complicated text multiple times to unpack its full meaning. The basic
outline of this familiar narrative is worth briefly summarizing. The Con-
fessions begins with a prayer and then describes Augustine’s childhood. He
was born to a good family, excelled in school, and won rhetorical contests.
At age seventeen, he moved from the schools of his home town of Thagaste
to Carthage, the intellectual center of North Africa. By age twenty-one, he
had begun teaching. By age twenty-nine, Augustine moved on to teach
in Rome. One year later, at the extremely young age of thirty, Augustine
won a chair of rhetoric in Milan, the de facto imperial capital. To put this
in context, Augustine’s older contemporary Libanius was judged a remark-
able prodigy when he was offered a similar position in Athens at roughly
the same age. Augustine rose so fast that he even contracted a marriage
with the daughter of a prominent courtier – a clear sign that this son of
a middling North African curial family had pushed his way into the most
elite circles of the Roman world. Practically speaking, he had achieved
all of this because of his intellectual training.
Augustine describes his education in great detail. The love of Latin
literature that Augustine developed as a young student of grammar in book
 spurred him to read Cicero’s Hortensius. In book , Augustine describes
Cicero turning him on to the love of wisdom and pulling him away
from emotional attachments. He then began to explore different religious
traditions. His initial disappointment with the Bible in book . pushed
Augustine to the more intellectually polished arguments of the Manichees.
Their use of astrology spurred him to study this science (a pursuit described
in book ) and ultimately to attend a lecture by the renowned Manichaean
teacher Faustus in book . When he realized the limitations of Faustus’s
teaching, Augustine then turned to Academic Platonism. In book , he
described his initial encounter with the teaching of Ambrose. Then in

 As Conf. . suggests, Augustine himself seems to understand (and expect) his audience to read
the text in this fashion.
 It is worth noting that Origen and Pythagoras both began teaching at a similarly young age.
 An incident described in Libanius, Or. ..
 On these sorts of marriage alliances, note the contribution of C. Sogno in this volume.
 edward watts

book  and the beginning of book , Augustine described his experience

with Neo-Platonic study and his excitement in learning about the Christian
Platonist Victorinus. Later in book , his experiences with ascetic ideas
and the lifestyle that they encouraged represents the culmination of this
process of formation. Once Augustine was able to incorporate these ideas
into his intellectual arsenal, he found himself prepared to begin his life in
philosophy – and, ultimately, his ministry at the young age of thirty-seven.
By any measure, this extremely impressive career offered Augustine
tremendous intellectual authority, should he choose to claim it. And, in a
way, he does. After bringing his narrative up to the present (the Confessions
were written sometime between  and ), Augustine begins to teach
his readers. Books  through  shift from a narrative to a discussion of
God’s relationship to the world. The narrative then has set the stage for
this teaching and established the credentials of its teacher.
Augustine’s hostile portrait of his education has become so familiar to
us that it is hard to imagine his development framed in any other way,
but these achievements could easily have been described differently. The
intellectual wanderings of Pythagoras, for example, were not so different
from those of Augustine. Apollonius of Tyana, too, jumped from teacher
to teacher when he found that he had learned all that was useful from one
instructor (or when his instructor proved of limited skill). Even more
relevant are the experiences of Augustine’s younger contemporary Proclus.
Marinus, his biographer, uses an almost identical structure to describe his
subject’s intellectual journey. Proclus, like Augustine, was a precocious child
who, having excelled in rhetoric, then took up the study of philosophy.
Like Augustine, Proclus twice realized the limitations of his teachers and
abandoned their schools to pursue new intellectual directions. And, like
Augustine, Marinus attributed these changes in course to a God (in his case
Athena, the goddess of philosophy). Philosophical biographies could eas-
ily integrate the sort of intellectual evolution that the Confessions describes
into an argument demonstrating that a philosopher had always followed
an inspired path to a truly ecumenical wisdom.
Of course the Confessions does not present Augustine in this way. Instead,
Augustine describes his journey as one of a man stumbling back toward
 Augustine, Conf. .–.  For the date of composition, see O’Donnell : xli–xlii.
 E.g., Philostratus, VA .–.  Marinus, Vit. Proc. .
 Ibid. –. Typical of his reaction to these teachers is Vit. Proc. , which describes how “after
passing his time with them in Alexandria and enjoying their company in so far as they were able,
Proclus looked down upon these teachers when, in a common reading of something it seemed to
him that they no longer bore true to the intention of the philosopher in their interpretations.”
 Ibid. .
 Three generations of Christian philosophical biography 

a truth from which his learning had pushed him. He classifies all of the
intellectual achievements that traditionally marked men for greatness as
evidence of his intellectual confusion and base character. To further rein-
force his negative portrait of the effect of conventional learning, Augustine
claims that every step he took back toward God occurred because of
divine action. Indeed, in a letter to Count Darius, a military commander
and imperial ambassador to North Africa, Augustine makes clear that he
hoped his readers appreciated this as the central point of the Confessions.
Augustine very deliberately turned the rhetoric of philosophical biography
on its head. He described an education that included all of the learning
that philosophical biographies used to demonstrate the authority of their
subjects – and then argued that this learning prevented him from grasp-
ing true wisdom. The narrative of the Confessions then stands as a sort of
antiphilosophical biography in which the authority of Augustine’s ideas
is demonstrated by his turning away from conventional philosophical
training.
In light of this, it is important to consider the personal backdrop against
which Augustine composed the Confessions. If we take  as the date he
began work, it had been six years since Augustine had come to serve as
bishop of Hippo and ten years since he had abandoned his promising
career in rhetoric, his well-born young fiancée, and the financial and social
security that each promised him. These were costly decisions, and the ensu-
ing decade had not been kind to Augustine. For the six years prior to his
work on the Confessions, Augustine had, in the words of James O’Donnell,
suffered from a severe and frustrating case of writer’s block. Many of
his friends had also abandoned promising worldly careers to serve in the
church. They could all perhaps be forgiven for looking back with some
longing and wondering whether they had made the right decision. The
Confessions, however, slams the door shut on such thoughts. By simultane-
ously employing and subverting the rhetoric of philosophical biographies,
the Confessions emphasizes what these men gained by turning away from
classical culture.
 “These words are not learned more easily through this filth, but through these words filth is
committed more confidently. I do not accuse words, which are like choice and precious goblets, but
the wine of error which is poured into us by inebriated teachers, so that we should drink . . . was it
surprising that I was lured into these fruitless pastimes and wandered away from you, my God?”
(Conf. ., ).
 “Accept, then, my son – accept, O excellent man, Christian not by outward profession merely, but
by Christian love – accept, I say, the books containing my ‘Confessions,’ which you desired to
have . . . If anything in me please you, join me, because of it, in praising Him to whom, and not to
myself, I desire praise to be given.” Augustine, Ep. . (to Darius,  CE), NPNF trans.
 O’Donnell : xlii–xliii.
 edward watts

The Confessions then can be read as a response to the personal and

emotional challenges faced by elite followers of the Christian philosophy
expressed so powerfully by Athanasius’s Life of Antony. They needed reas-
surance that they had done the right thing in forsaking the learning of the
world (and the tangible rewards it could bring) for the Christian philos-
ophy of Antony. The narrative of the Confessions shows the correctness
of this decision, and the rest of the work endeavors to explain how God
ensures that his followers will find this true path. It is, in a sense, a text
that tidies up some of the intellectual uncertainty of a post–Life of Antony
world.
When considered alongside works like Iamblichus’s On the Pythagorean
Life, Eusebius’s Life of Origen, and Athanasius’s Life of Antony, Augustine’s
Confessions shows a real evolution in Christian attitudes towards traditional
learning. But the power of the Confessions, like the force of those other
texts, comes from how it expresses new ideas while remaining rooted to
recognizable modes of expression. These Christian texts are, in their own
way, riffs on a standard tune. If we accept that, despite their thematic
and structural idiosyncrasies, philosophical biographies all use narratives
to introduce their readers to a set of esoteric teachings and argue for their
importance, we begin to see surprising cultural continuities that bring
together seemingly disjointed elements of the late antique cultural world.
As the late Roman cultural landscape changed, the philosophical sys-
tems for which these texts argued also evolved. If this is recognized, it even
becomes possible to resolve the philosophical paradox with which this essay
began. Whether or not one judges Simeon Stylites a proper philosopher,
it is impossible to deny that Theodoret’s life of him draws upon the same
rhetoric that had been used to define philosophical authority for much
of the past millennium. His Life of Symeon uses the archaized (albeit
sometimes slightly awkward) Attic Greek that served as the lingua franca
of elite discourse, and it is structured according to the basic model of
a philosophical biography. Theodoret begins with Simeon’s upbringing.
Theodoret describes Simeon as a shepherd “like Moses and David” who
decided, as a youth, to pursue a life of “the highest philosophy” after hear-
ing a reading of Scripture in church. Simeon progressed slowly through

 Hence P. Brown’s perspicacious description of how conversion is presented in the later books of the
Confessions (: ).
 It is notable in this regard to consider that Theodoret published his life of Simeon earlier than the
rest of the Religious History and structured this life differently from those that make up the rest of
his collection.
 Theodoret, RH ..
 Three generations of Christian philosophical biography 

philosophical training at the foot of many masters, first spending two years
learning from neighboring ascetics and then enrolling in the “philosophical
retreat” of two abbots. Simeon’s final stage of training involved imitating
the examples of great men of God like Moses and Elijah. Once Simeon
had absorbed these various lessons, Theodoret describes at length how he
demonstrated the power of his philosophical training through acts of fore-
sight and adjudication. The portrait then concludes with a discussion of
Simeon’s teaching. Indeed, it is even possible to read Theodoret’s biogra-
phy of Simeon as an illustrative complement to the ideas expressed in his
Cure of the Hellenic Maladies, a systematic redefinition of classical philos-
ophy and an explanation of a Christian sort of philosophical thought.
Like other philosophical biographies, Theodoret’s Symeon provides con-
crete examples of the way that a particular brand of Christian philosophy
could be practiced and shows the great feats it enabled a philosopher to
accomplish. The failure of Gibbon to take this new form of philosophy
seriously does not mean that we should fall into the same trap. Theodoret
calls Simeon a philosopher because he truly believes him to be one – and
imagines that his audience will too. If this claim and others like it are taken
seriously, a more detailed and nuanced map of late antique cultural change
reveals itself. The nature of philosophy may have changed and the category
of philosopher may have expanded to include a strange assortment of new
characters, but the slowly evolving rhetoric with which philosophers were
described suggests that fundamental cultural continuity existed amid this
change. It then becomes possible to speak about a literary bridge joining the
philosophical biographies of the tetrarchy to their ostentatious Theodosian
progeny.

 Ibid. ..  Ibid. ..  Ibid. ..  On this work see Siniossoglou .
 Indeed, this point is ably made by Ashbrook Harvey (: –).
 chapter 7

The education of Paulinus of Pella: learning

in the late empire
Josiah Osgood

A hallmark of John Matthews’s first book, Western Aristocracies and Impe-

rial Court, is a fine appreciation of the Latin literature of the age not only
as a window onto the world of late antique aristocrats but also as an artifact
of their ease. We encounter Symmachus celebrating in a letter Ausonius’s
new masterpiece, the Mosella, “pick[ing] out for special mention that part
of the poem which might to our tastes seem the most unsympathetic,
the resourceful ‘Catalogue of Fishes’ – where, said Symmachus, were listed
more varieties of fish than he had ever seen at Ausonius’ table!” (A footnote
duly notes, “At Mosella  f., Ausonius pays tribute to the gudgeon, not a
culinary delicacy.”) We encounter, too, Ausonius and the emperor Valen-
tinian himself, in friendly rivalry (or so Ausonius says) stitching together
wedding poems from lines of Virgil, Ausonius’s half of the exchange surviv-
ing in his Cento nuptialis – “in its concluding section, a hilariously indecent
example of the art of quoting out of context.” Yet the characteristic touches
of humor here point to a most important theme of Matthews’s work on
this period and a lesson he offers to all historians: political history cannot
be fully understood without an awareness of cultural and social history;
both the “public” and “private” lives of a ruling class have to be kept in
balance; otium matters as well as office.
Matthews’s most recent work, The Journey of Theophanes, has made this
point even more remarkably. By devoting an entire book to a six-month
trip from Egypt to Syria and back undertaken by a public figure for reasons
that in detail remain largely unknown, Matthews shifts the focus onto such
questions as how one traveled in the Roman Empire, what one ate, and

I would like to thank the editors of this volume as well as the anonymous referee for significant help
with this paper – and John Matthews, for making the Roman Empire come alive for me.
 Matthews : .  Matthews :  n. .
 Matthews : . The Loeb edition of Evelyn White (–) prints the Latin and then reduplicates
it on the facing page instead of offering an English translation. See now McGill () for a full
study of the cento.


 josiah osgood

with whom. “The dossier of Theophanes is – indeed, it is literally – an

everyday sort of history, but it is here that its real interest lies,” Matthews
remarks. “To know the conditions of travel from Hermopolis to Antioch
and the time it took to get there is to know something just as important to
the Roman empire as the change of religion that was lurking around the
corner at just the time Theophanes made his journey . . . . The latter may
be the more glamorous story, but it is not necessarily the more important.”
The ordinary things matter; “A society cannot be understood without a
knowledge of the more ordinary, as opposed to the more spectacular,”
Matthews suggests.
It is this insight that helps account for Matthews’s important, and orig-
inal, treatment of my subject in this essay, a technically inferior poet from
late antique Aquitania, known only through the  lines of his verse
autobiography in which he tried to make sense of his long life. Paulinus’s
Thanksgiving to God in the Form of My Journal (Eucharisticos Deo sub
ephemeridis meae textu), to give the work its full name, written in 
CE, the author’s eighty-fourth year, is primarily valued by historians as
an eyewitness source for the barbarian invasions that afflicted Gaul in the
early fifth century. Yet as the title and, even more, the prose preface to the
work suggest, this is a poem in which one also encounters something of
the Roman empire’s everyday history (at least, the everyday history of the
empire’s upper classes). Men of fame, Paulinus announces (praef. –), have
handed on an ephemeris of their deeds to perpetuate their glory – and here
he may be thinking even of so early a figure as Julius Caesar, whose deeds
were still in the fifth century the subject of schoolboy declamation. But
Paulinus’s opusculum, while an ephemeris too, will tell not of his res gestae but
of his “whole life” from his birth in Macedonian Pella, his family’s return
to Gaul (where he met his grandfather, Ausonius), and his childhood there
onwards; the barbarians only burst into the “vitals of the Roman realm”
in line  of the poem (Romani . . . viscera regni). While historians have

 Matthews : .
 For the Eucharisticos, the three principal editions are Brandes , Moussy  (whose text I follow
here), and Lucarini ; the last is discussed by Coşkun (), and on the text see also Coşkun
. A full translation into English prose, based on the text of Brandes, is provided in Evelyn White
–; selections are rendered into English verse by Lindsay (), who also offers a sensitive
appreciation. The full bibliographic resumé of Coşkun (: –) obviates the need for one here.
This article of Coşkun itself is especially valuable, as is McLynn b, discussed further below.
On matters of chronology, see especially Coşkun a, which slightly revises the tabulation in
PLRE, s.v. “Paulinus” (). I agree with McLynn and Coşkun in seeing the poem as a single product
completed in , rather than (as was earlier maintained) written in two phases.
 Sid. Apoll. Ep. ..; on the ephemeris of Caesar, cf. Moussy : –, citing Symm. Ep. ...
 The education of Paulinus of Pella 

traditionally focused on its later sections, Matthews, in Western Aristocra-

cies and Imperial Court, gave full recognition to its earlier parts, drawing
attention to its evidence for the pleasures a young man of upper-class soci-
ety in Aquitania could enjoy and, at the same time, Paulinus’s account
of himself in his twenties, strenuously trying to renew the vineyards on
the estates of his new wife. The grandson of Ausonius – known to us,
let us recall, only through this poem – absented himself from public life
and was even, on his testimony, devoid of political ambition. “It may be,”
Matthews suggests, “that those Gauls who engaged in politics have unduly
distracted attention from the unknown, but certainly far greater, number
of those who did not.” And even for those who did engage in politics,
that political engagement could itself have been a derivative aspect of their
social station.
If Paulinus’s verses are a window onto Roman Gaul in the late fourth
century, as well as the barbarian invasions of the early fifth, we must keep
in mind that the poem is an artifact of an even later time, the middle fifth
century, and so is a product not merely of ease but of a period of great
difficulty for Paulinus, in which, after witnessing the barbarian invasions
and serving under the usurper Priscus, and after quarreling with his family
members and losing his estates, he had to struggle to make a life for himself.
What is more, as Neill McLynn has urged, in an important reevaluation
of the poem, the Eucharisticos – a life painfully squeezed into hexameters
that must have caused their author some labor – should be seen not simply
as a chronicle of the poet’s life but as a rhetorical effort too. McLynn’s
hypothesis that Paulinus was trying to defend himself late in life against
his fellow Christians in Marseilles – Salvian, for one, has some harsh words
concerning the ruined possessores of Aquitania – may be hard to prove,
but he is right to urge that no straightforward equation can be drawn
between poetry and personality; the relationship has to be teased out.
The ephemeris of Paulinus, written in verse, is not the papyrus ephemeris of
Theophanes found by chance in the sands of Egypt.
In the rest of this paper, in the spirit of Matthews’s invitation for his-
torians to examine the more ordinary, I shall look at a part of Paulinus’s
ephemeris that has received less critical attention, the account the poet gives
 See, e.g., the treatment of Paulinus – to be traced in each volume through the index – in Courcelle
: –; Van Dam ; Drinkwater and Elton ; Mathisen ; Ward-Perkins ; and
Heather .
 Matthews’s treatment of Paulinus is also to be traced through the index; note especially Matthews
: –, –, and –.
 Matthews : .  McLynn b.  De gub. Dei .–.
 On the possible rhetorical context, see especially McLynn b: –.
 josiah osgood

of his formal education. And I shall relate this to the important theme of
education in the poem as a whole. My basic argument is that the Eucharis-
ticos offers valuable evidence not only for educational practice in the fourth
century (and earlier) but also for new attitudes to education in the fifth.
I will begin by discussing Paulinus’s description of his own instruction,
showing what in it is typical and what irregular through the poems of
his all-important grandfather, Ausonius, as well as through a remarkable
school text, first published in  and thought to originate in third- or
fourth-century Gaul. Next, I will show how Paulinus’s account of his
formal education (abandoned, he says, just as he was taking up rhetoric)
reinforces the larger theme of the poem that worldly trappings, including
eloquence (eloquium), ultimately matter little, and that true instruction
comes through lived experience; on this level, the Eucharisticos repudiates
the legacy of Ausonius. My third section shows how in his treatment of his
education, as in other parts of his poem, Paulinus owes a particular debt to
Augustine’s Confessions: both authors were reevaluating the role of formal
education, which had remained much the same for hundreds of years,
despite Christianity. Paulinus shows the spread of Augustine’s novel ideas,
while Augustine also offered Paulinus a way to justify his own deficiencies,
which had nothing to do with Christianity. Finally, I will embrace the
paradox that Paulinus chose to treat his whole life in hexameter verse with
frequent reminiscence of Virgil, an author he says caused him difficulty in
school. In part, I suggest, that seeming contradiction helps implicitly to
cast Paulinus as a new kind of teacher for his reader. And it also brings
us closer to Paulinus the man: the specific contents of the poem aside,
simply by expressing himself in this way, he has finally exposed something
about himself too. Paulinus cannot fully detach himself from his learning,
however imperfect it was, because, as was true of education in the Roman
Empire for centuries before, it above all conferred status.

scenes from a schoolroom: paulinus’s early education

What made up Paulinus’s formal education? The poet describes it in some
detail, in a passage that should be important for historians, because it gives
firsthand insight into one of those more ordinary, everyday sort of events
that took place in the Roman empire. He not only reveals entirely typical
features but adds some new small details to the general picture. And yet,
 The standard account in English on ancient education was for long Marrou , based on the third
edition of his work in French; it was supplemented by Bonner . Two recent works, Morgan
 and Cribiore , reflect valuable recent scholarship and contribute new insights of their own
through the use of papyri in particular: education emerges as less of a unified “system”; see also
 The education of Paulinus of Pella 

at the same time, Paulinus shows that his education did not turn out as
it was supposed to: his verses point to something triumphalist historians
of education traditionally have thought less about, the failures that the
Roman educational system must have had, which before Paulinus’s time
tended simply to be written out of history or (perhaps it is better to say)
never written in at all.
Paulinus prefers to divide his early life into discrete periods of time,
which allows him to employ a varied lexicon of temporal expressions (e.g.,
nostra trieteride prima []; primi . . . lustri []; duo . . . decennia []).
His education helps determine these divisions, and is envisioned as falling
in three main phases, the last aborted early into its undertaking. Initially, at
the end of his third year of life, Paulinus was trained at home in speaking,
deportment, and basic morality; then, at the end of his fifth, he went
to school to learn grammar through a reading of literary texts, including
poetry; and finally, at the end of his fifteenth, he was to learn rhetoric.
The basic sequence is entirely normal; it had been much the same for
hundreds of years and was still obviously functioning in the fourth-century
Bordeaux where Paulinus grew up. Ausonius himself not only celebrates
in the Professores his fellow teachers of grammar and rhetoric, but, in the
Protrepticus addressed to his grandson (who should be Paulinus’s older
brother ), he plots his own career as one that took him from training
infants to boys to youths who had put on the clothes of a man and were
to acquire “force in speaking” (fandi . . . vigorem) – until he was bidden to
the “golden palace” (aurea . . . palatia) to undertake the instruction of the
young prince Gratian (–).
In the preliminary phase, which took place at home, Paulinus represents
his parents as most directly involved, taking great care to train their young
son using a system of rewards. This is the sort of learning that focuses on
clear “rights” and “wrongs.” The goal was obviously a basic socialization.
Paulinus’s family also wanted him to learn the alphabet, combinations of
letters, and words – the so-called “first elements” (prima elementa [] ) –
Cribiore  and . Watts  underscores the importance of local contexts for understanding
changes in ancient education and the inadequacy of Christianity alone as an explanation.
 One of the strengths of Cribiore  is her sensitivity to the variety of lived experience, and her
picture of Egypt is now supplemented with her study of the school of Libanius in fourth-century
Syria (Cribiore ).
 Cf. Roberts : –.
 On the sequence of schooling, see the important article by Booth (), supplemented with Kaster
 (especially –).
 See PLRE, s.v. “Ausonius” (); and for the father, “Thalassius” (); see also PLRE I, stemma no. .
Coşkun b is the fullest study of the whole family.
 For the passage, see Booth : – and Kaster : –; and in general for Ausonius’s academic
career, Booth ; Green ; Kaster : –; and Coşkun b.
 The prima elementa are also mentioned by Aus. Prof. . and ..
 josiah osgood

and to take preliminary steps in reading. For this, they (or a teacher within
the house) might have used cakes, even shaped into the letters themselves,
a practice attested elsewhere: not even Jerome disdained it. Paulinus’s
parents finally were also concerned that he acquire the rudiments of correct
speech; the “ten special signs of ignorance” (decem specialia signa amathiae
[]), which he mentions, otherwise unattested, probably were a set of
simple rules passed on to avoid “barbarisms” in speaking. More puzzling
are the vitia mentioned in the following line – perhaps also solecisms or
perhaps other types of bad behavior. Late in life Paulinus highly valued
this training, more than what followed; he intriguingly calls it “the old
Roman practice” (Romana . . . vetustas []) and claims that it has fallen
out of favor. As we shall see, this articulation may have something to do
with the subsequent suspension of his education as well as with attitudes
to learning he formed later in life after the barbarian invasions.
Paulinus’s instruction in more advanced reading and grammar, at a
school, began early (nec sero []), and, he says, it was under compulsion.
One can safely guess that Ausonius put pressure on his grandson to make
as rapid a progress as possible. Education was for Romans an investment
by families in their young to maintain or achieve status, and hard work was
therefore owed to those who made it possible; the aim was to reach their
measure, if not surpass it. With Ausonius, the attitude would have been all
the stronger; in the Protrepticus addressed to his other grandson (Paulinus’s
brother), he famously hopes “that mindful of your father and of me, you
may always aim through eloquence for the hard-won rewards of the Muses”
(ut patris utque mei non immemor ardua semper / praemia Musarum cupias
facundus [–]). The young boy is to tread the same path that Ausonius
took to his consulship, Paulinus’s father Thalassius to the proconsulship of
Africa, and Paulinus’s uncle to the prefecture of Italy (–).
The curriculum of authors included in the Protrepticus (–) partially
overlaps with the reading list of Paulinus, but Paulinus’s discussion is delib-
erately pointed: “I was compelled to read and learn the beliefs of Socrates
and the martial fictions of Homer and the wanderings of Ulysses; and then
straightaway I was compelled to traverse the books of Virgil too” (dogmata
Socratus et bellica plasmata Homeri / erroresque legens cognoscere cogor Ulixis;
/ protinus et libros etiam transire Maronis [–]). Christian texts played no
part in this schooling. The young Paulinus had to read instead the dogmata
 See now especially Cribiore : –.  Jer. Ep. ..  So Moussy : –.
 On this phase of education, see now especially Cribiore : – and also Morgan : –.
 For details of the careers, see PLRE, s.v. “Ausonius” (), “Thalassius” (), and “Hesperius” (). On
the context of Ausonius’s activities as patron, note the contribution of Garnsey in this volume.
 The education of Paulinus of Pella 

of Socrates, probably a collection of gnomic sayings or injunctions, like the

“Answers of the Wise” or the “Delphic Precepts” known to have functioned
as ancient school texts; he also had to learn Homer, whose epics are “fic-
tions” (plasmata) and “errors” as well as a story of “wanderings” (errores);
he had, finally, an unrewarding traverse (transire) through the books of
Virgil, the centerpiece of the Latin curriculum. Used to conversing with
Greek slaves, Paulinus maintains, he had trouble acquiring an “unknown
tongue” (ignotae . . . linguae []), a somewhat unconvincing claim, given
what was said before, but evidence that complements Ausonius in showing
that bilingual education on some level, perhaps fairly basic, could still be
expected in this milieu, and that within it the Greek Homer still came
before the Latin Virgil. Hence it was not, as is sometimes implied, Pauli-
nus’s birth in Pella or his Greek slaves that led to his reading Homer first,
nor, at the same time, is his claim inherently dubious. Learning at least a
little Homer would have been expected.
This now may be confirmed by an unexpected piece of evidence, the
text of a bilingual Greek and Latin schoolbook that Carlotta Dionisotti
resurrected from a transcription made in  by Conrad Celtes of a
manuscript at Sponheim, and that she published brilliantly, but perhaps
not completely invitingly, in . Like other extant schoolbooks, this
text includes a series of vignettes drawn from the everyday life of the boy
or girl who was to use the text – and indeed who would probably have
written down a copy of the very text in question as dictation. (These
collections have since humanist times gone under the title Colloquia, but
misleadingly, as they are not primarily dialogues.) Taking into account a
very interesting scene (secs. –) that ends the series, in which advocati,
causidici, scholastici, and evocati come to the secretarium of a judge and
plead many cases, “each to the best of his ability in accord with literary
eloquence” (quisque ut potest secundum literarum facundiam), Dionisotti
 See Goetz : . (responsa sapientum) and . (praecepta in Delphis ab Apolline in columna
scripta), preserved in the so-called Hermeneumata Stephani.
 Error is a favorite word of Paulinus and it here points to later uses in the poem, especially at line
, seemingly in reference to heretical Christianity, where are also mentioned the dogmata prava
(cf. dogmata Socratus) that Paulinus temporarily embraced. For interpretation of the latter passage
and Paulinus’s so-called heresy, see now Coşkun : –.
 On Virgil’s place in the curriculum, with discussion of how the author was presented to students,
see the contribution of McGill in this volume.
 Homer precedes Virgil in the reading list of Protrepticus; cf. Dionisotti : –. On Ausonius’s
evidence for knowledge of Greek in fourth-century Gaul, Green  provides an overview. The
classic study of Courcelle () picks up at a later point.
 So, e.g., Moussy : – and Green : ; Matthews (: ) also seems to express doubts
and notes the Greek slaves.
 Dionisotti .  The texts are found in Goetz .
 josiah osgood

proposes a date for the text in the third or fourth century CE, and on
the basis of some of the garments and foodstuffs mentioned plausibly
suggests that it originated in Gaul. Hence, when the student in Celtes’s
text (secs. –) discusses the readings in the schola and begins with the
Iliad and the Odyssey and then continues with other Greek and Latin
texts (including a number of Latin poets, “Maro, Persius, Lucan, Statius”),
we should have further support for the picture Paulinus gives of his own
instruction.
Celtes’s scenes, like the other so-called colloquia, though obviously styl-
ized to some degree, are of great value in giving a vivid sense of what would
have taken place in at least some Roman schools. Reading through them
can help one to envision something of what the young Paulinus’s training
was like: it began at dawn, or even earlier, with the child rising from bed,
getting washed and dressed, and walking to school, where a long morning
of drills in reading, speaking, and writing ensued. As he – the child is
clearly a boy in most of the scenes – dresses in a way fitting for a “son of
the household” (filium familias [sec. ]), and as he orders around a slave,
it is clear that he is to imagine himself of high status, and that he was
being taught not just to read or write but also in some ways an even more
important lesson, that education helped determine status.
But rather than review all of Celtes’s material, much of it already familiar
from earlier studies of education, let me instead mention, because of an
important detail we discover, what follows the morning lessons, a late,
and much-needed, lunch: “Since I am hungry” (our pupil is envisioned
saying), “I say to my slave: bring a table and cloth, a napkin; and go to your
mistress, and bring bread and spread and a drink of wine, beer, spiced wine,
absinthe, milk. Say to my mother that I have to go back to the house of the
teacher again. So hurry up in bringing us the bread” (Quoniam esurio, dico
meo puero; Pone mensam et mantele, mappam; et vade / ad tuam dominam,
et adfer panem et pulmentarium et potionem vini, cervisiae, conditi, absinthii,
lactis. Dic meae matri quod iterum habeo reverti ad domum magistri. ideo
ergo festina nobis adferre prandium [secs. –]). The list of beverages, going
well beyond the child’s drink of milk, exemplifies the school text’s tendency
to accumulate items of vocabulary, including cervisia, a beer with a Celtic
name, commonly drunk in Gaul – one of the clues Dionisotti pointed
to in locating the text. Dionisotti does not mention the milk, but this

 Dionisotti : –.

 Dionisotti : . Absinthe, by contrast, was widely defused: it features among the foodstuffs in
the archive of Theophanes: see Matthews : , with .
 The education of Paulinus of Pella 

too, I suggest, points to northern Europe. While on the subject of food,

one may add, this whole passage is also of interest for providing unique
evidence for a quick “school lunch,” to be contrasted with an evidently
more elaborate meal consumed by the adults. It is a pity that the cycle
of poems Ausonius wrote called Ephemeris only partially survives, with
the invitation to lunch issued – to include six guests (six is a convivium,
but more a convicium [.–]) – but the lunch itself missing, so as to
preclude direct comparison. As Dionisotti suggests, Ausonius’s cycle may
be a reworking of the school texts from which he learned and taught, a
point to bear in mind when we consider Paulinus’s own reflections in his
ephemeris and the influences that shaped them; the educational account in
the Eucharisticos is, in a sense, not a reworking but an inversion of a school
text. Paulinus ends up not in the Forum but on a vineyard.
To return to the Eucharisticos, we can more quickly dispose of the final
stage of Paulinus’s education because it never took off. Imagining that he
was, after the demanding study of grammar, on the road to rhetoric –
and a “fit return” (fructum . . . dignum []) for his effort – Paulinus was
struck down by a quartan fever (quartana, i.e., malaria) “when the fifth
triad of my life was barely completed” (vix impleta aevi quinta trieteride
nostri []). His parents, in consultation with doctors, recommended a
rest from studies and, we are told, during a slow recovery the adolescent
spent his time riding horses and hunting with his father, who had recently
put aside these pastimes to try to set an example for his son; this “cure”
is a telling comment on the physical demands education in rhetoric was
thought to, and could, entail. Never inclined to study again, the fully
recovered Paulinus was not compelled to finish his training. Were Ausonius
still alive, he doubtless would have been horrified – and one might even
guess from this testimony that he was not still alive, for could this have
happened otherwise? It is no accident that Celtes’s school text ends with
the hearings at the secretarium and then two great public trials (or that

 In the Mediterranean world, milk was less often drunk and was not an ordinary foodstuff; in the
account books of Theophanes, for instance, among the many comestibles purchased, milk does not
come up, though cheese does (Matthews : ).
 Only one other schoolboy’s lunch is described in any detail in the so-called Colloquia: accipio panem
candidum, olivas, caseum, caricas, nuces. bibo aquam frigidam. pransus revertor iterum in scholam
(Goetz : –). The preparations made for lunch in the Colloquia Monacensia (Goetz :
–) – including a vinum bonum domesticum – clearly mimic those of an adult meal. See the
additional references at Dionisotti : .
 Dionisotti : –.  See the final section below.
 On the rhetorical phase of education, see now especially Cribiore : – and Morgan :
–.
 Ausonius died sometime in or after  CE, the year Paulinus turned sixteen.
 josiah osgood

we also find employed as a school text the “Opinions of Hadrian,” curtly

expressed views on a variety of practical legal problems ): eloquence and
its expression in public life were the summit to which a Roman education
ideally led, and Ausonius had expressed the same idea in his Protrepticus.
When an education such as this was cut short or abandoned, Paulinus’s
text can help show, intermediary phases of it – all that parsing, memorizing,
and endless repetition – may have later seemed less than worthwhile,
not least compared to the earlier, more basic, foundation. And yet, it
is precisely because one had that education that one could knowledgeably
condemn it. Augustine was keenly aware of this problem, and his reflections
on the matter in the Confessions have obviously shaped Paulinus’s less
searching examination on his own training, a point to which we will
return, after looking at the theme of instruction in the poem as a whole.
But here one can make a more basic observation, that in linking the value
and reward of education to appearances in public life, or to service in
the government, as the school texts or Ausonius in his Protrepticus did,
educators failed to develop other justifications for their practices. The
traditional Roman education, in particular at its highest levels, was in
alliance with the government – especially its more ceremonial aspects –
and after the empire came to an end, the fortunes of the more basic
grammar on the one hand and rhetoric on the other could diverge. Basic
types of grammar could and did survive as a way to learn Latin, even as a
second tongue or a purely literary language, but not as prelude to rhetoric.

learning late: paulinus’s education in hindsight

If something of an Ausonian taste for the ephemeral – the picturesque
scenes of daily experience – informs the vignettes of Paulinus’s early life,
at the same time we need to see those vignettes not just as a source for
educational practice in the Roman empire but as a key part of Paulinus’s
poem as a whole. Indeed, the account Paulinus gives of his formal learning
contributes to a fundamental theme of the Eucharisticos, that the education
that mattered in Paulinus’s life came not in school but through experiences
afterward, in adulthood; a frequent shifting between verbs in the present
and past tense underscores this increased knowledge on the poet’s part
(and, incidentally, gives the poem much of its emotional depth). The
 The Hadriani sententiae are found once again in the so-called Hermeneumata Stephani in Goetz
: –.
 To take the first hundred or so lines, notice the concentration of present tense verbs in lines –,
–, and –, and perfect (or historical present) verbs in the other sections.
 The education of Paulinus of Pella 

poem, though addressed to God, attempts to make an argument for its

reader that a different kind of education is to be valued, one in which God,
or Christ, is the ultimate teacher, who knows all and has given an ignorant
Paulinus fleeting pleasures only to teach him, through subsequent adverse
circumstances, not to love prosperity and its trappings. This includes elo-
quence itself, which allows Paulinus, in the account of his schooling, to
repudiate the Ausonian heritage under which he grew up and, ultimately,
to become a sort of teacher himself, despite his disavowal of any claim to
that title.
After its checkered start, Paulinus’s education takes a turn for the prac-
tical when he finds himself, newly married, in the decayed vineyards of his
wife’s estate. Unlike his work in school, this was a task he came to enjoy,
and, he says, he “learned the way to renew exhausted vines” (fessis . . . vinetis
comperta mihi ratione novandis [–]). But it was really from his thirtieth
year, when both his father died (with his will contested) and the barbarian
invasions began, that his new education started. It was the quarrel over
the inheritance with his brother, McLynn has convincingly argued, that
led to most of Paulinus’s problems and encouraged him, reluctantly, to
take up with those who jockeyed for power in Gaul after  CE. “For
experience easily showed me,” he writes obliquely, addressing Christ, “how
great, thanks to you, the favor bestowed on me by the powerful was, when
often I was granted unknowingly the brilliant distinction of leading men,
before I gained and held possession of it as my own” (Namque et quanta
mihi per te conlata potentum / gratia praestierit facile experiendo probavi, /
saepe prius claro procerum conlatus honori / ignorans proprio quam praeditus
ipse potirer [–]). But it was subsequent misfortunes that taught him
the most, as he explains in an important passage: thinking back to his
prosperous estate, with its throngs of clients – themselves, like his edu-
cation, a centuries-old sign of status – he writes, “I now regret having
loved in that time those things destined to perish so quickly and at last
with better perception as I grow old, I recognize that it was useful to have
them removed, so that with worldly and fleeting riches lost I might learn
to seek instead those that will endure” (Quae peritura cito illo me in tem-
pore amasse / nunc piget et tandem sensu meliore senescens / utiliter subtracta
mihi cognosco fuisse, / amissis opibus terrenis atque caducis / perpetuo potius
mansura ut quaerere nossem [–]). With the striking phrase sensu
meliore senescens, Paulinus announces that in fact his learning is still, in

 Again, see the final section below.  See especially McLynn b: .
 For Paulinus’s turbis . . . clientum compare the contribution of Garnsey in this volume.
 josiah osgood

his eighty-fourth year, under way. Unlike his childhood education, which
began early, this instruction, he emphasizes, has come late.
As he understands it only now, his real teacher has been Christ, who,
showing a mixture of solicitude and severity to his pupil, teaches by
experience. As the poem repeatedly emphasizes, Paulinus’s experience has
been a mixed one: early joys, later sufferings, and in the suffering occasional
moments of good fortune, too, that show divine mercy, culminating in the
purchase of his estate by a Gothic settler. The prose preface makes espe-
cially clear Paulinus’s understanding of this prolonged instruction. Now,
Paulinus says, he knows that God’s mercy has always been around him; in
early life there were “fleeting pleasures” (temporariis voluptatibus), and in
the present part, too, he knows that “the care of his providence has been
of help to me, because, while training me with unremitting misfortunes,
he has clearly taught me that I ought neither to love too immoderately
present happiness . . . nor be greatly overawed by adversities in which I
had found that his mercies could support me” (curam mihi providentiae
ipsius profuisse, quod me adsiduis adversitatibus moderanter exercens eviden-
ter intruxit, nec impensius me praesentem beatitudinem debere diligere . . . nec
adversis magnopere terreri, in quibus subvenire mihi posse misericordias ipsius
adprobassem [praef. ]). The image here of God as teacher, and Paulinus as
pupil, is all the more striking, as Paulinus has already by this point in the
text disavowed any great faith in his own eloquence (eloquium) such that
he could dare to rival a great writer.
Like all such disavowals, the rhetoric here cannot be taken purely at face
value. Paulinus may claim his is a carmen incultum and wish that it not
come to the notice of the doctiores (praef. ), and the curious reader who
happens upon it may be invited not to praise anything in the poet’s doings
or verses (in gestis vel in versibus) and instead to confine it all to oblivion
(praef. ), but the reader, of course, has to compare this poem with others
and has to evaluate it, and the reader might think in particular of the works
of Paulinus’s grandfather, Ausonius. In suggesting to his reader that deeds
and verse do not matter, but in then offering a poem over six hundred lines
long recounting his life, Paulinus is showing that they do, but not in the
way an Ausonius might have thought. Who in the end, we are to ask, really
is doctior?

 Euch. .
 See esp. Euch. praef. , –, –, –, –. For this Augustinian idea, see the next
section and also the contribution of Watts in this volume.
 Note such phrases as ambigua sorte (), instabilis mundi (), lubrica tempora vitae (), insta-
bilis . . . aevi (), and ambiguae . . . vitae ().
 The education of Paulinus of Pella 

With all this in mind, let us return to the lines concerning Paulinus’s
actual instruction. There, in varied form, he repeats the claim of the preface
that his own attainments do not place him among the distinguished; the
very page on which he writes, he says, reveals it. The “double learning”
(doctrina duplex []) of Greek and Latin with which he struggled, he also
says, though it does convey a “twofold radiance” (gemino . . . splendore [])
is suited to more powerful minds: “Its wide range easily drained dry the
slender vein of my mind, too barren, as I now understand” (sterilis nimium
nostri, ut modo sentio, cordis / exilem facile exhausit divisio venam [–]).
The seemingly generous compliment to a traditional ideal of learning seems
somewhat undercut by Paulinus’s simple assertion that he now knows such
learning was simply beyond him. To be sure, Ausonius once said something
similar in his Professores about his own learning of Greek (.–), but the
clear implication of Ausonius’s comment, coming as it does in this work
celebrating his fellow teachers, is that Ausonius made good on his early
inadequacy later. With Paulinus, there is no remorse; his failure did not in
the end, he suggests, matter at all. Paulinus, instead, has found a way to
justify his failure to himself – and, perhaps, to his reader. It was part of a
larger plan, which he says he now understands: the pleasures he enjoyed in
place of the arduous training in rhetoric prepared him, with the hardship
that followed, to make way for the pity of the God that he believes will
be with him for his last days. But how, one must now ask, leaving aside
Christ and seeking a historical explanation, did Paulinus come to this
interpretation of his life?

the debt to augustine

He came to it, at least in part, through Augustine. Anyone reading through
the Eucharisticos, especially in its earlier sections, cannot fail to be struck by
its similarities with the Confessions. These similarities are not only formal
(Paulinus, like Augustine, addresses his meditation to God and ostensibly
disarms the curious reader), but they also concern the sequence of events in
Paulinus’s life. God grants the nursing Paulinus sustenance; Paulinus grows
conscious; he is compelled to learn a foreign tongue and has difficulties
with it; a wish to take a vow of continence is refused by his parents; Paulinus
knows illicit love, and a child is born to him out of wedlock. Of course, one
 “And this now, against my wishes, this page of mine reveals” (quod nunc invito quoque me haec mea
pagina prodit []).
 First properly emphasized by Courcelle (: –); see further Moussy : –, McLynn
b: –, and Coşkun : –.
 josiah osgood

might object, these might be common enough, even essential, experiences

in a young man’s life; but the language used to describe them at times
overlaps too. Even more profoundly, there is the sense in both authors of
a late but powerful appreciation of divine providence: God acts in a man’s
life, ultimately for his own good, even if at first unappreciated; and God is
the ultimate teacher.
Seventy years ago, Pierre Courcelle drew attention to all these similari-
ties and pointed out that the appearance of Augustine’s text must account
for the pronounced differences between the style of Paulinus’s autobiog-
raphy and that of his grandfather’s self-portraits in verse, but Courcelle
commented less on the motive or effect of the appropriation. McLynn
has more recently suggested that Paulinus, in so obviously advertising an
affinity with Augustine, is trying to establish his “own credentials” and to
burnish “the status of both his life and his art.” But one can sympathize
with Coşkun’s suggestion, too, that the reading of Augustine’s great work,
with its justification of at least some of a life’s failures, must have had
psychological appeal to Paulinus as well as a rhetorical one. There was
explaining to do to others, but also to himself.
Augustine was not the least useful when Paulinus came to think of his
failure to pass through all stages of the curriculum set for him by his elders,
for Augustine had in the Confessions written devastatingly of the educa-
tional system of the day, especially in its higher levels. Learning to speak,
to read, to write, at home and at school, was, Augustine believed, of value,
but at the time, “poor wretch, I was unaware of what use it was” (quid
utilitatis esset ignorabam miser [.]). (The temporal gap here between the
events in question and the time of writing points to another of Paulinus’s
useful borrowings from the Confessions.) The objective of those who forced
his education on Augustine was merely “to satisfy insatiable desires for vast
amounts of nothing and for a glory that is disgraceful” (ad satiandas insa-
tiabiles cupiditates copiosae inopiae et ignominiosae gloriae [.]). Augustine
came to reject their motives, but he still was able to “make good use of
letters” (bene uti litteris [.]). Words are only precious vessels, not the
wine of error poured into them (.). This belated understanding of an
education conveniently provided a template for Paulinus’s own experience
and must have intensified his sense that he had to value the founda-
tion of his education – with its clear sense of “right” and “wrong” neatly
overlapping with the categories of a simple Christian morality – but he

 Courcelle : .  McLynn b: .  Coşkun : .

 Note the parallel language of Euch. – and –.
 The education of Paulinus of Pella 

could dismiss his inability to have achieved the eloquium favored by his
elders.
Yet we must observe that Paulinus in some ways alters Augustine’s cri-
tique, since their stories in key ways diverged. Whatever his troubles with
Greek, Augustine, on the one hand, ultimately flourished in school, learned
rhetoric, and went on to teach the subject himself; for him it was a passport
to higher standing. Paulinus, on the other hand, was, on his own testimony,
a dropout at age fifteen; he did not (so he says) love the emotional poetry
that enraptured Augustine. Because Paulinus went less far, he has less need
to reevaluate and to dwell on the error of what he acquired. Put differ-
ently, unlike Augustine, he did not have to love literature, in order to learn
not to love it (as he did with some of the other delights of life). The full
attainments of the traditional education for Paulinus form a summit never
reached, not one taken up to be later abandoned; in my final section, I will
suggest how this accounts for a paradoxical feature of the poem. Let us note
here a different point, that despite the divergences between the two texts,
the Eucharisticos is significant because it shows the spread of Augustine’s
ideas as put forth in the Confessions, not least those concerning the reevalu-
ation of education. Education, remarkably slow to change throughout the
centuries of the Roman empire, was not to change overnight in the fifth
century either. But Paulinus, however inconsequential his views for his
contemporaries, or posterity, helps us to understand how gradually they
would change, with grammar (say) being put to use to read and learn the
Bible. In an earlier time Paulinus would have remained simply the dropout
that he was; in his lifetime he carves out a place for himself in the age’s
enduring literary culture.

the debt to ausonius

Any satisfactory interpretation of the theme of education in the Eucharisti-
cos must, finally, take into account another divergence with the Confessions,
one that at first blush seems a real oddity: Why, given his forced march
through Virgil, and his self-proclaimed lack of attainment, did Paulinus
choose to sum up his life, and make sense of it, in the form and language
of classical verse? It is not just the hexameters, however much they have
been criticized; it is also the numerous echoes of Virgil, several dozen

 The classic study on Augustine’s role is Marrou ; Kaster : – provides an overview of the
whole problem. For a valuable new perspective, see also now Watts  as well as his contribution
in this volume.
 josiah osgood

according to the list assembled by Wilhelm Brandes. Why this sort of

language, and not the prose of Augustine, so famously suffused with echoes
of the Psalms? Two answers, somewhat in tension with one another, can be
suggested, bringing us closer to Paulinus’s text and also to the man himself.
Other contributors to this volume point out how a great range of late
antique biographical texts – from Athanasius’s widely read account of the
ascetic Antony to Phocas’s Life of Virgil to Gregory Nazianzen’s two ora-
tions against Julian – can be seen to function as “teaching texts,” aimed to
inspire, and instruct, students of one sort or another. The Eucharisticos
is profitably seen in this light too. Like Augustine also, Paulinus, we have
already seen, aims to instruct – addressing his Roman contemporaries in
Gaul in particular, who, whatever Paulinus’s own exaggerations, had wit-
nessed severe dislocations in the last two generations. The use of Virgilian
hexameters reinforces the didactic goal, insofar as poetry might be thought
to hold greater appeal, especially to younger readers, than (say) a long
prose tract such as Salvian’s Government of God. This need not mean
that Paulinus thought only of a readership of the young; for older readers,
the Virgilian language of the poem, combined with Paulinus’s prominent
recollections of his own schooling, evoked their own experience of edu-
cation – where Virgil held so important a place – and would encourage
reevaluation of it. Paulinus is thus a sort of antischoolmaster.
And yet, while we can acknowledge that there were biographies in verse
(or perhaps even more apposite, the autobiographical poems of Gregory
Nazianzen), the fact is that in electing to write in the form and language of
classical verse, rather than in a simpler prose, Paulinus perforce acknowl-
edged the appeal of classical verse to a reader. And at least some of its
appeal, in fifth-century Gaul, could be separate from the emotional or
intellectual pleasure poetry might give. Now, while Virgil loomed large for
Paulinus, Brandes, in his concordance, also notes a number of parallels with
Ausonius, and here we start to find a second explanation for Paulinus’s
decision to write in verse: Ausonius was not just any poet but, of course,
Paulinus’s grandfather. One of these Ausonian echoes, not in fact noticed

 Brandes : –; cf. McLynn b: . In general, my reading in this last section aims not to
challenge McLynn’s explanation of the seeming divergence between the poem’s form and contents
but to add another perspective.
 See the contributions of Watts, McGill, and Elm.
 Compare here especially McGill’s discussion of Phocas’s decision to write a life of Virgil in verse;
on the appeal of poetry more generally, see the discussion in Cameron  (especially –).
 On Gregory’s poems, again note Cameron  ( and ) and see Meehan  for translation
of three of the most important texts.
 Brandes : .
 The education of Paulinus of Pella 

by Brandes, is especially significant and derives from Ausonius’s description

of his own career as a teacher: “Many in their infant years I brought up,
and, holding them snugly in my lap and soothing their cries, snatched their
tender years from their fond nurses. Then, as boys, with mild warnings
and gentle threats, I lured them to seek through sourness for ripe success
and pluck the sweet fruit of a bitter root” (multos lactantibus annis / ipse
alui gremioque fovens et murmura solvens / eripui tenerum blandis nutricibus
aevum. / mox pueros molli monitu et formidine leni / pellexi, ut mites peterent
per acerba profectus, / carpturi dulcem fructum radicis amarae [–]). The
“sweet fruit” here, after the “bitter root,” is, of course, rhetoric after gram-
mar and literature, the “worthy fruit” Paulinus might have won had he not
been struck with illness and dropped out of school (Euch. ). Paulinus,
on the testimony of his poem, never had training in rhetoric, but he did
have the poetry that his grandfather also taught and wrote, the poetry that,
whatever efforts it cost the young pupil, could paradoxically be a sign of
ease and so of high status, the status of a Roman senator – think of the
Mosella or the Cento nuptialis.
The homage Paulinus pays to poetry by adopting it as the form of his
autobiography must be related to the praise of all the trappings of status
in the early part of the poem. It has been pointed out, more than once,
that the manner in which Paulinus lingers, lovingly, on the days of his
life as a rich young noble in Aquitania seems inconsistent with the rest
of his poem. These scenes are not riven through with the same depth
of anguish we find in Augustine, and they have more of a tone that is
straightforwardly elegiac. It could be that it is part of the poem’s rhetoric
to share these pleasures fully with the reader, only to let the reader see them
knocked away. But why, then, still wish at the time of writing, as Paulinus
did, that the pleasures had been more lasting, along with the “time of
peace” (tempore pacis [–])? The pleasures of life are not, in the end,
empty, Paulinus is at moments honest enough to recognize, and it becomes
difficult not to sympathize with the view put forward by John Matthews
in Western Aristocracies: “The autobiographical poem of Paulinus is cast in
the form of an extended prayer of thanksgiving; yet it is pervaded by, and
does not always try to disguise, a profound pessimism – a sense that, after
all, everything had not happened for the best.”
As Paulinus grasped to put his life into the lines of verse and justify it to
himself and his readers, he inevitably clung to the poetry of his grandfather
 In a stimulating chapter, Mathisen (: –) explores (without reference to Paulinus) the value
of classical literary culture to fifth-century Gallo-Roman aristocrats.
 See most recently Coşkun : –.  Matthews : .
 josiah osgood

and the distinctions that a literary culture can bring. The Latin poet’s
effort to summon up his own past overlapped, too, with an ambivalent
recollection of his greatest predecessor, Virgil. He may have dismissed his
learning, but he did not dismiss it fully. Poetry was still for Paulinus the
mark of a Roman aristocrat’s life of ease, an ease that the poet once knew
and that he showed himself struggling to hold onto in a Gaul then ruled
by barbarians. Christianity alone, as if it were like some powerful force in
nature, did not bend his view of education or that of many others. Western
aristocrats also required other, more concrete, reasons to revise or repudiate
their centuries-old traditions: disinclination for the old public life or a lack
of opportunities to engage in it. Ultimately, the fate of learning was tied
more closely to that of the government.
 chapter 8

Another man’s miracles: recasting Aelius Donatus in

Phocas’s Life of Virgil
Scott McGill

his fave dictis! retegenda vita est

vatis Etrusci modo, qui perenne
Romulae voci decus adrogavit
carmine sacro.
Look kindly on these words!
Now the life of the Etruscan bard must be made known,
who claimed eternal glory for the Latin language
through his sacred poetry.
So the biographer Phocas ends his invocation to the muse Clio that precedes
his hexameter Life of Virgil. A grammarian and teacher at Rome, Phocas
likely dates to the late fourth or fifth century. His Vita, with a twenty-
four-line preface in sapphics, a concluding lacuna, and textual problems in
lines –, is the only example we have of an ancient Virgilian biography
written in verse.
Scholarship on Phocas has consistently traced his Vita back to earthlier
sources than Clio. These are the Virgilian biographies of Suetonius and
It is a pleasure to contribute this essay to a Festschrift for John Matthews, who generously agreed to
advise my dissertation on the Virgilian centos. I am grateful for his willingness to work with me on
that project and for the warm support he has since provided. I thank Robert Kaster, Michael Putnam,
Cristiana Sogno, Ed Watts, and the reader at Cambridge University Press for their suggestions and
encouragement.
 I use the text of Brugnoli : –. On the textual tradition of Phocas’s Vita, see Stok a. All
translations are my own, unless otherwise indicated.
 On Phocas, see PLRE ., s.v. “Phocas” () and Kaster : –. Phocas’s terminus post quem
has been a matter of dispute. While Cassiodorus (GL .., Keil) groups him with grammarians
of the late third and early fourth century, critics have argued that he should be placed later, due to his
citation practices and his debts to Aelius Donatus (more on this to come). So Kaster : , with
bibliography. See Mazhuga  as well. The grammarian Priscian (fl.  CE), meanwhile, provides
a terminus ante quem for Phocas by citing him. But Brugnoli : vi–viii notes that critics date
Phocas more precisely to the late fourth or early fifth century, and argues further for that position.
 Together with some lines of verse in Phocas’s grammatical work, the Vita thus reveals him to have
been something of a poet-grammarian, a category well represented in late antiquity; see Ausonius,
Prof. .– and  (and Ausonius himself ), as well as Alan Cameron :  and –. For ancient
autobiographical texts in verse, see Osgood, in this volume (pp. –) and n.  below.


 scott m c gill

the fourth-century CE Aelius Donatus, who, the evidence is strong, essen-

tially reproduced the now lost Suetonian text. (The abbreviation VSD
conventionally designates Donatus’s work, with its presumed origins in
Suetonius.) While some identify Suetonius as Phocas’s model or maintain
an agnostic stance with regard to which of his precursors Phocas used, the
majority position is that he relied upon his fellow late-antique biographer
Donatus. Lending credence to this viewpoint is Phocas’s treatment of how
Virgil got caught up in the land confiscations that followed the Battle of
Philippi (–). In that passage Phocas echoes the content and language
of the equally full account of the same episode in Donatus’s prefatory
comments to the Eclogues that do not derive from Suetonius. Because the
parallels forcefully indicate that Phocas was indebted to Donatus for this
material, it seems reasonable to conclude that he also turned to that same
source throughout the rest of his biography.
This essay will thus follow majority opinion and suppose that Phocas
depended upon Donatus specifically. I will then center my inquiry on a
particular approach to recasting his model that Phocas displays. For an
extended period in Phocas’s treatment of events surrounding Virgil’s birth,
similarities in expression between his text and the VSD argue conclusively
for his reliance upon that work. Yet mixed in with the passages that Phocas
adapts from Donatus are two scenes without precedent in his predecessor.
No other Virgilian biography that survives from antiquity so dramatically
departs from the Donatan narrative frame. Phocas’s radical deviation from
 Naumann  provides a good summary of the relationship between Suetonius’s and Donatus’s
biographies. For an overview of Virgil’s biographical tradition, see ANRW II...– and EV
∗ –, s.v. “Vitae Vergilianae.” Studies of the Suetonian/Donatan biography that I have found
useful include Rostagni : –; Horsfall ; and Bayer . I use Stok’s  edition of the
VSD.
 Critics who posit Suetonius as Phocas’s source are Strzelecki () and Mazzarino (–: –).
Kaster (: ) is undecided on whether Suetonius or Donatus was Phocas’s model.
 Brugnoli (: vi–vii), who believes that Phocas relied upon Donatus, gives a bibliography of critics
who hold the same view, to which I add Hardie : vii–viii and Callu : .
 See causa in line  and VSD  (following Stok’s numbering); fretus amicorum clipeo in  and
fretus . . . amicitia quorundum potentium in VSD ; and meritis . . . carmen in – and merito
carminum (fretus) in VSD . A less forceful parallel is victor in line  and VSD . The details
in Phocas that Virgil was personally attacked (–) and that Augustus restored Virgil’s land to
him (–) also match up with Donatus (VSD ). Finally, the fullness of Phocas’s and Donatus’s
treatments, which contrasts with VSD –, suggests intertextual contact. (The fact that Servius’s
Vita and the probably fifth-century Vita Philargyriana I each have at least some of these elements,
meanwhile, seems to result from their own reliance upon Donatus; so Stok b: esp. – and .
I therefore consider it unlikely that Junius Philargyrius, whom critics place in the second half of the
fifth century CE, was a source for Phocas – a conclusion that, if accepted, would affect how we date
Phocas. Ziolkowski and Putnam [: ] are inconsistent on this point.)
 H. Naumann : – takes this view. See Vidal :  as well.
 Another man’s miracles 

Donatus amid a part of his Vita modeled upon that forerunner forms the
subject of this chapter. I will first identify Phocas’s sources for his original
material and investigate what his debts suggest about how he interpreted
the VSD and about the movement of content between ancient biographical
texts. The essay will then consider the patently unreal nature of the scenes
that Phocas wove into his Life. Interest will lie particularly in probing
how Phocas set out to utilize rhetorically what he must have known was
fictional material and in exploring issues related to the ancient reception
of the extra-historical anecdotes.
The passage at issue deals with the prodigies that accompanied Virgil’s
birth (–). A web of verbal correspondences signals Phocas’s dependence
upon the VSD when describing three of the marvels: the dream of Virgil’s
mother during the evening before his birth in which a laurel grows with
miraculous speed (–; VSD ); the uncanny serenity of the baby Virgil
(–; VSD ); and the poplar planted by Virgil’s father at the boy’s birth
that also grew with supernatural quickness (–; VSD ). Yet balancing
Phocas’s debts to the VSD are two prodigies he presents with no equivalent
in that source, in other ancient Virgilian biographies, or, as far as I can
discover, in all passages and works on Virgil that survive from the Roman
world. Both come after the account of Virgil’s birth and his preternatural
calmness as a newborn, where the echoes of Donatus are particularly strong.
The most striking is Phocas’s infantem vagisse negant (they say that the
infant did not cry []), which closely resembles the VSD’s ferunt infantem
ut sit editus neque vagisse (they say that the infant, when he was born, did
not cry []); but there are other similarities as well. Upon describing the
pacific Virgil, Phocas proceeds to relate how nature reacted to his birth
(–):

 The most extensive echoes appear in connection with the second of the prodigies, which I will
examine shortly. Other parallels are ramus (; VSD ), populea virga (; VSD ), and brevi tempore
(; VSD ).
 Ziolkowski and Putnam () have done Virgilian scholarship a great service by providing a
sweeping anthology of primary material on Virgil.
 Thus Phocas’s use of editus to describe Virgil’s birth () parallels ut sit editus in VSD , while his
mitis in line  (quo tempore Chelas / iam mitis Phaethon post Virginis ora receptat [at the time when
Phaethon, now gentle, withdraws to the Claws of Scorpio behind the face of Virgo]) matches up with
miti in the VSD’s et adeo miti vultu fuisse (and he had such a gentle expression []). The substance of
Phocas’s fronte serena / conspexit mundum (for with a gentle countenance he looked upon the world
[–]), moreover, corresponds to et adeo miti vultu fuisse. (The resemblance between vultu and
Phocas’s ora in line  is also worth mentioning.) Only the VSD and Phocas share these elements
among surviving ancient Virgilian biographies. Critics who note the similarities between Phocas
and the VSD include Oroz (: ) and Brugnoli (: ).
 scott m c gill

nam fronte serena

conspexit mundum, cui commoda tanta ferebat.
ipse puerperiis adrisit laetior orbis:
terra ministravit flores et munere verno 
herbida supposuit puero fulmenta virescens.
For with a serene expression he looked upon the world,
to which he was bearing so many gifts.
The globe itself altogether happily smiled upon the birth of the child;
the earth furnished flowers and growing lush and green
placed a grassy cradle under the baby in its vernal munificence.

Phocas adapts this passage from a section of Virgil’s own fourth Eclogue
that tells of how nature will respond with abundance to the birth of the
poem’s Golden Age-heralding child (Ecl. .–). The close thematic
agreement between Phocas’s and Virgil’s accounts supports this idea: both
texts describe the earth producing flowers and a cradle lush with vegetation.
Verbal parallels (ipse [] and ipsae [Ecl. .]/ipsa [Ecl. .]; flores [] and
flores [Ecl. .]; munere [] and munuscula [Ecl. .]; and puero [] and
puer [Ecl. .]) and mirroring adjective–noun hexameter framing patterns
(ipse . . . orbis in line  and ipsae . . . capellae in Ecl. .) then confirm the
connection.
What led Phocas to think of the portrayal of nature’s happy response to
the newborn in Eclogues  at this point in his biography? An answer emerges
from the similarities between VSD  and the end of Virgil’s poem. Sources
ranging from late-medieval Virgilian biographies (the Vita Monacensis II
and the Life of Zono de’Magnalis) to modern criticism have noted the
resemblance between Virgil in the VSD and the child in Ecl. .–,
whom the narrator enjoins to recognize his mother with a smile. The idea

 At tibi prima, puer, nullo munuscula cultu / errantis hederas passim cum baccare tellus / mixtaque
ridenti colocasia fundet acantho. / ipsae lacte domum referent distenta capellae / ubera, nec magnos
metuent armenta leones; / ipsa tibi blandos fundent cunabula flores (But, child, for you the earth will
pour forth its first little gifts: ivy spreading everywhere with cyclamen (?) and the Egyptian bean
mixed with smiling acanthus. The goats on their own will bring home their udders swollen with
milk, and the herds will not fear the powerful lions. Your cradle will spontaneously produce sweet
flowers for you). On this act of imitation, see Duckworth : –; Mayer : ; Brugnoli
: –; and Vidal : – and : –.
 Incipe, parve puer, risu cognoscere matrem / (matri longa decem tulerunt fastidia menses) / incipe, parve
puer: cui non risere parentes, / nec deus hunc mensa, dea nec dignata cubili est (Begin, small child, to
recognize your mother with a smile [ten months brought long labors to your mother], begin, small
child: those upon whom parents do not smile no god considers worthy of his table, no goddess of
her bed). On the reading cui non risere parentes, see n.  below.
 Another man’s miracles 

is that the baby Virgil behaves in a manner that corresponds to how the
Virgilian baby is told to behave in the fourth Eclogue.
That Phocas drew the same connection explains why he followed up a
sketch of the infant Virgil with origins in the VSD by recasting a section
of the fourth Eclogue. In this reading, Phocas goes from reworking VSD
 to adding a scene derived from Virgil’s poem to his biography because
he had identified a connection between those models in the figure of
the happy newborn they share. What at first glance look to be discrete
acts of adaptation are, in fact, linked: the recognition of a direct point of
thematic contact between VSD  and the fourth Eclogue steers Phocas to
Virgil’s poem, from which he proceeds to draw an apposite aretalogical
wonder.
Phocas’s use of adrisit in line  (ipse puerperiis adrisit laetior orbis)
bolsters the idea that he saw the links between VSD  and the conclusion
of the fourth Eclogue. While the participle ridens also appears in Ecl. .
(mixtaque ridenti colocasia fundet acantho [and the earth will pour forth the
Egyptian bean mixed with smiling acanthus]), the syntax of Phocas’s term is
closer in form and function to ridere in Ecl. . (incipe, parve puer: cui non
risere parentes [begin, baby boy: those upon whom parents do not smile]).

 Brugnoli (: ) cites the Vita Monacensis II (–). On the fourteenth-century Zono
de’Magnalis, see Ziolkowski and Putnam : –; and see  for the remark on the resem-
blance between the VSD and the fourth Eclogue. A modern critic who recognizes the intertextual
link between the VSD and the end of Virgil’s poem is Horsfall (: ), who understands it as an
example of how details from Virgil’s texts migrated into his Life. (Zono de’Magnalis, meanwhile,
suggests that Virgil had behaved just as the baby does at the end of the fourth Eclogue, meaning
that the poem reflects his autobiographical experience.)
 I take tanta commoda in line  as a further sign of how Phocas linked Virgil in VSD  and the
smiling baby in the fourth Eclogue. Having made that connection, Phocas has the newborn Virgil
bringing gifts to the world, just as the puer in Ecl.  does, and then proceeds to adapt Ecl. .–. I
should add that this obviously does not imply that Phocas interpreted the end of the fourth Eclogue
autobiographically, like Zono de’Magnalis. That reading of the baby, anyway, seems confined to
de’Magnalis in the surviving biographical record.
 As Hadas and Smith (: ) observe, in an aretalogy (i.e., an account of the deeds of a god or
hero) “circumstances of [the subject’s] birth or his death involve elements of the miraculous.” For
more on aretalogy, see Cox : – and –.
 Also worth noting is the (surely overconfident) assertion of Duckworth (: ) that Phocas’s
conspexit mundum in line  “comes directly” from Ecl. ., aspice convexo nutantem pondere
mundum (behold the world nodding with its arching weight). Vidal (: ), meanwhile, cites
the similarity between laetentur in Ecl. . (aspice, venturo laetentur ut omnia saeclo [behold, how
all things rejoice in the coming age]) and laetior in Phocas’s line . See Bayer :  as well.
If we accept this evidence for imitation (though I am not sure that we should), further grounds
emerge for supposing that Phocas was recalling more in the fourth Eclogue than lines – and, in
fact, was thinking of the late material in the poem.
 Duckworth (: ) also calls attention to the resemblance between Phocas’s adrisit and Ecl. ..
 scott m c gill

The appearance of puerperiis as the object of adrisit then connects Phocas

further to Virgil, who combines puer and ridere in Ecl. .. Phocas’s
language thus has more than one element capable of corroborating what
the progression of topics in his work suggests: that the end of the fourth
Eclogue served as a bridge between one model, VSD , and another, Ecl.
.–.
If Phocas moved between Donatus’s biography and the fourth Eclogue
as I suggest, his actions provide an example of how adaptation can be
a barometer of interpretation. While versifying material he found in an
earlier Life, Phocas anticipates later biographers and modern criticism and
spots a likeness between the baby Virgil and Virgil’s baby in Eclogues .
What ensues is a change to Donatus’s biography that grows out of Phocas’s
reading of it.
Following the account of nature’s happy and lush reaction to the baby
Virgil, Phocas describes another miraculous event that does not appear
in the VSD. Upon commenting in line  on the factual accuracy of the
scene he is about to describe (praeterea, si vera fides, set vera probatur
[moreover, if the evidence is true, as it is proved to be]), Phocas tells of
how bees alighted on the resting Virgil’s mouth and made a honeycomb
there. Phocas proceeds to note that antiquity tells of the same marvel in
connection with Plato, and to use that detail as a springboard for extolling
Rome (–):
laeta cohors apium subito per rura iacentis
labra favis texit dulces fusura loquellas.
hoc quondam in sacro tantum mirata Platone 
indicium linguae memorat famosa vetustas.
set Natura parens properans extollere Romam
et Latio dedit hoc, ne quid concederet uni.

 Phocas would seem to take puerperium from VSD  (virga populea more regionis in puerperiis eodem
statim loco depacta [a poplar branch immediately planted in the same place where the baby was born,
as was the custom in the region]). But Ecl. . could have prompted him to think of puerperium
and to use it where he does. The syntactical parallel between Phocas and Virgil is strongest if Ecl.
. reads cui non risere parentes, which we find in the MSS (P, R, and w) and in Servius (ad
Ecl. .), rather than the preferable qui non risere parenti (on this textual issue, see Clausen :
). Virgil would use ridere with a dative relative pronoun whose antecedent is puer, while Phocas
has arridere with the dative puerperiis. Given the evidence that cui non risere parentes was known
in late antiquity, it seems entirely plausible that Phocas’s text likewise contained that reading.
(It bears mentioning that the Vita Monacensis II and Zono de’Magnalis also read cui non risere
parentes.)
 My reading contrasts with that of Brugnoli (: ), who proposes that the descriptions in the
VSD of the laurel and poplar that grew with wondrous speed occasioned Phocas’s turn to the fourth
Eclogue.
 Another man’s miracles 

A happy band of bees swarming suddenly over the fields

covered the lips of the resting Virgil with a honeycomb,
lips that would pour forth sweet speech.
Having once wondered at this only in connection with holy Plato,
antiquity, conferring renown, records the token of eloquence.
But our parent Nature, hastening to elevate Rome,
also gave the prodigy to Latium, so that it not yield at all to anyone.

While the VSD records that Virgil’s father kept bees (), it seems unlikely
that the earlier Life triggered Phocas’s story about the labial honeycomb,
given how entirely the wonder differs in content and function from Dona-
tus’s remark. A more plausible scenario has Phocas simply moving in his
literary memory from one text describing a marvel in nature to a second
text that does the same. The biography of Plato indeed contained the
anecdote on the bees; and we can understand Phocas’s reference to that
work as a quasi-footnote disclosing his source for the apian prodigy.
Accounts of bees weaving honeycombs on (and in) authors’ mouths
appear in ancient biographical texts other than the Life of Plato and Phocas’s
work. Thus in the Greek tradition, biographies of Hesiod and possibly of
Sophocles and Menander contain the anecdote, which finds its way into
Pindar’s Lives as well. On the Latin side, bees alight on Lucan’s mouth in
the sixth-century-CE Vacca’s biography. Like Phocas, Vacca notes further
both that the prodigy augured his subject’s later eloquence and that it
prevented the Roman world from lacking a wonder ascribed to a Greek (in

 I follow Vidal :  and Ziolkowski and Putnam :  in translating tantum adverbially. For
famosa vetustas meaning “antiquity, conferring renown,” see Lucan, BC . (aevi veteris custos,
famosa vetustas).
 Phocas himself does not reproduce this detail in VSD . It is to my mind equally unlikely that
Virgil’s discussion of bees and beekeeping in Georgics  led Phocas to include the anecdote, because
of the sharp thematic and contextual differences between the texts.
 Just maybe, moreover, Virgil’s reference to honey in Ecl. . (et durae quercus sudabunt roscida
mella [and hard oaks will ooze dewy honey]) provided the mnemonic cue.
 See Lefkowitz :  n. , with bibliography. Other Latin authors who cite the honeycomb story
in connection with Plato are Cicero (Div. . and .) and Valerius Maximus (.. ext. ).
 I disagree here with the proposal of Strzelecki : , that Phocas derived the bee story from
some other, now lost Virgilian Vita. The question arises whether Phocas had direct contact with
the story in the Greek Life of Plato or knew of it through a Latin intermediary or simply through
hearsay. (For evidence of his knowledge of Greek, see GL ..– and ..–..) But in
any event, the citation of Plato reveals Phocas’s familiarity with the story’s place in the Platonic
biography.
 So Riginos :  n. . Deserving notice here as well is Christodorus (ca.  CE), who relates in
a poem describing a statue of Pindar (AP .–) that the poet once awoke to the buzzing surprise,
and who describes a late Hellenistic bronze statue of Homer found in the Baths of Zeuxippos in
Constantinople that has a bee bringing forth a honeycomb from Homer’s mouth (AP .–).
 scott m c gill

Vacca’s case, Hesiod). So, too, Paulinus of Milan (or Paulinus the Deacon)
includes the portent in his early-fifth-century Life of Ambrose (.–),
where the bees are once more explicitly connected to future eloquence.
Although Phocas may well have been aware of the anecdote on the
honeycomb only as it related to Plato, he thus connects up with the broader
biographical tradition in giving space to a commonly described wonder.
In the process, he joins with varied sources to illustrate that the story
was a topos in ancient literary biography. By indicating that he derived
the scene from a Platonic source, moreover, Phocas sets his passage up
as a signposted example of how a biographer might consciously transfer
a motif from one person’s Vita to another’s – a practice with analogues
elsewhere in antiquity. A writer recalls a predecessor in the biographical
tradition and applies material he extracts from that source to a new life
story.
Immediately after the honeycomb episode, Phocas casts his attention
back to the VSD and adapts with verbal echoes Donatus’s account of
how the poplar that Virgil’s father planted at the boy’s birth grew with
miraculous speed. The return to Donatus makes it clear that Phocas’s
approach was to break up with fresh details the account that served as
his primary model for the Virgilian portents: the biographer relies upon
the VSD for the first two and the last of the wonders he presents, but
between them interjects scenes that are external to that source. In opting
both to reproduce and to expand upon Donatus, Phocas demonstrates that
he considered him a guide for the handling of the prodigies, but not an
authority demanding strict adherence.
It is not just in the passages adapted from the fourth Eclogue and Plato’s
Life that Phocas’s content differs from Donatus’s. Along with omitting a
good amount of material found in the VSD, Phocas changes pieces of
information – for example, the occupation of Virgil’s father and name of

 As Brugnoli (: –) notes. The parallels in how the biographers comment on their passages
raise the suspicion that Vacca was in fact following Phocas.
 Operabatur enim iam tunc Dominus in servuli sui infantia, ut impleretur quod dictum est: favi mellis
sermones boni (Already then, in his little servant’s infancy, God was hard at work, so that the baby
would be filled with his language; for good speech consists of the hive’s honey). Brugnoli : 
again provides the reference.
 See Fairweather : esp. –. Riginos (: ) discusses the same topic and refers to Phocas,
but does not discuss him as I do. See also Reifferscheid : .
 For the verbal parallels, see n.  above.
 Thus VSD – (and some of  and ) and most of – have no equivalent place in Phocas’s
work. Phocas also compresses the discussions of the Culex (VSD ) and of Virgil’s fateful trip in 
BCE (VSD ).
 Another man’s miracles 

Virgil’s mother, the time it took him to write his poems, and the reason he
departed on the journey where he contracted his fatal illness – and greatly
fills out the account of Virgil’s experiences in the land confiscations after
Philippi. But nowhere else in his Vita does Phocas present entire anecdotes
having no antecedent in Donatus, as he does when cataloging the wonders
that arose around the baby Virgil. This indicates that Phocas felt most free
to depart from the substance of the VSD when dealing with the prodigies.
The result is a striking combination of change and intertextual continuity,
with Phocas presenting non-Donatan anecdotes within a framework that
he takes from Donatus.
It is scarcely credible that Phocas believed in the factual accuracy of
the non-Donatan wonders he presents. Ancient interpreters of the Eclogues
saw Virgil himself behind some of his characters and held that in certain
passages he was alluding to his own experiences. Yet the account of nature’s
response to the child’s birth in the fourth Eclogue was not so allegorized
and, anyway, seems to have been too clearly a lavish laudatory fiction to
induce any such biographical speculation on Phocas’s part. Considering
this and the textual evidence that Phocas turned from Donatus’s Vita to
Virgil’s poem, the conclusion presents itself that he was simply led via a
metonymic child to what he knew to be an extra-biographical passage,
with which he then chose to expand upon the VSD’s narrative.
Having ascribed to the newborn Virgil an experience with no grounding
in his actual biography, Phocas then continues along the same path. Or so
his consecutive turns to separate external sources and his quasi-footnote in
lines – lead one to suppose: Phocas follows up a first literary embellish-
ment on the story of Virgil’s infancy with another. Nor is the line praeterea,
si vera fides, set vera probatur (), with which Phocas introduces the bee
anecdote, an obstacle to this interpretation. On the contrary, I would argue
 See lines – and –. Other small changes come when Phocas writes his own version of Virgil’s
epitaph on Ballista (–) and when he identifies Virgil’s teacher as Siro (). Finally, Phocas’s poem
in several places differs from the VSD because he interjects narrator’s comments on his material in
the form of apostrophes, rhetorical questions, and explanatory glosses (see –, –, –, and
– [as well as –]).
 See lines – and n.  above.
 Notable examples are the comments of Servius ad Ecl. . and ., on the land confiscations (cf.
VSD – and –), and ad Ecl. ., on how Corydon stands in for Virgil (cf. VSD ). See also ad
Ecl. . (along with .) and VSD  for programmatic statements on reading Virgil allegorically.
I thank Patrick Glauthier for his help with this topic.
 Evidence that ancient Virgilian scholarship (which Phocas, as a grammarian, was certainly in a
position to know) recognized the poem’s laudatory character comes from Servius, ad Ecl. .,
where he describes Virgil’s material as a laudatio, and ad Ecl. ., where he further notes the
presence of laus.
 scott m c gill

that Phocas, having decided to use a scene with no biographical founda-

tion, went on to acknowledge questions about its truth precisely because
he had the material’s lack of historicity in mind. Because the proof that
affirms the story’s accuracy has to be the eloquence that, Phocas relates,
Virgil later displays (labra . . . dulces fusura loquellas []), moreover, we can
suppose that the biographer was aware that he was dealing in symbolic
truth, even as he also preserves space for a naı̈ve reading that takes him at
his word.
Given poetry’s traditional association with fiction, to which Phocas
himself alludes in the preface to his work, we might suspect that the
biographer felt free to include his new scenes because he was writing his
Vita in verse. Yet seeing that Phocas strays so sharply from Donatus’s narra-
tive only when treating the prodigies described in VSD –, it seems better
to conclude that something about that material specifically emboldened
him to adapt it as he did. One possibility is that Phocas put no stock in
the historicity of Donatus’s wonders and felt that this gave him the license
to make them even more unreal. It is entirely plausible that Phocas consid-
ered factual most of the information about Virgil’s life that he found in the
VSD. This would have been a natural stance to take in Latin antiquity,
where the sharp skepticism that modern scholars display toward so many of
the elements of the Donatan account has no parallel, and where several
biographical details appearing in the VSD are accepted as fact in other
sources. But Phocas could have excluded the VSD’s prodigies from his
historicist reading and denied their veracity. Such an approach appears to
have a partial precedent in Donatus himself, who introduces the story of
the serene newborn Virgil with ferunt, “they say” (VSD ). The verb dis-
tances the biographer from the account and indicates that he cannot vouch
 I will return below to the possible responses to Phocas’s material. Suffice it to say now that we should
not discount the possibility of such naı̈ve readers, given the widespread credulity that individuals
in Latin antiquity showed toward even extravagant portents. A late antique illustration of that
credulity comes from the grammarian Priscian, Praeex. .
 Phocas’s variation on this age-old idea in classical literature comes in his preface (–): sola fucatis
variare dictis / paginas nescis, set aperta quicquid / veritas prodit, recinis per aevum / simplice lingua
(You [Clio] alone do not know how to vary pages with false words, but whatever unvarnished truth
presents, you repeat it through the ages with plain speech). The words fucata dicta suggest the
distortions of dressed-up language, i.e., of poetry. At the same time, Phocas claims that history and
poetry are compatible; see lines – of his preface.
 To follow up on the previous note, this would mean that the invocation to Clio is broadly consistent
with his understanding of his text’s historicity. Phocas programmatically emphasizes historical truth
in his preface and feels that his narrative, like his model Donatus’s, generally contains it.
 Horsfall  is a notable example of a skeptical modern reader.
 The allegorical readings of the Eclogues (see above) provide examples. For other instances, see e.g.
Martial, Ep. ..–, ..–, and ..–; Apuleius, Ap. .; and Macrobius, Sat. ..–
and ... (I omit other ancient biographies of Virgil from consideration.)
 Another man’s miracles 

for its truthfulness. For his part, Phocas, who reworks Donatus’s ferunt
(see negant in line ), could have more definitively rejected the anecdote’s
reality and then viewed the other portents with the same suspecting eye.
Because the prodigies were not historical anyway, Phocas might have pro-
ceeded to reason, there was no need for him to refrain from incorporating
more fiction into his own treatment.
At the very least, Phocas’s actions point to someone who deemed the
account of the marvels suitable to the kind of fictional expansion he pursued
in ways that other elements in the biography were not. This openness to
fabricating wonders matches up with Menander Rhetor’s statement in
his treatise on the imperial encomium that one should not hesitate to
invent miraculous stories in connection with an emperor’s birth. Like
Menander, Phocas leaves room for making up miracles when describing
his infant subject and, in the process, shows that an ancient biographer
no less than an ancient theorist of panegyric considered it possible to get
creative with such material.
The wonders with which Phocas fills out the VSD’s account have a
further link to encomium in how they operate rhetorically: through them
Phocas aims to add laudatory elements that underscore Virgil’s extraordi-
nary nature and, indeed, to make him look even more special than he did
in Donatus. In the case of the passage derived from Eclogues , Phocas also
engages in intra-Virgilian play, as he uses Virgil to describe Virgil. Yet this
blending of elements does more than just produce a kind of mise en abyme

 Ferunt thus seems to function somewhat similarly to how vulgatum est does in VSD . There
Donatus notes that it was “common gossip” that Virgil had an affair with Plotia Hieria but refuses
to credit it and, indeed, cites Asconius Pedianus to refute it. I should note as well that because
siquidem in VSD , which introduces the story of the poplar (et accessit aliud praesagium, siquidem
virga . . . ), appears to be epexegetic and to mean “in that” (OLD s.v. “siquidem” []) rather than
to be a strong conditional and to mean “if it is really possible that” (OLD s.v. “siquidem” [a]), we
cannot take it as a mark of further skepticism.
 K‹n m•n § ti toioÓton perª t¼n basil”a, –xerg†sai, –†n d• o³on te § kaª pl†sai kaª poie±n
toÓto piqanäv, mŸ kat»knei (if there is anything like this [the miraculous events associated with
Romulus and Cyrus, among others] in connection with the emperor, work it up; if it is possible
to invent, and to do this convincingly, do not hesitate) (Bas. Log. .–). I use the text and
translation of Russell and Wilson : –. See n.  below as well.
 It bears observing that as part of the late antique reception of the fourth Eclogue, prose and
verse panegyrists recast its lines, including Ecl. .–, when describing how nature responded
to the birth of an emperor or other distinguished personage. Examples of later-fourth- and early-
fifth-century panegyrists who rework Ecl. .– are Symmachus, Or. . (to Gratian in /)
and Claudian, Carm. Min. .– (the Laus Serenae). Perhaps these developments had come
to Phocas’s attention, so that he recognized that his reworking of Ecl.  paralleled a practice
belonging to the conventional storehouse of panegyric devices. But obviously, all we can do here is
speculate.
 scott m c gill

effect, as it works alongside the honeycomb episode to elevate Virgil to

aretalogical heights exceeding those in Donatus.
Consistent with this effort to create a larger-than-Donatan Life of Virgil
are changes that Phocas makes to the VSD in his treatment of the prodigy
of the poplar. Whereas Donatus in VSD  has the shoot growing so fast
that it quickly equals the height of far older trees, Phocas describes a
scene in which the poplar stands taller in a brief time than all the aged trees
around it (tempore quae nutrita brevi, dum crescit, in omen / altior emicuit
cunctis, quas auxerat aetas [–]). Making this all the more fantastic is that
Virgil’s father had planted the branch in “sterile sand” (populeam sterili
virgam mandavit harenae []), a detail that Donatus omits. The ways in
which Phocas’s changes to the VSD augment the astonishing nature of the
prodigy point to the same overarching intention that the increased number
of miracles implies: to exalt Virgil with a greater intensity than Donatus
does. Having surely picked up on how in the earlier biography the wonders
stand as omens of the greatness Virgil would achieve, Phocas takes steps
to enhance Donatus’s message.
Of course, where Virgil will ultimately display the sublimity that the
portents augur is in his poetry. Phocas points precisely to this outcome in
the miracle of the bees when he connects the favi to the future sweetness
of Virgil’s language, thereby drawing a link between the story and his
subject’s poetic excellence. The biographer likewise makes it clear that
the miracles portended Virgil’s literary achievement when he asserts that
the dream vision of the laurel made Virgil’s mother sure of his future as
a writer (–), and when he sums up his account of the prodigies by
observing that they together led to the decision to commit Virgil to the
Muses and, thus, to show him the way to eternal fame (–).
 In the bee anecdote, Phocas also has the secondary aim of speaking up for the Latin world against
the Greek, a goal that he displays elsewhere as well (–).
 Ita brevi evaluit tempore, ut multo ante satas populos adaequavisset ([the poplar branch] grew in such
a short time, that it equaled in height the poplars planted much earlier).
 Along with the logic of Donatus’s account, the remark in VSD – that the uncrying Virgil provided
a certain glimpse into his unusually prosperous destiny (ut haud dubiam spem prosperioris geniturae
iam tum daret) would have alerted Phocas to this message.
 On the association in Greco-Roman culture generally of bees and honey with poetry, see Waszink
. (Obviously, the other examples of the miracle of the bees in ancient biographies of poets
exemplify this association as well.)
 O sopor indicium veri! nil certius umquam / cornea porta tulit. facta est interprete lauro / certa parens
onerisque sui cognoverat artem (O sleep, portent of truth! The gate of horn never bore anything
surer. His parent was made certain through the interpreter the laurel and knew the nature/make-up
of her unborn child [–]). The message is clearly that the laurel, the symbol of the literary arts,
made Magia Polla aware that her son would be a literary artist, and a crowning one at that.
 Haec propter placuit puerum committere Musis / et monstrare viam victurae in saecula famae (On
account of these things it was decided to commit the boy to the Muses and to show him the way
 Another man’s miracles 

The manner in which Phocas amplifies the VSD’s account and makes
Virgil’s literary achievement all the more miraculously foretold and, there-
fore, all the more uncanny naturally implies a desire to communicate to
others just how wonderfully extraordinary a poet his subject was. But for
whom exactly was Phocas lionizing Virgil? What audience did he anticipate
for his Vita? A reasonable assumption is that Phocas intended the poem for
general circulation, as he did the Ars de nomine et verbo, a grammatical
treatise that he published. Certainly, moreover, the preface to the Vita
suggests literary ambitions and displays a generalized, programmatic mes-
sage in ways that fit with a text meant for dissemination in the wider world.
Phocas might have in particular targeted as readers his fellow grammarians
and those educated in the schools of grammar, in Rome and elsewhere.
It seems natural that he would have expected those individuals to have
an interest in the author who held a central place in the grammatical
curriculum, as he had since the first century BCE.
Other potential recipients of Phocas’s Vita were his students. As Donatus,
Servius, and other sources in Virgil’s reception indicate, instruction in
Virgilian poetry included an examination of the author’s biography. If
the order of study laid out in the evidence is any guide, teachers in the
grammatical schools provided accounts of Virgil’s life before turning to
the study of his text – an order perhaps reflected in the way Virgilian
Vitae precede Donatus’s and Servius’s commentaries. (One also imagines
to fame that would live through the ages [–]). Warranting mention here is Servius, ad Ecl. .,
who notes that Virgil’s reference to errantis hederas portends the child’s future as a poet (nam hederae
indicant futurum poetam). Could Phocas have been aware of this piece of Virgiliana when deciding
to recast Ecl. .–?
 Phocas displays the same desire in lines – and –.
 See GL ..–, ., and .–, with Kaster : .
 By publication in the ancient context, I mean simply putting a work into circulation, whether
through friends or through channels like bookstores and libraries, without any restrictions on who
could read and copy it.
 Phocas’s presumed Christianity (see Kaster : ) does not imply that he was writing for a largely
Christian audience, as Mazzarino (–: –) suggests. Certainly works like Phocas’s could have
been meant for non-Christian audiences or for Christians like Phocas who had secular, classical
interests. Nor does Phocas’s description of Virgil’s poetry as a carmen sacrum (praef. ) complicate
the issue. What that phrasing demonstrates is simply that Phocas, like other lay Christians, could
represent Virgil in reverent and even quasi-religious terms. On this practice generally, see Alan
Cameron : –.
 On Virgil in the schools, see, e.g., Bonner : –; Kaster : ; and Comparetti :
–.
 See VSD  and the opening of Servius’s biography. For examples from the Expositiones and
Periochae, see Ziolkowski and Putnam : – and –.
 See again VSD  and Servius’s Life.
 Donatus, moreover, dedicates his work to Lucius Munatius, apparently a grammarian himself and,
therefore, someone who would have been in a position to recognize how Donatus reflected what
occurred in the schools.
 scott m c gill

that they would return to the subject as necessary.) Seeing that Phocas
was himself a teacher, the possibility emerges that he gave his biography
just such a scholastic use. In that setting, Phocas could have intended
his hexameters to have a particular function: to make the story of Virgil
more attractive, compelling, and memorable to students. The notion that
poetry sweetens material and makes audiences more receptive to it has
a long history in the ancient world and continues to find expression in
late antiquity. Of particular relevance is Gregory of Nazianzus’s assertion
that one writes in verse to appeal to young people. Engaging the young
through poetry seems a plausible consideration for Phocas: the teacher
would have sought to present a Life that held his students’ attention and
excited them about the poet figuring so prominently in their schooling,
even if it familiarized them with only a selection of the VSD’s details. The
prodigies that Phocas adds to his biography, moreover, would have been
entirely suitable to the scholastic context. Not only does Phocas expand
upon Donatus in a manner liable to capture his students’ fancy, but he also
promotes Virgil the school author by affirming his special status through
more elaborate, aggrandizing imagery than what appears in the VSD.
When encountering the non-Donatan miracles, students and general
readers alike would have had the choice of accepting or rejecting their his-
toricity – and to judge by the Ars de nomine et verbo, which Phocas wrote for
broad circulation and with an eye to the schools, he could well have meant
the biography both for the public and for his charges. We might won-
der if grammatical students would have been in a position to identify the

 Lucretius, DRN .– and .– provides the locus classicus. For a late antique example, see the
fifth-century Christian poet Sedulius, who explains that he wrote his Carmen Paschale in verse to
attract those educated in the secular schools (CSEL..–.).
 See Carm. ...–, a poem on his own verses, and Carm. .., Gregory’s verse autobiography.
Alan Cameron :  cites and discusses these passages.
 Obviously, Phocas would have been seeking to appeal to young people for reasons different from
those of Gregory, who claims to want to make his Christian message more palatable. As Watts
relates in this volume (pp. –), moreover, ancient teachers used philosophical biographies in a
roughly similar fashion.
 How thorough the biographies presented to the students would have generally been is, anyway,
uncertain, considering that the surviving ancient examples are all shorter (and in some cases [e.g.,
in Servius] dramatically so) than the VSD.
 The attention Phocas gives to Virgil’s verbal excellence in the bee anecdote, moreover, accords with
the scholastic focus upon Virgilian language and style. Vidal (: ) connects Phocas’s biography
to the schools, though not in the terms that I do.
 See GL .., .–, and .–, with Kaster : .
 If we accept this possibility, it must still remain unclear whether Phocas first wrote the Vita for his
school and later put it into circulation, perhaps with some embellishments – one might question,
for instance, if Phocas would have written for his students the preface in sapphics – or first wrote it
for general circulation and later decided to use it in his school.
 Another man’s miracles 

non-biographical origins and fictional character of the scenes that Phocas

added to his Life. The supposition seems safe, however, that members of
the public would have recognized that the miracles were not historical,
and that Phocas anticipated such a response. Knowledge of Eclogues  is
an important consideration. As Phocas must have imagined, at least a
good portion of his general readership – and not just fellow grammarians
but educated adults on the whole – would have been informed enough
about Virgil to recognize the biography’s parallels with Ecl. .–. In a
world where, as the evidence (and logic) indicates, the reach of biographical
allegory did not extend to that Virgilian passage, not to mention where
imitatio was so common, the default response would presumably have been
that Phocas simply recast a vignette in Virgil’s poetry with no connection
to Virgil’s real life. Because it is reasonable that an author working within
those cultural parameters expected such a reception from his Virgil-savvy
readers, the conclusion also follows that Phocas envisioned audience mem-
bers who would accept the presence of a non-biographical prodigy and
activate the laudatory function with which he had invested it.
Phocas would appear to cue his readers who rejected the biographical
reality of the Eclogues-derived wonder to respond in the same way to the
ensuing miracle of the bees when he brings up the topic of the latter anec-
dote’s truth in line  (praeterea, si vera fides, set vera probatur). To return
to an earlier point, Phocas’s words are consistent with an author who rec-
ognized that the story was not literally vera. But appearing where they do,
the remarks can also be understood as an authorial signal to the audience
members who had just encountered what they saw to be a fictional anec-
dote to continue to be skeptical and to understand that Phocas was again
 To return to a point made in n.  above, I hesitate to rest my argument upon the inherent
implausibility of the prodigies, considering the evidence we have for extensive credulity toward
wonders, including fantastic ones, in antiquity.
 We cannot be as confident that Phocas’s students would have known the passage, because it is
unclear whether Phocas would have taught the Eclogues as he no doubt did the Aeneid. Hence,
the students might not have had the knowledge needed to locate the literary origin of the miracle
derived from Ecl. .–. See, too, n.  below.
 If audience members knew the VSD, as at least many must have, even stronger conditions would
have been in place for them to respond in this way. The idea is that they would have been in a
position to conclude that Phocas strayed from his biographical model and the transmitted details
of the poet’s life.
 Surely Phocas also intended to provide those who recognized his reuse of the fourth Eclogue with
a moment of intertextual entertainment at the realization of how two Virgilian planes meet in
his poem. It also bears recalling here Menander Rhetor’s piqanäv in his discussion of invented
miracles. In relating that the fictional prodigies need to be believable, Menander suggests that the
panegyrist should aim to have them be believed. Thus while Phocas and Menander are similar in
how they leave room for fictional wonders, they can also be seen to offer divergent information on
the reception of the material’s truth-value.
 scott m c gill

incorporating an unreal prodigy into his work in order to brighten even

more the miraculous sheen around Virgil. (The citation of Plato would
have then offered them a clue that the scene had a specific literary prove-
nance.) This interpretation implies further that Phocas expected readers
who, having picked up on the fictionality of the scene, would recognize
the symbolic nature of the proof he mentions in line  (set vera probatur).
The idea is that Virgil’s verbal excellence confirms the metaphorical truth
of a wonder that figures Virgil’s possession of such linguistic virtue.
Of course, by raising the issue of veritas in line , Phocas signals his
awareness of his text’s affiliation with history. This way of thinking finds
even stronger expression in Phocas’s preface, where he invokes Clio and
presents her as the Muse who preserves the past and historical truth.
As the prodigies of happy nature and the bees indicate, however, Phocas
did not feel completely bound to and by biographical accuracy. Among
the extant Virgilian Vitae from antiquity, Phocas alone expands upon
the VSD’s portents with those additional scenes. Yet this only makes the
largely obscure author working in the shadows of Donatus worthier of
notice, in that he alone reveals a particular approach both to writing the
poet’s Vita and to rewriting the VSD that a Virgilian biographer could
pursue. The result is an example of how an ancient author could couple
his work with history and, indeed, quite possibly understand his narrative
as fundamentally historical while still using fiction as a means of recasting
and even inventing certain episodes.
To suppose that Phocas knew he was dealing with fictional material
when he incorporated the non-Donatan prodigies into his Vita, as well as
that he anticipated audience members who would do the same, is to add to
the story told by other critics of the place of fiction in the Lives of ancient
authors. One can distinguish between the results of naı̈veté and deliberate
rhetorical embellishment and conclude that a biographer might present
fabricated details in order to highlight for his responsive readers something

 As noted earlier, however, Phocas still leaves room for naı̈ve readers to accept that the story has
been proven true. Perhaps grammar students would have been among them; and if they were also
credulous toward prodigies and failed to see the reference to Plato as a quasi-footnote, they could
have missed the fictionality of the scene.
 See nn.  and  above. This runs counter to how Plutarch in the Life of Alexander () and
Cornelius Nepos in the Life of Pelopidas () separate biography and history. One way I would not
want to take line  or the emphasis on truth in Phocas’s preface is as a sign that he was anxious
about meeting disapproval in some quarters at his use of fictional anecdotes. Certainly an author
who must have believed that at least a significant portion of his audience would see that he had
incorporated fiction into his text seems to have been at ease with what he was doing.
 I echo Bowersock :  in a discussion of the varied (and sometimes paradoxical) relationships
between fiction and history that ancient writers establish more broadly.
 Another man’s miracles 

about his subject. Exactly how much fiction Phocas imagined that his
audience would see in his section on the prodigies is debatable: perhaps he
expected to have that reaction limited to the two new events he describes or
perhaps, by contact and analogy, he surmised that readers would consider
the dream vision, the serene newborn Virgil, and the fast-growing poplar
to be just as suspect as the other wonders were. But what we can conclude
is that he believed his account of the portents was a place to do more than
just record the facts of Virgil’s life story, and that he assumed an audience
that would pick up on what he was doing. One of Phocas’s aims in his Vita
was to celebrate Virgil the poet; and fictional miracles were a tool at his
disposal to do just that.

 I paraphrase Lefkowitz (: ix), who notes the difficulty of making the distinction I describe
(although she uses “deceit” rather than “deliberate rhetorical embellishment”) in other ancient
biographies. Lefkowitz and Fairweather  remain fundamental to the study of fiction in poets’
biographies in antiquity.
 To revisit an earlier topic, it is tempting to think that Phocas himself questioned the truth of the
VSD’s portents – a line of approach that he could have envisioned audience members replicating.
 chapter 9

Gregory of Nazianzus’s Life of Julian revisited

(Or. 4 and 5): the art of governance by invective
Susanna Elm

For [Ammianus, and others of his mind], much more was at stake than
the acceptance of military defeat. The death of Julian was nothing less
than the death of their hero, and a fatal blow to their hopes that the
Roman empire might be renewed upon the principles of an earlier
age.
[I]t is hard to imagine a writer more responsive to the issues and
personalities of his time, and hard to think of a topic on which,
however peripheral to his own preoccupations, he does not make
some contribution to our understanding.
John Matthews, The Roman Empire of Ammianus, x

On  December  CE Flavius Claudius Julianus, aged twenty-nine,

arrived in Constantinople as sole ruler of the Roman empire. A few months
later an equally young man from a small city in Cappadocia apologized
to his family and friends for a choice that few had anticipated: he wished
henceforth to lead “the true philosophical life.” Two men, two signal events
in their personal lives that stand paradigmatic for a phenomenon central
to John Matthews’s work: the later Roman empire, seen either as a “new
age” described with “more or less traditional way[s]” or as a period, as once
again argued, of decline and fall.
The fate of the late Roman empire has captured the imagination of his-
torians for a very long time (as the introduction suggests, John Matthews’s
more or less explicit interlocutions with Gibbon merit a study in them-
selves), and much of that fascination found focus and drama in the person
of Emperor Julian. Julian’s fate is, of course, inextricably linked to that of
another historian dear to John Matthews, Ammianus Marcellinus. Thus,
Charles de Montesquieu’s extravagant praise of Emperor Julian in his Con-
sidérations sur les causes de la grandeur des Romains et de leur decadence from
An earlier version of this paper was delivered at the University of British Columbia, in the presence
of John Matthews. I would like to thank Mark Vessey and all the conference participants for their
helpful suggestions.
 Matthews : ix–x.


 susanna elm

, which represented a rare exception to the general view in which

Julian was held at that time and eventually led to the reassessment of his
person we now find in modern scholarship, resulted directly from the redis-
covery of “the noble panegyrick made by Ammianus Marcellinus on this
prince, lib. XXV.”
Between  and  eight printed versions of Ammianus’s Res gestae
were produced that ushered in a new view of Emperor Julian and what he
stood for. Influenced by Ammianus, Jean Bodin argued in his Methodus
ad facilem historiam cognitionem, in , that Ammianus’s presentation of
Julian ought to be a model for religious tolerance as applied both to the
“present age” and to one’s judgment of historical persons, a clarion call
taken up by Montaigne, who referred twice to Ammianus in chapter  of
the second book of his Essays of  to demonstrate that great (ancient)
men such as the emperor “dubbed Apostate” were indifferent to invective
(ch. ), and that even adversaries of “our” religion could have positive
characteristics and lead a worthy and virtuous life.
Ammianus’s “panegyrick” of the emperor Julian as an ideal philosophical
ruler and a model of “temperantia, prudentia, iustitia, fortitudo” (..)
only slightly impaired by his overeager attachment to religion had faced stiff
competition before gaining wider acceptance, even after Bodin, Montaigne,
and Montesquieu. The reason for the slow rehabilitation of Julian “dubbed
the Apostate” was not so much due to the fact that prior to John Matthews
Ammianus had remained “merely a relatively, rather than an almost totally,
neglected author.” What Ammianus’s portrayal of Julian was up against
was instead the picture of the emperor as painted by Gregory of Nazianzus,
the young man from Cappadocia who was Julian’s contemporary and, as
it turned out, nemesis.
In two orations against Julian, written after the emperor’s death, Gre-
gory of Nazianzus shaped Julian as “the Apostate,” as a symbol of divine
providence sent to punish but also to reveal divine clemency. In the person
of Julian, God had sent the Christians a new persecutor, too wily to shed

 In his view, it had been Julian’s “wisdom, and perseverance, joined with œconomy, conduct, and
valour, and prospered by a noble series of actions” that had repelled the Germanic barbarians, made
“his name . . . a terror as long as he lived,” and hence postponed the eventual overrunning of Rome,
a sentiment repeated by Voltaire (Secondat : –; Voltaire : –; Mervaud ).
 Secondat : .
 One of the first to rediscover Ammianus’s Julian was Lorenzo de’ Medici, who owned one of the few
manuscripts and who incorporated this new Julian into a stage play on the fate of several martyrs
(Nesselrath ). For the rather tenuous manuscript history of the Res gestae, see Seyfahrt .
 Mesnard : .  Screech : –; Demarolle .
 Matthews : .  Matthews : x.
 Gregory of Nazianzus’s Life of Julian revisited 

blood but clear-sighted enough to remove them from the schools, to chas-
tise Christians for their misuse of the power he had so recently bestowed
upon them. Yet, in his divine mercy God had also preordained that Julian’s
rule would be short-lived. That view of Julian became the dominant one
for centuries to come: as the apostate and Christ-hater destined to die
in Persia as a sign of God’s providence. To call Julian anything but “a
rabid dog” and an enemy of the Christians risked protests from all sides,
as Bodin, Montaigne, and Montesquieu speedily found out, even though
by the eighteenth century many could agree with Abbé de la Bléterie that
Gregory ought to have seen the good as well as the bad in the emperor and
his actions, just as Ammianus had done.
What, then, had Gregory done to shape Julian’s afterlife to such a decisive
degree? And conversely, now that Ammianus’s view of Julian has become –
mutatis mutandis – that of the scholarly mainstream, how has the scholarly
communis opinio dealt with Gregory and his characterization of Julian?
The question is of interest in the context of the manner in which the
“long fourth century” continues to be assessed because, as I stated at the
outset, both Emperor Julian (and hence Ammianus) as well as Gregory of
Nazianzus and his writings on Julian stand paradigmatic for many of the
methodological and even “ideological” ways in which this long century
continues to be discussed either by those concerned with the “secular
aspects of the end of the Roman world, such as its political, economic and
military history” or those engaged with “the new Late Antiquity” that has
“privileged religious history over social and many other kinds of history to
a rather excessive extent.”
The ways in which scholars approach Julian, the emperor, and Gregory of
Nazianzus, “the Theologian,” stand paradigmatic even today for precisely
these divisions between “secular” history and the “new late antiquity.” One
belongs qua emperor to real historians and the other qua father of the
church to those excessively concerned with religion. And yet Julian and
Gregory inhabited the same world and were deeply imprinted by the values
and concerns of their shared world. For example, both evoked the triad of
logoi, hiera, and the polis – the city and what it stood for – when seeking
 For an overview, see Rosen : –; also instructive are the marginalia on several manuscripts
commenting on Julian’s words. As Rosen points out ( n. ), an overview of Julian’s image in
medieval Byzantium remains a desideratum. For the Latin West, see Braun and Richer .
 Jer. Ep. . ; Vir. ill. Prol.; La Bléterie : –, –; of course, he maintained with Gregory
that Julian’s early death proved God’s providence (–).
 For the evolution of Julian’s assessment since Voltaire, see Braun and Richer .
 Harris : ; Ward Perkins : ; Liebeschuetz b; see also Giardina : –;
Liebeschuetz a. See also the incisive remarks of Lim (: –).
 susanna elm

to formulate prescriptions for the world they both inhabited. This was
a world in the process of changing its normative, conceptual framework
along with its reality – from polis to oikoumene, from a curially governed
city to one that functioned within the larger constraints of an empire, itself
governed by new types of men with a different economic base. To construct
these changes according to categories such as “real” history concerned with
secular matters versus “religious history” or Christianization does not suffice
to explain “the unexpected features of this new age” nor how these two men
marshaled “the more or less traditional way” of using the city as metaphor
and reality to achieve their respective goals: to instruct their audience
“how to govern and to safeguard the oikoumene of the Romans.” In the
following, I will briefly revisit Gregory’s so-called invectives against Julian,
his Life of the emperor as “the Apostate,” to show how he used the figure of
Julian as a teaching tool to instruct his audience of public men how to reject
all within the classical polis he considered “theirs” (pagan) in order to claim
the considerable rest as “ours” (Christian). In so doing I have tried to be
“responsive to the issues and personalities of [t]his time,” as John Matthews
has taught me to be, both in person and through his writings, to which I
return time and again, not only because I find it “hard to think of a topic
on which, however peripheral to his own preoccupations, he does not make
some contribution to [my] understanding,” but also from the comfort I
derive when “the elegant, slim volume I had originally planned” eludes me
and I am, once again, “encounter[ing] great difficulties of structure and
organization, above all in the deployment of detailed material” – and John
Matthews shows me the way. I could not have wished for a better teacher.

gregory’s julian: orations 4 and 5

The orations in which Gregory shaped Julian as “the Apostate,” his so-
called invectives against Julian, are his only orations known to the majority
of historians who deal with Julian the emperor since they are, after all,
one source for the reconstruction of Julian’s life. Nevertheless, despite their
value for Julian’s life, these orations have received relatively scarce attention
for their own sake since de La Bléterie, even though most historians are
milder in their verdict than Johannes Quasten, according to whom “hate
and anger so predominate in them that their historical value is almost nil.”
Indeed, many follow Glen Bowersock in granting Gregory the ability to
have gathered interesting gossip and useful tidbits, though Raymond Van

 Matthews : x–xi; Matthews : xi.

 Gregory of Nazianzus’s Life of Julian revisited 

Dam agrees rather wholeheartedly with Quasten’s assessment that these

“simultaneously hysterical and pedantic, fiercely caustic and pompously
ponderous” products exhibit more than anything their author’s lack of
historical sensibility. Those engaged in “religious history” or theology
likewise regard these orations with a certain diffidence, not least because
“cette espèce du danse allègre du scalp sur le corps d’un enemie morte de
la veille a quelque chose de gênant,” not easily reconciled with the picture
of Gregory as paragon of an “esprit de haute culture, brilliant et grâcieux,
âme douce et tendre, mal armée, faute peut-être de sens pratique, pour
soutenir les lutes dans lesquelles le hazard de la vie le jettera” – the Gregory
of historiography in a nutshell.
In fact, the metaphor of the danse du scalp is apt, because Gregory
performed with these orations on several levels. First, he explicitly and
emphatically did not wish to write history (.). Second, Gregory claimed
to perform a sacred ritual: his orations were a celebratory hymn (.;
.; .) and a bloodless sacrifice of words to the Word (.). As such,
his performance stood in sharp contrast to that of the emperor. Gregory
performed a sacred ritual, whereas Julian’s entire reign had been nothing
but theater, skene, a fiction without any roots in reality (. and ),
“making one laugh rather than cry” (.). Gregory’s aim was to make
Julian lose face, and so he presented him masked as a mime or histrio.
For him, Julian the man and his rule were theater, histrionics, and thus
as transient by definition as all theatrical performances. And through his
performance of Julian’s rule as merely transient theater, Gregory shaped
Julian’s image more than anything the emperor had ever done or written
in the short twenty months of his reign.
Julian had died the night of  June , fatally wounded on the battle-
field while on the retreat from Ctesiphon. Inspired by Platonic concepts
of the ideal ruler, Julian had understood himself as a philosopher, priest,
and emperor, called to rule and divinely inspired by the gods who had
been the lawgivers and hence “creators” of Greece’s greatness, in particular
Zeus-Helios, Athene, and Hermes. Therefore, immediately upon his acces-
sion, Julian began to publish imperial letters and to issue edicts aimed at
strengthening and invigorating the triad that had been the basis for Greek’s
greatness, namely logoi, hiera, and the polis, i.e., the Greek language and
the culture it created, the sacred, and the city. These he also considered
 Quasten : ; Bowersock : ; Van Dam : –.
 Bernardi : ; P. Godet, as cited in Lugaresi : –. For the historiographic Gregory, see
Elm  with further bibliography.
 Matthews : .
 susanna elm

the wellspring of Rome’s supremacy and hence guarantor of its imperium,

without which the oikoumene of the Romans could not be safeguarded.
Rome, of course, was for Julian the divinely appointed heir to Greece.
As he stated in his Hymn to Helius Rex (d–a), “the Greek colonies
civilized the greater part of the oikoumene thus preparing it to be easier
submitted to the Romans. For the Romans are not only Greek by origin [ou
genos monon Hellenikon], but also the sacred laws and the pious belief in the
Gods, which they have instituted and preserve, are from beginning to end
Greek.” Greek paideia, worship of the gods that had inspired it, and the
invigoration of the municipalities were therefore his “political program.”
On  June, members of the army command selected Jovian to succeed
Julian and to get them back behind Roman frontiers. Jovian, a Christian,
complied and concluded an unfavorable peace treaty. Half a year later, the
new emperor succumbed to accidental carbon monoxide poisoning. He
was succeeded by Valentinian, another Christian, who raised his brother
Valens to the purple as co-ruler responsible for the East on  March
. Gregory conceptualized Oration  during Julian’s reign but wrote
and revised it while Jovian was emperor and during the first months of
Valentinian and Valens’s rule, while Oration  was written about a year
and a half later, entering a lively debate among non-Christian as well as
Christian intellectuals concerning Julian’s apotheosis, the significance of his
reign, and that of his Persian campaign. Both orations address a particular
group of persons: Gregory’s peers, members of the new service aristocracy
that was then in the process of becoming an empire-wide elite. These men
and their prosperity fueled the economic growth and urban expansion of
the fourth-century East, and most of them – Gregory’s brother Caesarius
among them – were Christian. It was for their benefit that Gregory tried
to formulate in Orations  and  the correct ways in which Christian public
men ought to conduct the affairs of the politeia.
Gregory shared Julian’s assessment regarding the centrality of logoi, hiera,
and the polis: Greek culture, the sacred, and the city. He also saw them
as intrinsically linked with each other as well as foundational for the
well-being of the Roman empire. Gregory differed sharply, of course,
with regard to the divinity that had originated all three. Those whom

 For Julian and his policies, see Matthews : – and Rosen : – with further
bibliography.
 Matthews : –; Rosen : –.
 For a discussion of the dating, see Bernardi : – and Lugaresi : – with further
bibliography. I follow Lugaresi in assuming a later date for Or. .
 Banaji : –; Sarris , esp. .
 Gregory of Nazianzus’s Life of Julian revisited 

Julian had considered gods Gregory considered demons, and the figure
Julian considered the pseudo-divinity of a marginal people, the Galileans,
was Gregory’s universal pantokrator. A small but significant difference:
because Julian had, according to Gregory, misconstrued the sacred, his
understanding of logoi and the polis was equally faulty. In a private person
such a mistake might be merely reprehensible, but in an emperor such
“comportment . . . alone suffices to pillory [stēliteusai] the moral charac-
ter of a ruler,” because a ruler’s comportment affects the fundaments of
the state (.). And that is one of the central issues at stake in Gregory’s
allegedly ahistoric orations: the Roman empire (archē), the roots of its
power, and the manner in which the community of the Romans ought to
be governed – according to Gregory (.). To clarify his views regarding
these matters, Gregory focused on two central places in which the emperor
performed imperial rule and where his impact on the triad logoi, hiera, and
polis was most directly visible and most directly felt: the theater and the
marketplace.

theater
The theater – as is well known and was abundantly clear to Gregory – was of
crucial importance for the self-representation and self-constitution of any
city. Here the city’s elites were seated in hierarchical order and proportional
overabundance, viewing the results of their own financial munificence; here
they were seen by the non-elites as doing so (.). In addition, it was “the
primary vehicle for the inculcation of classical culture” because here mimes
and pantomimes performed the tragedies and comedies as well as all the
essential myths regarding the gods of the Greeks and Romans: the essence
of logoi and the hiera.
These very same logoi and the hiera at their center were what Julian
had sought to reinvigorate, albeit not their theatrical performance. That
interpretation was Gregory’s and an obvious defamation. Like many intel-
lectuals, Julian and Gregory shared the ambivalence the theater evoked.
It was powerful and attractive, it transmitted Greek culture, but it did so
in a popular, mocking, and often rather explicit fashion. To be a mime
and pantomime was considered shameful, infamus. Theater was by its

 Leyerle : –, quote .

 Compare Ammianus’s denunciation of the Roman senators’ lust for the theater in Matthews : 
and Lugaresi : . For an anthropologically based analysis of the nature of such ambivalence,
see Goody : –.
 Lim : –.
 susanna elm

very nature dissimulation, fiction, lie. To call Julian’s entire reign nothing
but “theater” (.) said it all. Julian had erected “a polis constructed of
words” that could not be sustained by deeds, because it lacked “the force
of a system that is derived from divine inspiration” (.; .; Pl. Rep.
c). All of Julian’s imperial acts were simulations of reality destined to be
fleeting – recall that the men whom Gregory here addressed, all Christian,
were also the ones who continued to fund just such theatrical games for
public enjoyment and at immense costs to themselves.
Gregory’s argument works on several levels. First, if Julian’s acts and
deeds, including his legislative acts and deeds, were “theater,” demonically
inspired dissimulations destined to be transitory, then Gregory’s words
were “reality” rooted in deeds and lasting because they expressed a divinely
inspired system: thus, his words overwrote Julian’s (legislative) acts and
deeds. Second, in a long disquisition about the nature of the myths por-
trayed in the theater so dear to Julian, Gregory dismantled the religious
and thus ethical foundations upon which Julian’s transient “polis” – that
is, his rule – rested: like their theater performances, the myths of Julian’s
gods were devoid of all the ethical prerequisites of appropriate rule, such
as restraint, justice, prudence, philanthropy, and, above all, piety; no won-
der, therefore, that Julian’s disastrous performance as ruler showed such
deficiencies (.–). For Gregory’s attack aimed not only at Julian’s
deeds and works but also at his person. At every stage Gregory sought
to portray the emperor as a bad actor who did not comport himself as a
real emperor should. When Gregory had first laid eyes on Julian in Athens
and observed him twittering and twitching, he could already divine what
the future would bring: “What more can I say? I looked at the man prior to
his deeds and recognized him by them . . . and I said to [my companions]:
‘what disaster is the Roman state nurturing here!’” (.).

agora
The theater was a place where the emperor interacted directly with his
subjects, where he performed imperial rule and was seen doing so. It was

 As Gregory elaborates in Or. .–, Julian’s claim to Greek culture and the sacred was nothing but
a recourse to the “fictions and vain words of the poets” (.), particularly those of “Homer . . . the
great comediograph, or better tragediograph of your gods” (.).
 Lim : –; French : –, –; also Sarris : .
 This is the famous and widely quoted characterization of Julian’s physique (Bowersock : ;
Rosen : ).
 Gregory of Nazianzus’s Life of Julian revisited 

also one of the places where imperial letters and decrees were received and
read aloud in a highly ritualized manner, as if the emperor were present
himself. A second site where the same interaction occurred was the agora,
the marketplace, and it is here, too, that Gregory engaged Julian.
Even though Orations  and  follow the fundamental structure of the
invective and are therefore usually characterized as such, Gregory with them
explicitly created a new genre, making it clear that he did not intend to
present his audience with an invective nor, for that matter, a history (.).
An invective, or psogos, was a set genre in which the author accused, judged,
and implicitly convicted a specific person, and, to reiterate, Oration  in
particular follows its basic structure. Nevertheless, Gregory’s distinction
is relevant. By forgoing the genre psogos, Gregory chose to forgo the role of
the prosecutor arguing his case to elicit conviction. Instead, he called his
orations a stēlographia – a “writing-on-a-stele” or pillar (.). The term is
a neologism and derives from the verb stēliteuein, to inscribe the name of
convicted criminals after their execution onto a pillar in the marketplaces.
Persons thus inscribed or pilloried were marked and shamed forever and for
all to see: to subject Julian to this fate was Gregory’s stated intention and
to achieve it he employed yet significantly altered a “traditional” genre.
Phrased differently, Julian had already received his sentence. By divine
decree he had been convicted and executed in the theater of war. God had
been prosecutor, judge, and executioner – Gregory merely proclaimed the
sentence and explicated it to all present and to those in the future. God had
permitted Julian’s rise to chastise the Christians, who had not used their
new power wisely, but he had also decreed his demise by ensuring that his
rule was as transient as a theater performance.
That Gregory called this sentence a “writing-on-a-stele” was no accident.
The act of stēliteuein was a public proclamation of legal acts, and such public
proclamations were the domain of those who ruled. On just such pillars in
the public places, Julian’s edict and imperial letters had been posted, includ-
ing those where he had presented himself as Platonic philosopher-king, the
Misopogon among them, and his famous decree excluding Christians from
logoi – all still in effect. In his metaphorical counter-stele Gregory, in a
 Matthews : –.
 Koster (: –, –) focusing on Latin authors is still fundamental, though he does not treat
Christian writers.
 Kurmann : –. For the inscribing of convicts see Ps.-Plut. V. decem Or. b; Philo, Quis
rer.div. haer. ; Jos. Ant. ..; Philochoros, Fr. Gr. Hist.  F ; Iambl. V. Pyth. .
 Gleason ; for a summary of the scholarly debates regarding Julian’s so-called school laws, see
Watts : –, and Matthews : –.
 susanna elm

performative act of his own, overwrote all of Julian’s acts and most of his
concepts, one by one. And while Julian’s words, written on stone, would
pass like dried grass, Gregory’s words – though written on flimsy linen or
wax tablets – would last: “Here our stele for you, higher and more visible
than the stelai [pillars] of Hercules . . . . This stele . . . will even be received
in the future, of that I am certain, to pillory you and your works and to
teach everyone not to attempt such a rebellion [apostasia] against God,
so that they may not be punished in like manner for having committed
similar crimes” (.).
The marketplace was also a site for the encounter of imperial deportment
and theater and theatricality, especially during the great festivals with their
processions. The Kalends of January, for example, saw processions through
the agora to the circus in which actors carried masks of gods and goddesses,
made fun of things and people, including the emperor, and enticed every-
one to dance the night away. Such behavior served Julian right. After
all, it was he who had, according to Gregory, contributed to such mis-
uses of the marketplace, making it possible for those “citizens capable of
composing such charming witticisms” (Misop. a) to abuse Julian in the
agora.
The agora had been, according to Gregory, a central place where Julian
had enacted his form of persecution by excluding Christians from it. He
had done so by mixing his gods with the symbols of his rule. “It is an
imperial rule,” says Gregory, “among the Romans that the emperors are
honored through official images (eikones) . . . . They require also proskynesis
to appear more venerated, and not only that directed to their person
but also that directed towards their statues and painted images” (.).
Julian had added poison to these images and statues because he “mixed
in with the honors traditionally rendered the emperors the impiety by
combining into one the laws of the Romans and the proskynesis in front
of the idols” (.). Whilst not at all criticizing the honor due to imperial
icons, Gregory chastised Julian for having excluded Christians from the
agora. Those who held public positions and exercised power were especially
hard hit because Julian’s actions forced them to choose between lèse-
majesté and obedience to their Christian commandments. Such behavior
revealed Julian’s perversion of Roman law and hence called into question
his own legitimacy as an emperor. A ruler who is unjust cannot issue
just laws, as Gregory is at pains to point out throughout Orations  and
, so that all the laws that Julian had issued and had displayed on the

 J.Chrys. in Kalendas  and PG ., ; Jul. Misop. c, a; Gleason : –.
 Gregory of Nazianzus’s Life of Julian revisited 

stele in the marketplace were invalid and ultimately overwritten by divine

decree.

empire or governance
What was at stake for Gregory was the future of the relationship between
Christian God, emperor, and priest. By calling his discourse a “writing-on-
a-stele,” Gregory signaled that he was posting divinely sanctioned codes
of public conduct that overwrote Julian’s imperial decrees. And by attack-
ing Julian as lacking imperial gravitas, he defined what being a Christian
emperor entailed. Gregory juxtaposed Julian’s negative image with a pos-
itive portrayal of Constantius as a paragon of an ideal Christian ruler.
Constantius “knew very well, because he reflected upon such problems
in a manner more elevated and imperial compared to many others, that
the Roman power had grown together with that of the Christians and
that the imperium had arrived together with the coming of Christ” (.).
To assure the continuing greatness of the Roman empire thus required
the appropriate nurturing of all things Christian: the unity of Christian
teaching or logoi; the protection and patronage of Christian hiera and, of
course, through tax exemptions, removal of “pagan” symbols; the appropri-
ate conduct of festivals (.–), the creation of the Christian polis, and so
on. Julian, this “best of all governors of the commonwealth [koinon],” had
failed to realize that to eradicate Christianity “now that it ruled sovereign”
was to threaten Roman rule and to place the entire commonwealth into
grave danger, “all because of this newfangled and wonderful ‘philosophy
and rule,’ which has given us happiness [eudaimonia] and brought us back
to the golden age and to that politeia that does not know rebellions and
wars” (.).
Gregory, to recall, wrote these two orations during the reign of Julian’s
successors who were all Christian. But much remained uncertain: how
would these new emperors comport themselves as Christian emperors?
Julian’s edicts needed to be rescinded, but would they do that given that he
had been a legitimate and duly divinized emperor? How would they treat
Christian logoi? After all, Julian had been the first baptized Roman emperor,
but he had deserted. Instead of adhering to what his priests and bishops
had taught him, he had gone off to invent his own logoi – a stern warning
to Christian emperors not to deviate from the words of their bishops such
as Gregory. And what about the theater and the marketplace? How was a
Christian emperor to comport himself there? For example, as Neil McLynn
has so persuasively pointed out, the manner in which a Christian emperor
 susanna elm

should perform imperially in the new theater – the church – was at that
point entirely unclear: how was one to integrate the emperor into the
liturgy?
By creating Julian as “the Apostate,” Gregory used the traditional model
of the invective in order to create something “positive” ex negativo: Julian
became his teaching tool to explicate and to demonstrate to his audience –
his peers – which central values of the dominant culture to claim and which
ones to discard and how they ought not to conduct (and thus, inversely, how
to conduct) themselves in claiming and controlling these values. In so doing
Gregory used the city and the oikoumene as metaphors but also to give
clear directives – from the manner in which imperial iconography ought
to be displayed to ways in which public men ought to govern as Christians
within the space of the polis and the empire and how they should view
the instructions of their priests. At stake was the appropriate governance
of the empire now that it was Christian, and that meant at the same time
the correct interpretation of what the Word, Logos, had wished. Julian (as
constructed by Gregory) had been a divinely sent warning sign on how not
to rule – and Gregory had been chosen to explicate what that implied to his
audience of men engaged in fostering and safeguarding Rome’s greatness.
To say it differently, Gregory’s invectives became so powerful – I suggest –
because they were far more than either a rote invective or a hysterical
screech. By refashioning the genre into the stelographia, Gregory not only
denounced and shamed the deceased emperor as deserter, aka Apostate,
but used him as divinely mandated pedagogical device sent to help the new
men understand how God wished them to use their power as Christians in
the agora, the polis, and the oikoumene – as instructed by their priests. And
if one lived in a world where the correct understanding of the divine was
the precondition for the safety of everyone and everything, such matters
were as “real” as the economy and the military.

 McLynn : –.

 part iii

Faces of Theodosius I
 chapter 10

Liar in winter: Themistius and Theodosius

Peter Heather

While Jerome was busy inventing the Latin Christian Chronicle tradition,
two huge political stories were breaking around him in Constantinople in
. The first was less out of the ordinary. A new emperor was in town,
letting loose all that scrambling for office, honour, and alliance, which, as
John Matthews’s work has done most to illuminate, marked out regime
building, late Roman style. This particular scramble was a bit unusual since
Theodosius had been in office since January , so this was hardly the first
few hectic months of power broking. But the emperor had only entered
his capital in November/December , so his regime was new there, the
beating political heart of the eastern Mediterranean. It was also a regime
which needed desperately to reinvent itself: because of the second of the
stories. The Goths who had killed and defeated Theodosius’s predecessor
Valens at Hadrianople on  August  were still at large in the Balkans and
entirely unsubdued. Theodosius came from a highly distinguished military
family, and could boast his own decent military track record. He had been
appointed to beat the Goths and given command of all the affected areas
of the Balkans, contrary to normal late Roman political geography, to
facilitate unified operations. But, unthinkably, he too had failed to beat
the Goths. His army fell apart in the summer of , command of the war
reverted to the western Emperor Gratian’s generals, and Theodosius beat a
hasty retreat to Constantinople, tail firmly between his legs. In the winter
following defeat, Theodosius’s regime was in crisis, its credibility stretched
thin.
To help navigate a course to calmer waters, the emperor turned to a
trusted pair of hands, the veteran by now of thirty years of high-profile

 The scholarly consensus is that Theodosius was given East Illyricum, esp. Moesia and Macedonia
(but not Pannonia and Dalmatia): Grumel , based on Sozomen, Ecclesiastical History .–. For
the suggestion on the basis of Sidonius, Poems . that it was the whole diocese, see Errington
b: –. But Gratian’s forces continued to operate in Pannonia, which to my mind makes
Sozomen’s account preferable. On Theodosius’s defeat, see Heather and Moncur : ch. .


 peter heather

politics, several changes of regime, and a number of imperial defeats: not

least the fallout from the death of Julian on Persian soil. On the second
anniversary of Theodosius’s accession to office, in all probability, on 
January , the philosopher and orator Themistius stood to address the
great and good of the eastern empire in what is now Oration  in the
standard editions. This extraordinary speech points up some of the key
ways in which the regime was responding to crisis. What neither he nor his
audience had the slightest conception of, however, was the earth-shattering
ramifications of the lines of policy that he started to air. It would not only
help bring down an empire, but have a measurable, if certainly indirect,
impact on the practical working out of Jerome’s literary projects.

emperor and philosopher

There is a backstory to Themistius’s appearance in front of Theodosius
that we cannot tell in full. It was not his first oratorical effort for the new
emperor. That had come in the course of an official embassy from the Senate
of Constantinople conducted in late spring or early summer the previous
year, before the military debacle. Themistius had followed on behind the
rest of his colleagues, pleading illness. He may really have been ill, but
the delay between the emperor’s accession (January ) and Themistius’s
first speech in front of him (April/May/June? ) fits a broader pattern
in the philosopher’s career. As regimes changed, there always was such a
delay while incoming emperor and established philosopher negotiated the
terms of their relationship. What was said by Themistius to whichever
it was of Theodosius’s confidants who first approached him, we have no
way of knowing, but the outcome was beneficial to both. What emperor
could offer philosopher was continued prominence in public life, and
Themistius grabbed it enthusiastically. He had never, in fact, been slow
to indicate to potential rivals the closeness of his relationships with the
sequence of his imperial patrons, and the early speeches in front of the
Emperor Theodosius did not fail to emphasize that, once again, he had
managed successfully the leap from one regime to the next.
The script of Oration  emphasized that Themistius was speaking at
Theodosius’s personal command (c), reminded the audience that in the
 It was given in winter (Or. .b–c), in the third year of Theodosius’s reign: hence on or after 
January . The speech also discusses the arrival in Constantinople but not the subsequent death
of the Goth Athanaric (c–b). He arrived in the city on  January , and died there on 
January (Cons. Const. s. a.  = C.M. , ), making it likely that Themistius was speaking on 
January itself, Theodosius’s third anniversary. This is the generally accepted dating.
 Illness: Themistius Or. .c. Pattern of delay: Heather and Moncur : ch. .
 Liar in winter: Themistius and Theodosius 

course of his career he had kept company with a multiplicity of emperors

(a), and, in a lengthy passage, equated himself with the poet Tyrtaeus,
whom the Athenians had sent to the Spartans to put fight back into them
(c ff.). Oration  of January , likewise, left the listener in no doubt
as to Themistius’s continued importance. It emphasized that both the new
Consul for that year, Saturninus, and the Emperor Theodosius himself had
omitted nothing that would advance his honour (b), the same passage
also happening to refer – again – to the amount of time Themistius had
spent as an insider to a succession of imperial regimes. Striking too is
the end of the speech, where Themistius returned to Theodosius’s son
Arcadius, portraying himself as the boy’s guardian and educator (a–b).
That he chose to make the point so strongly suggests that there had been a
real chance that Theodosius wouldn’t be open to an accommodation, and
that some vultures had been hovering around hoping to feed on the corpse
of Themistius’s political prominence. But if it is obvious what Themistius
gained from the friendship, the benefits to the emperor are worth a bit
more comment.
In part, Themistius offered Theodosius precisely what he had been
delivering for the regimes of a succession of Christian emperors for the
last thirty years: cultural reassurance. Themistius was a non-Christian, a
philosopher with a specialism in Aristotle but huge knowledge of Plato
too, whose Hellenic cultural credentials were in one sense impeccable. He
exploited the long-established tradition of parrhesia – the philosopher’s
right to free speech undaunted by men of power – to claim to speak
nothing but the truth, even in front of an emperor. If Themistius gave
his stamp of approval to the regime of a Christian emperor, it offered
reassurance to the still many non-Christians in the governing classes of
the empire that they had nothing to fear, whatever religious novelties the
regime might be introducing. To do this, Themistius had set himself in
deliberate opposition to other Hellenic philosophers, particularly some of
the Neo-Platonic holy men, who saw their wisdom as entirely incompatible
with Christianity, and whose hostility to it was unrelenting. Themistius had
been happy to fight them by staking his claim to a middle ground of Greek
cultural heritage which could be revered by Christians and pagans alike,

 Themistius’s final imperial patron would in fact prove – in formal terms – the most generous,
granting Themistius the Urban Prefecture which guaranteed his position among the leading group
of Constantinopolitan senators, perhaps the kick upstairs that marked his retirement from active
politics (Heather and Moncur : ch. ).
 peter heather

and whose watchword was not ascetic withdrawal, but practical wisdom to
be employed in the interests of the state.
By the time of Theodosius, however, these were old contests, and
Themistius’s stance had long since lost its novelty. But that did not dilute
its usefulness. If anything, the stamp of Hellenic approval was even more
necessary to the new emperor than his predecessors. As Neil McLynn’s paper
explores, in these early years Theodosius’s regime was busily attempting to
establish its Christian credentials, with Theodosius the first emperor to
renounce the old pagan imperial title of Pontifex Maximus on his acces-
sion in January . This was followed in February  by a clear statement
of how Christian orthodoxy was to be defined, and, as Themistius was
speaking, a major council of eastern bishops in Constantinople had been
or was about to be called for May . Making these Christian moves was
an act of huge symbolic importance for potential Christian supporters,
but ran the risk of alienating traditionally minded non-Christians among
the elites of the east, of whom there were still many. A simultaneous and
equally public accommodation with Themistius, the symbol of traditional
culture, thus allowed Theodosius to establish a balanced ideological profile
at the start of his reign. The fact that the emperor should also have taken
the trouble to court Themistius adds further weight to McLynn’s argument
in this volume that we should be highly suspicious of the traditional picture
of an unrelentingly Catholic Theodosius.
But Themistius had far more to offer than symbolic cultural reassurance.
He was a consummate orator as well as a philosopher. He himself claimed
it to be a distinctive feature of his own philosophy, one inherited from his
father, that, unlike some, he did not shy away from using skilled oratory.
This emphasis had spawned a noisy controversy around him in the s,
when he first broke into the public life of the capital. His opponents claimed
he was not a true philosopher, but a sophist: an individual not interested in
true wisdom, who used words to get ahead. One who lied, in other words,
for money. Themistius denied the charge vigorously, claiming that as long
as the message was sound, there was no harm in using every rhetorical
trick in the book to get it across. Whatever view you take on this, there
is no denying Themistius’s rhetorical prowess. It was certainly appreciated

 For this view of Themistius in more detail, see Heather and Moncur : ch. . Themistius’s
self-presentation stands in strikingly self-conscious contradistinction to the Neo-Platonic holy men
discussed in Fowden . On parrhesia, see esp. Brown : chs. –.
 On the pontifical robe issue, see Alan Cameron . Alan Cameron  has now thrown this into
doubt, but I’m happy to stick with the traditional view. Definition of Faith and Council: CTh ..
(cunctos populos) with McLynn in this volume, ch. .
 Liar in winter: Themistius and Theodosius 

by the younger Libanius, a letter from the s recording how he and a
friend spent the entire morning poring over the copy of the latest speech
that Themistius had sent them, but also by a succession of emperors, who
trundled him out when there was some difficult policy twist that needed
justification.
There had been some problems along the way that were beyond even
Themistius’s powers to spin. Trying to claim that Jovian’s peace treaty with
the Persians – which cost the empire Nisibis and Singara, five provinces
beyond the Tigris, and fifteen forts – had actually been a victory was one
argument that was never going to fly. But the decision to make that claim
was clearly one taken by the regime, not by Themistius himself – it also
appears on Jovian’s coinage – so he just had to make the best of it, which he
duly did by giving it maybe three minutes in a fifty-minute speech, which
otherwise concentrated on more plausible Jovianic successes. Otherwise,
Themistius’s speeches were much more effective, whether in selling less
than outright victory as a good outcome to Valens’s first Gothic war, or
in helping to insulate Christian emperors keen on a policy of religious
toleration from the pressures of Christian lobby groups. He claimed to be
an independent philosopher, using the right of parrhesia to tell the truth
without fear or favour. But such is the closeness of the fit between the
needs of his different imperial masters and the lines of policy advocated in
Themistius’s different speeches that it is impossible to avoid the conclusion
that he was in fact a spin doctor, selling regimes and their policies to
the Senate of Constantinople, of which he had been a member since his
adlection by the Emperor Constantius II in .
It was in this role again that Themistius stood up in January , and the
results do not disappoint. In his opening lines, Themistius noted that the
army was in training, in winter quarters, but held out the expectation that
it would take the field in due course (b–c). Towards the end, the Goths
were labelled ‘the Hounds of Hell’, and Themistius looked forward to the
Emperors Gratian and Theodosius inflicting such a defeat on the Goths that
it would be remembered, after Homer, in the ‘far hereafter’ (b–a).
Sandwiched between these briefly bellicose remarks, however, was a long
disquisition on the general nature of the imperial office, and on Theodosius
in particular. Rather than celebrating the emperor’s military capacities, as

 Controversy: Heather and Moncur : ch. . For the reading, Libanius, Ep. , trans. Norman,
, as Ep. ; cf. the more general comments on Themistius’s rhetorical skills in Gregory of
Nazianzus, Epp. , ; Libanius, Ep. .
 Persian defeat: Themistius Or. .a–c with Heather and Moncur : ch. . For full argumentation
justifying this view of Themistius, see Heather and Moncur : ch. .
 peter heather

he had done the previous spring, Oration  concentrated instead on

Theodosius’s possession of the moral qualities required to bring good
civilian government to the eastern empire. In the speech, Themistius was
quite explicit about this change of emphasis, noting right at the beginning
that he would be combining the peaceful subject matter of Hesiod with
Homer’s grander rhetorical tone (b–a). In its central portions, he
then went on to note that Theodosius’s virtues had been displayed not
only in his direct conduct of the imperial office, such as tempering justice
with mercy and showing proper generosity to his subjects, but also in his
ability to appoint officials who shared the same moral qualities as himself.
Overall, Themistius argued, while it was all well and good for emperors
to be capable commanders, their chief concern had to be good civilian
government. Actually commanding troops in battle was a desirable extra,
not an essential qualification for the job (b–d). This was a strikingly
different portrayal of Theodosius to that of Oration , which had hailed
him as the man who would win the Gothic war.
The explanation for this striking change of tone lies in the course of the
Gothic war itself. The collapse of his army in summer  had forced Theo-
dosius to hand back military control of the war to Gratian. Themistius’s
demilitarized account of the imperial office in January  was an auda-
cious attempt to find an alternative ideological justification for a regime
whose initial self-presentation had been vitiated by military disaster. The
extent of the problem should not be underestimated. Victory was the
prime hallmark of a fully legitimate Roman emperor, because it showed
that God’s Hand was behind him. What did Divine support mean if not
victory on the battlefield? Having failed on the military front, Theodos-
ius’s regime needed to establish its ideological legitimacy by another route,
and quickly. Themistius’s speech answered that need. Portraying Theodos-
ius as the upholder of divinely ordained civilized order exploited a second
aspect of Roman imperial discourse. Emperors as triumphant conquerors
of barbarians provided one of its central images, but no less important was
the civilized, rational order at home – civilitas – which such conquests were
designed to protect. With the Gothic war still raging, Themistius could
not afford entirely to abandon the image of Theodosius the war leader, but
his rhetorical strategy concentrated on the other main set of Greco-Roman
ideological criteria, by which Theodosius might be accounted a ‘good’
emperor. Essentially, Themistius took the opportunity represented by a
 The point emerges powerfully from McCormick .
 The moral education of individuals and the ordering of society according to written law were the
central planks of this order (civilitas): see, for instance, Barnish ; Heather a, b.
 Liar in winter: Themistius and Theodosius 

keynote speech on the second anniversary of the emperor’s accession to

relaunch the regime.
As a piece of oratory, it works brilliantly, but there are two things we
don’t know. First, how much input had Themistius had into the decision to
reinvent Theodosius’s image in this way, shifting the picture from ‘imperial
commander to win the Gothic war’ to that of ‘supreme upholder of
Divinely ordained civilitas’? I suspect this was a decision taken by the
regime as a whole, not Themistius’s initiative, but he may well have had
considerable input into how to make the new image stick. Second, how
well did it work? The speech is coherent in argument, and clearly touches
all the right ideological bases as far as the new image was concerned, but
did it actually convince any doubters in the audience that Theodosius was
a ‘good’ emperor, despite his failure to defeat the Goths? Again, there is no
way of knowing, but the speech was clearly only one salvo in the battle to
win over hearts and minds within the ruling elites of the eastern empire.
Themistius’s initial speeches for Theodosius again point us towards some
of the other measures being taken at the same time, and there is every
reason to suppose that the philosopher himself was deeply implicated in
the process of regime building.

constantinopolitan supporters of theodosius

A rightly famous John Matthews paper explores a cluster of allies and
supporters that Theodosius brought with him from the west, and who
proceeded to hold high office in the east. The extraordinarily cosmopolitan
nature of top-level careers, military but above all civilian, in the fourth
century is, likewise, one of the key themes of Western Aristocracies. Such
men had or came to possess a widely dispersed portfolio of landed assets and
might find themselves serving a single emperor or a succession of emperors
just about anywhere from Trier to Antioch. You can find many examples
without the slightest difficulty from the Tetrarchs to Stilicho and beyond.
There is thus every reason to suppose Theodosius entirely typical in having
been trailed to his new seat of power by a group of old supporters and their
hangers-on in January . But Theodosius was equally typical of late
Roman emperors in another way too. The irregularity of top-level politics
in the fourth-century empire, which might be described with maximum
generosity as ‘semi-dynastic’, meant that emperors regularly came to the
purple with no established powerbase in the regions they were to rule.

 Matthews ; .

 peter heather

Theodosius was a westerner, whose previous military career was western,

and who, as far as we know, had never even visited Constantinople and the
east before Gratian elevated him to imperial command there.
Governing the late Roman empire, or any major part of it, was not a
straightforward task at the best of times. Theoretically an emperor enjoyed
untrammelled powers, and, when motivated to do so, could mobilize a
terrifying level of legal and military force against individuals or groups who
incurred his imperial wrath. The codes are full of intimidating legislation,
and the pages of Ammianus record many a nasty investigation – not least
the famous magic trials in Rome or Antioch – complete with torture and
grisly ends for many of those caught up in them. But this is only one half
of the picture. Equally important was the staggering size of the empire.
Even just the eastern half of it ran from the Succi Pass to the Euphrates
and from the Danube bend to Cyrenaica. Across these vast landscapes
most things moved at no more than forty kilometres a day where the
modern average would be easily ten times that. In terms of how long it
took administrators, information, and decisions to circulate, therefore, you
have to figure the empire ten times as large even as it appears on a modern
map. Fourth-century eastern emperors, moreover, tended to spend their
time moving up and down a line of force stretching between two key
centres: Constantinople close to the Danube and the west, and Antioch
close to the Persian front. Even when they had been in office for a while,
therefore, there were vast stretches of their empire and even economic and
political centres of huge importance such as Alexandria, of which such
rulers had no personal knowledge whatsoever. This, of course, was even
more true of a neophyte like Theodosius, parachuted into an entirely new
domain.
All this meant that the Empire could not but be run for the most part
by an administrative bureaucracy, whose numbers were growing geomet-
rically as the fourth century progressed. In the old grand narratives of
imperial decline, the nature of this bureaucracy was summed up by a few
of Ammianus’s horror stories combined with a couple of pages of rant
 Vanderspoel : – unconvincingly argues that Theodosius started to create an eastern power-
base in late , visiting Constantinople in November. This is based on an obviously misdated
reference in the Chronicon Paschale to Theodosius’s first entry to the city (put in November 
instead of November ) in a passage which misdates a whole series of other major events such
as the death of Valentinian I (trans. Whitby and Whitby  with nn.), and a forced reading of
Themistius, Or. .a–b.
 Again, the work of John Matthews has done much over the years to bring out the sheer nastiness
of maiestas proceedings: Matthews : –; : chs. –.
 No fourth-century emperor ever visited Alexandria, it seems, although Constantine did once plan
to: Barnes : –.
 Liar in winter: Themistius and Theodosius 

from Libanius. Bureaucrats were new men, over-promoted nobodies who

were centralizing creatures of the empire, unafraid to tax the old local
curial elites to extinction to fund the imperial armies at a time of over-
powering economic recession. As the twentieth century wore on, much
of this picture was revised. The ‘new men’ were real enough, but not char-
acteristic. Most imperial bureaucrats were recruited exactly from the old
curial elites of the empire they were supposed to have been so willing to
persecute, and, as a result, complex social ties bound bureaucrats and curi-
als together. The supposed economic decline of the empire has itself also
collapsed under a staggering weight of archaeological evidence, much of
it ceramic, showing that most of the fourth-century countryside was pop-
ulated as never before in the entire Roman imperial era. Even the extent
of centralization needs careful qualification. The late empire did intrude
itself into the formerly self-governing cities to an unprecedented extent in
matters of taxation. Old revenues were confiscated, and regular attempts
made to catalogue revenue-generating agricultural assets in great detail.
But this much central government taxed the capacity of the bureaucracy to
its limits, both in terms of numbers (though much bigger, numbers were
small in global, comparative terms) and in its technological capacity to
capture and process information. Put simply, it just wasn’t possible for an
empire of this size (even half a one) to be governed in a generally centralized
fashion, and, in most ways, its constituent local units, particularly outside
of the Constantinopolitan–Antiochene axis, were in practical terms largely
self-governing.
This brings the task of regime building faced by Theodosius in Jan-
uary  firmly into view. Aside from extending his control over the main
concentrations of troops (or in Theodosius’s case recreating them after
the disaster of Hadrianople), he had also to form a working relationship
with key members of the departments of central government, and gener-
ate consent, somehow, among key figures in the revenue-producing local
communities, which made up his empire. He could in theory just com-
mand obedience, but everything would function infinitely more effectively
if positive support could be garnered. Essentially, Theodosius needed to
make central and local government want to work with him, and needed to
deploy the extensive powers of patronage he had available to generate this
effect.
 The relevant chapters of the last volume of the old Cambridge Ancient History and the first of the
old Cambridge Medieval History document the old orthodoxy.
 I have not documented these points, because they seem to me generally accepted and entirely
uncontroversial. They may of course eventually pass the way of other orthodoxies.
 peter heather

The process began immediately in Thessalonica, where Theodosius

established his first capital to prosecute the Gothic war raging around
him. A hostile account of the political activity this situation required can
be found in Zosimus’s picture (derived from Eunapius) of the early days
of the reign. Great streams of people rushed to Thessalonica from all over
the east, we are told, and all received favourable replies to their requests.
According to a second passage, Theodosius’s huge generosity caused such
an impoverishment of the imperial fisc that it became necessary to sell
official positions to the highest bidders. Though certainly hostile, these
passages capture some key aspects of regime building. Important men in
need of favours were themselves courted by Theodosius with gifts of money,
title, and privilege. At the same time, those wanting to carve out niches
for themselves in the imperial administrative hierarchy came to court the
emperor. This kind of political auction was unavoidable at the start of a
reign in any case, just as it was for medieval kings too, who also had to give
away far more at the beginning of a reign than subsequently. Because of the
circumstances of Valens’s death and Theodosius’s accession, however, the
new emperor could draw on no elements of continuity from the past, and
this quite possibly did mean that Theodosius had to be more generous than
an emperor with more opportunity to build up ‘connections-in-waiting’.
Certainly, that other great fourth-century western imperial parachutist,
Constantine, behaved in the east with similar generosity after his victory
over Licinius, so there may well be a pattern.
Other sources provide less overtly hostile visions of this activity. Neil
McLynn has suggested elsewhere that a series of legal measures passed by
Theodosius at this time, dealing with allocations of currently vacant prop-
erty, were responses to petitions presented to the emperor at this fertile
moment. There were probably plenty of them: Libanius mentioning two
of his acquaintances who took the opportunity to press their suits. Fur-
ther insight into the distribution of patronage going on in Thessalonica is
provided by the speech Themistius gave there: Oration . It closed with
the plea that Theodosius should rival Constantine in his attentions to his
new capital city, Constantinople, and in particular that he should make
it an equal second Rome, not only in terms of its buildings, but also in
its honours, by which Themistius meant quite specifically the total size

 Respectively, Zosimus ..; ..–.

 Constantine: Heather . Ammianus similarly comments that the start of a reign was the perfect
opportunity to press dubious claims before the regime had any real knowledge from which to judge
their validity (..).
 McLynn a:  on CTh ..–; Libanius: Or. ., .
 Liar in winter: Themistius and Theodosius 

of its senatorial body (a–a). As with most Themistian pleas, which

have sometimes been misinterpreted as expressions of genuinely indepen-
dent opinion, the rest of the speech tips us off that Theodosius had already
started implementing the recommended policy before Themistius spoke.
The likeliest explanation here is that Theodosius was using Themistius to
make the pronouncement that he was willing to increase eastern senatorial
numbers until they were equal to those of the Senate of Rome itself. Equal-
ity of privilege for the Senate of Constantinople had been granted under
Constantius II in the s, alongside an increase in numbers. Now Theo-
dosius was bringing a further increase to the party. And, as Brian Croke’s
chapter explores below, equality in numbers of senators was matched by a
determined building campaign which turned Constantinople into a mag-
nificent imperial capital in its own right.
How many new appointments Theodosius made is unclear. Themistius
reports in a speech of / that during his political lifetime the size of the
Senate increased from  to ,. But this includes the earlier expan-
sion under Constantius, and Valens too must have increased senatorial
numbers through the famous laws on precedence enacted with his brother
Valentinian I in . These made senatorial status the culmination of all the
different strands of honour and office within the imperial system. Nonethe-
less, Theodosius was in need of a lot of friends in , and I suspect that
we may be talking of at least a few hundred new appointments.
Thanks to those precedence laws, senator of Constantinople was a title
calculated to win friends and influence people at both ends of the crucial
administrative spectrum, in both central and local government. In central
government, senatorial status was, as the fourth century progressed, increas-
ingly the top status that an imperial bureaucrat could aspire to. Previously,
it had been top equestrian rank, but honours inflation was putting paid to
that. Making top imperial bureaucrats Constantinopolitan senators, there-
fore, was exactly what the doctor ordered. The title also perfectly fitted
the bill for stay-at-home landowning elites who dominated local society.
Here, fourth-century developments made the dominant group honorati:
men with imperial honours and retired imperial bureaucrats who sat with
the provincial governor when he tried cases and carried out periodic revi-
sions of the local tax allocations, amongst other key duties. Becoming a
senator of Constantinople instantly propelled you into this key class in
 See in particular Or. .c. See generally Heather and Moncur : ch.  on the underlying
patterns within Themistian pleas.
 Senatorial expansion: Or. .xiii; Valentinianic laws on precedence: Jones a: – commenting
on CTh ..; ..; ..; ..; ...
 peter heather

local society, even if you had no intention of turning up regularly in the

imperial capital.
Indeed, the two groups were closely intertwined. Between ca.  and
 CE, the number of attractive jobs in the imperial bureaucracy increased
from about two hundred and fifty overall to at least three thousand per
generation in each half of the empire: a twenty-fold increase. But the
new bureaucrats and senators who benefited from this process did not lose
contact with the home provinces from which they had come. After the riot
of the statues in , and again when advancing the senatorial claims of his
friend Thalassius, Libanius mobilized support among Constantinopolitan
senators of Antiochene origin. The letters of the Cappadocian fathers,
likewise, are full of requests for help to great imperial functionaries, several
of whom had Cappadocian origins. Here, common origin provided the
starting point of the relationship which the requests exploited. Imperial
officials active in provincial government might also have links to local
society. It seems to have been not uncommon, for instance, for men to
govern their home provinces, a career pattern which transformed but did
not break off existing linkages in local landowning societies. Such inter-
connections – based on travel, letter writing, marriage, peripatetic study,
and above all common interest – provided the wiring through which the
‘conductivity’ (as Peter Brown once labelled it) of the landowning classes
of the eastern Mediterranean flowed: their ability to act over vast distances
as a surprisingly homogeneous group of opinion.
The senatorial body which grew up in Constantinople thus represented
a fair cross-section of the upper echelons of the landowning, tax-paying,
tax-gathering, and locally dominant opinion by whom and for whom the
Empire was run, a cross-section which maintained strong ties to the local

 Honors inflation, the bureaucracy, and honorati: Jones a: chs. –. For a more political spin
on these developments, Heather .
 See further, e.g., Dagron ; Chastagnol ; Heather .
 : Libanius, Or. . (cf. .). For Thalassius, Libanius mobilized twelve senators of Antiochene
origin (Petit : –).
 Basil of Caesarea’s main court contacts were both natives of Caesarea: Sophronius (Magister Officio-
rum and Urban Prefect of Constantinople: Epp. , , , , , ) and Aburgius (Praetorian
Prefect of the Orient: Epp. , , , , ). After his time as bishop in Constantinople, Gregory
of Nazianzus had a wider range of contacts: Epp. – (to different friends after his return home);
cf. –, –, –, –.
 Inscriptions from Aphrodisias record the local roots of numerous late antique governors of Caria:
Roueché : nos. , , , –, –,  (the latter honouring Aphrodisias as his father’s
homeland).
 The interconnections of eastern elites still await the kind of treatment provided for the west by
Matthews . For some introductory thoughts, see Heather ; or, from a different angle,
Brown .
 Liar in winter: Themistius and Theodosius 

political communities from which they had, at this point, only recently
emerged. A major senatorial expansion was well calculated, therefore, to
generate political capital of exactly the kind Theodosius required, with
heavy emphasis at the political centre in Constantinople, but a strong
trickle-down effect into the elite groups who governed the localities.
The one potential risk was that too many promotions might annoy those
who were already senators by diluting their sense of privilege. Even by ,
this was a body of men to take seriously. For all his patron’s well-known
complaints about his cousin Constantius II, it is striking that Julian’s
panegyrist, Cl. Mamertinus, went out of his way to pay compliments to
the new imperial senate, and to note Julian’s marks of favour towards
it. Slighting or disbanding the senate which Constantius had done so
much to create was no part of Julian’s plan for distancing his regime from
that of his predecessor, presumably because this would have annoyed too
many people whose support he needed. There is also the famous case of
Thalassius – rejected, it seems, twice for membership – reported in loving
detail in the correspondence of Libanius from  CE, which shows that the
Senate was far from ready to share its privileges with complete liberality. In
expanding it, therefore, Theodosius had to tread a careful path: providing
sufficient levels of reward, while not alienating established opinion.
On balance, the emperor’s adventus into Constantinople would probably
have generated a further bout of regime building whatever its circumstances
and date. Military defeat at the hands of the Goths, however, made further
recruitment of favourable opinion all the more necessary, since the failure
of the emperor to fulfil his original remit of victory at least potentially
opened the door to possible rivals. Just as Procopius’s revolt against Valens
had taken some time to gain momentum and organization among dis-
contented elements in the east’s military and civilian hierarchies, so might
Theodosius’s military failure have provided the crucial catalyst even after
two years of power. Themistius’s attempt to relaunch the regime on a civil-
ian ticket was one important strategy deployed at this crucial moment, but
there were others too, and, again, something of these activities is reflected
in the early speeches Themistius gave for Theodosius in Constantinople:
both Oration  of January  and Oration  of January .
A striking subtheme of both speeches is a determined and thorough-
going critique of the government of Theodosius’s predecessor, the Emperor
Valens. Again, the unstinting post-mortem critique of his former imperial

 Mamertinus: Pan. Lat. . ; on the Thalassius story, see Liebeschuetz : – commenting on
esp. Libanius, Or. .
 peter heather

employers was a regular feature, one of the stratagems by which Themistius

negotiated the potentially difficult hop from one regime to the next. Crit-
icism of Valens had begun, if briefly, back in Oration  in Thessalonica.
Despite the closeness of his own past association with him, Themistius was
quick to damn Valens in hindsight as a man lacking in mildness and ‘love of
mankind’ (philanthropia). Back in Constantinople, the critique became
relentless. In Oration  above all (but also in ), Themistius placed huge
stress on the claim that the imperial office should be the culmination of
a properly progressive career in public life. This was aimed entirely at the
memory of Valens, whose relative inexperience Themistius had found ways
to praise when the emperor was in power, but which was now deployed
to point up Theodosius’s virtues as a much more suitable occupant of the
imperial throne. As Themistius developed the argument in Oration , it
was precisely Valens’s inexperience which led him into a whole series of
mistakes, all of which Theodosius avoided. Valens had suppressed proper
freedom of speech (parrhesia), where Theodosius cherished it (a–b).
Valens had generated fear among his subjects by sentences of exile, con-
fiscation, and death, where Theodosius had not issued a decree of death
in three years (b–c), and had even returned confiscated land to the
heirs of those condemned under Valens (d, d). Valens had promoted
unworthy subordinates, but Theodosius knew how to pick out virtuous
men (c–d).
Cherishing freedom of speech, not resorting to fear to control one’s
subjects, generosity, mercy, and picking good subordinates were all highly
traditional tests of good government within the established value systems
of the Greco-Roman world. In the west in , for instance, Ausonius cele-
brated the Emperor Gratian in his Gratiarum Actio with a discussion exactly
framed according to these categories. In general terms, Themistius was try-
ing to build consent to its rule by emphasizing how superior Theodosius’s
government was to its predecessor in areas that were generally recognized
as fundamental to determining the overall character of a regime. More
particularly, his comparative approach, holding Valens always before his

 Or. .b. On Themistius’s post-mortem critiques of his former imperial patrons as a general
phenomenon in his career, see Heather and Moncur : ch. .
 See Or. , esp. b–a; Or. , esp. d–c, c–d with the fuller discussion in the intro-
duction to the translation of Or.  in Heather and Moncur : ch. . Themistius’s posthumous
treatment of Valens must be compared to those from the emperor’s lifetime, which made virtue of
the emperor’s relative inexperience: e.g. Or.  trans. in Heather and Moncur : ch. , or Orr.
 and  trans. in Heather and Matthews : ch. . Themistius’s need to work around Valens’s
inexperience, even during that emperor’s lifetime, suggests that the ingenious attempts of Woods
() to find a distinguished career for him are misplaced.
 Liar in winter: Themistius and Theodosius 

audience, also reflected the careful manner in which the new regime sought
to appeal to one of its most natural constituencies: those who had lost out
under its predecessor. Particularly in the aftermath of the attempted usurpa-
tion of Procopius in , and again in the magic-cum-treason trials of the
early s, as John Matthews has again done so much to elucidate, Valens
had cut a considerable swathe through sections of the elite classes of the
east. These families formed a natural group of potential supporters for the
new regime, whom Theodosius carefully cultivated by reversing some of
his predecessors’ policies and decisions. In particular, Theodosius returned
part of the lands of condemned individuals to their relatives, legislation
which presumably lies behind the oft-quoted cause célèbre of some Gala-
tian youths rescued from poverty. The new emperor also carefully showed
himself much more generous with the consulship than the Valentinianic
dynasty. The year  was Theodosius’s vicennalia, his fifth anniversary,
and the Valentinians customarily held the consulship themselves on all
such occasions. Theodosius, however, granted it to Saturninus, the general
responsible for negotiating peace with the Goths, and Themistius made
determined play of his emperor’s greater generosity, praising it as a trait
which would encourage more men to greatness. Both very specifically
and more generally, then, Valens’s regime gave his successor a useful target
by which to orient and justify his own regime. Interestingly, Constantine
had done much the same after taking over the east from Licinius, targeting
those who had lost out under his rule.
What emerges from all this is something of how a beleaguered regime
sought to entrench its position using a mixture of ideological claims about
the manner in which it would govern, supported by practical measures
calculated to back up those claims and maximize political support. This
came on top of, no doubt, much careful work behind the scenes. It was
not enough just to offer lots of favours. To get the maximum benefit from
such distributions, they had to be given to the right people. Themistius
will have been involved in this process too. He knew many, perhaps even
all, of the senators of Constantinople personally. Some he had recruited
himself in /, when he had toured some of the leading cities of the
eastern Mediterranean looking for suitable recruits during Constantius’s
senatorial expansion. He then became a standing member of the committee
to vet further candidates for admission, and hence came into contact with
later recruits too. His later claim to have been personally responsible for
 Matthews : chs. –.
 CTh ..– with Or. .c–d and Or. .d; cf. Vanderspoel : –.
 Or. .–.  Heather .
 peter heather

expanding membership of the senate from  to , is presumably an

overstatement, but many Constantinopolitan senators clearly owed their
elevation at least in part to Themistius’s support.
Policy in the later Roman empire was largely made by the interaction of
factions and interests. With established influence over a substantial body
of senatorial opinion, it is not too much to see Themistius as one of these
faction leaders from the late s, the head of one of a series of interest
groups whose haphazard interaction shaped imperial destiny. A possible
analogy to the kind of faction management such an assembly of individu-
als required in a world without parties might be the desperate list-keeping
of that eighteenth-century English parliamentary grandee, the Duke of
Newcastle, whose efforts to garner support were famously explored in the
works of Namier. Newcastle had various types of influence over a wide
range of individuals, but this never translated into automatic support. Nei-
ther Themistius nor his emperors needed to win votes, but all imperial
regimes required a broad base of support, and no one was better placed
than Themistius to deliver the backing of a substantial block of senatorial
opinion when imperial policy required consent. Themistius had in New-
castle’s terms his own substantial ‘interest’, which, following his role in its
recruitment, would naturally have viewed him as its leader, or one of its
leaders, and would have been alienated from any regime which had not
included him in some way in its distribution of power and patronage.
Set in context, therefore, Themistius’s fifteenth oration provides insight
into some of the measures Theodosius’s regime adopted to defuse the
potential political crisis that faced it in winter /. New ideological
emphases were matched by redoubled efforts to recruit supporters and
convince key constituencies that this was a regime deserving support. The
chances of source survival mean that Themistius’s part in these processes
features centre stage for us. A pagan supporter of Christian emperors, who
had been busy recruiting senators for thirty years, his position was certainly
unique, making him a figure well worth careful cultivation, even if he was
not the only political fixer that the new regime needed to get onside.
Fortunately for us, however, Themistius also had another string to his bow.
His rhetorical prowess combined with the interlocking web of connections
 The claim is made at Or. .xiii. Themistius recruited six out of thirty new senators who appear in
letters of Libanius dated ca. /: Petit : –. A letter of Libanius from the Constantian
period pictured Themistius sleeping gently on the riverbank, while senatorial fish landed themselves
(Ep.  trans. Norman  as Ep. ). Standing committee: CTh .., which stayed on the books
despite Themistius’s relative eclipse under Julian, so that he presumably had some contact with all
subsequent promotees.
 On the Duke of Newcastle, see esp. Namier .
 Liar in winter: Themistius and Theodosius 

he enjoyed within the Senate of Constantinople made him an expert at

selling imperial policies to the political classes of the east. Not only could
he help Theodosius deal with the internal political problems involved in
relaunching his regime, but he also had a contribution to make to the other
great political story rumbling around the eastern capital in the winter of
/. In the face of a second major military setback, exactly how was the
Gothic problem to be resolved?

themistius, theodosius, and the gothic problem

It would seem that Themistius was originally going to say nothing explicit
in Oration  about the possibility that there might be an unexpected
solution to the Gothic problem. The fact that, as we have seen, the speech
made only brief reference to Theodosius the military commander, while
the bulk of the speech concentrated on the overwhelming importance to an
emperor of his civilian government, could, by itself, represent no more than
a fairly straightforward attempt to deflect criticism away from the emperor
for the military failures of the previous summer by not mentioning it. A
last-minute addition to the speech, however, suggests that a much more
radical response to defeat was being discussed within the inner circles of
the regime in winter /.
A key piece of evidence for dating the oration is the passage which
records the arrival in Constantinople of the Gothic king Athanaric (c–
b). As Malcolm Errington astutely observed, it looks like a late insert,
skilfully woven into an existing draft. The passage is short – half a page in
the Teubner edition – and were you to remove it the argument before and
after, concentrating on the beneficial effects of the justice of the emperor,
runs together without any kind of jump. Apart from shedding light on
how Themistius composed his speeches, the passage is striking for what
Themistius has to say about the Gothic king. His overall conclusion from
Athanaric’s arrival in Constantinople was that it proved that enemies were
better subdued by persuasion than by force (c–a). Athanaric had by
this date been abandoned by the majority of the Goths who had previously
followed him, and was of little practical importance. Old and sick, the king
arrived in Constantinople on  January , and died there thirteen days
later. Nonetheless, the passage has huge significance because it briefly
rehearsed the argument which Themistius would deploy at much greater

 Errington b: –. Athanaric in Constantinople: see above note .

 peter heather

length in Oration  to justify the full peace agreement of  with all the
hitherto unsubdued Goths.
In that speech, Themistius himself acknowledged that the Goths who
defeated and killed Valens and his army at Hadrianople in  had not
been wiped out in the four further years of war which followed (Or.
.a). This, it must be emphasized, was entirely contrary to the expec-
tations sown by Theodosius’s original self-presentation and the general
demands of Roman ideology, which considered victory over barbarians as
the explicit sign of God’s favour for any ruling regime. Themistius faced
a huge problem, then, in trying to justify a peace with the Goths which
obviously represented so much less than outright victory. In Oration ,
he attempted to satisfy public opinion by positing a range of arguments
which cumulatively suggested that the best possible outcome had nonethe-
less been achieved. The least subtle of them was the claim that the Goths
had been so thoroughly subdued that they might as well have been exter-
minated (a). More fundamentally, Themistius argued that, even if the
Goths could have been destroyed, it was much better overall that they
had not. In making this case, Themistius came close to admitting that
destruction of the Goths had not been a practical possibility:
For just suppose that this destruction [of the Goths] was an easy matter and that we
possessed the means to accomplish it without suffering any consequences, although
from past experience this was neither a foregone nor a likely conclusion . . . (a)
To state unequivocally that the empire had in fact been unable to defeat
the Goths would have compromised the martial dominance expected of
an emperor, so this thought was taken no further. Rather than focusing on
the degree to which circumstance had constrained the empire into a com-
promise peace, Themistius concentrated instead on the positive benefits
that flowed from ‘forgiving’ the Goths. They were spared, he argued, to
serve the empire as farmers and soldiers, increasing both imperial revenues
by paying taxation and the overall military power at Theodosius’s disposal
(Or. .a–d). Overall, the emperor’s decision and capacity to persuade –
rather than defeat – the Goths had brought everyone to a better outcome.
The inclusion of this basic line of thought in Oration  strongly suggests
that, already by January , Theodosius and his advisors had begun to
think the unthinkable: that a compromise peace might have to be made
with the Goths. This was so highly charged a thought, and so controversial
 In Oration , the paying of taxes on agricultural production was put in the future (Or. .a–d),
the speech concentrating on a description of the repopulation of Thrace after the devastations of
war. Oration . of / presented the paying of taxes as now an established fact.
 Liar in winter: Themistius and Theodosius 

compared to its elite audience’s expectations of victory, that the argument

absolutely could not have been included without the emperor’s express
approval. At the very least, the passage must rank as what is known in
British politics as ‘kite flying’: airing a possible policy shift in public, via
an indirect means, to begin to prepare public opinion for its adoption,
should it prove necessary. Perhaps no final decision had yet been made to
negotiate a compromise peace with the Goths in January , but we can be
confident that it was already being discussed. My own suspicion for what
it’s worth, in fact, is that the decision had already been made, in response
to the lamentable events of the previous summer, and that Oration  was
a first shot in the long propaganda campaign which the emperor and his
advisors knew would be required to prepare public opinion for such a
massive volte-face.
Whether all the likely details of a potential peace deal had been ham-
mered out already in January , we do not know. Indeed, it is precisely on
the details of the final treaty of October  that we are sadly ill-informed.
We know that it was presented to the Roman public as a Gothic surrender,
a deditio, but that its detailed provisions ran contrary in some key ways
to the normal range of expectations that that term conjured up: hence
Themistius’s long and convoluted argument showing why it was, in fact,
better that the Goths had not been militarily destroyed. The Goths received
lands to farm (under legal terms and in geographical areas unspecified) and
incurred obligations of some kind both to pay taxes and to fight for the
Roman state in return. The extent of the Emperor Gratian’s involvement
in the deal is also unclear. It is traditional to regard the policy as Theo-
dosius’s own, not least because Themistius claims as much in Oration ,
but Themistius would say that, and, by itself, the statement can carry no
more weight than any other of the many unsubstantiated claims that our
orator made for different emperors in the course of his thirty-year career
as an imperial spokesman. What we do know is both that Gratian’s troops
had been responsible for the military side of the Gothic war from late
summer  when Theodosius handed control back to him, and that the
 So too Errington b: .
 Halsall (: –) argues that there was no single peace deal and no ceremony in Constantinople.
He is not the first to do so (so too McCormick : ), but this ignores the very specific date
given for the peace of  October  and Themistius’s clear references to a ceremony at Or. .b:
“But since I stood an eyewitness of that day on which he brought in peace quietly and calmly as if
at a sacred rite, and, to those who were then in the deepest despair, presented the barbarians giving
up their weapons voluntarily, and we saw clearly and in the flesh, what we had previously known
through pictorial representations.”
 On the norms associated with the term deditio in late Roman diplomacy, see Heather  with
refs.
 peter heather

western emperor hovered over the Balkans in both  and . This makes
it entirely possible that the peace deal was the agreed policy of both the
eastern and western halves of the empire.
But whatever its other uncertainties, there is no real doubt, despite some
recent arguments to the contrary, that it broke new ground by licensing
the continued autonomy on Roman soil of the Tervingi and Greuthungi
who had crossed the Danube in  and killed the Emperor Valens two
years later. A series of contemporary commentators emphasize that this
was the bottom line of the treaty, and it was precisely this point which had
Themistius squirming so hard in Oration  of January . Not only was
he then forced to argue, as we have already seen, that it was a good thing
that the Goths hadn’t been destroyed, but he closed with this thought
(c–d):
Our times are not the first when it has come to pass that those who have transgressed
have found forgiveness and thereafter been of use to those who had been wronged.
Look at these Galatians, the ones on the Pontus. Yet these men crossed over into
Asia under the law of war, and, having depopulated all the region this side of the
Halys, settled in this territory which they now inhabit. And neither Pompey nor
Lucullus destroyed them, although this was perfectly possible, nor Augustus nor
the emperors after him; rather, they remitted their sins and assimilated them into
the empire. And now no one would ever refer to the Galatians as barbarian but
as thoroughly Roman. For while their ancestral name has endured, their way of
life is now akin to our own.
The fact that assimilation had to be forseen as a future development under-
lines that continued Gothic autonomy was the new – and potentially
worrying – key feature of the treaty of .
Not only was Themistius implicated in efforts to head off Theodosius’s
internal political problems in January , then, but his rhetorical skills
were also mobilized to deal with their underlying cause: the lack of a
straightforward victory over the Goths. In the short term at least, both
gambits seem to have worked. Theodosius’s Gothic policies did not lack
critics, even in the east, but the regime never faced the direct challenge
of a usurpation originating within the territories given to Theodosius in

 For this argument in more detail, with full refs, see Heather and Moncur : ch.  (introduction).
 Continuing Gothic autonomy is signalled, beyond Themistius, in sources both sympathetic to
Theodosius and his treaty (Pacatus, Pan. Lat. ..–, where the Goths are one of a series of
foreign peoples serving Theodosius, the continuity of their existence up to ca.  confirmed by
the others with which they are grouped there) and hostile to it (Synesius, De Regno – with the
commentary of Heather ). Halsall (: –) oddly argues that there is no evidence that any
continued Gothic autonomy was licensed in . He appears not to have read the closing words of
Themistius’s speech closely enough. Kulikowski  is similar.
 Liar in winter: Themistius and Theodosius 

January . The emperor’s fifteen-year reign saw no attempted coups

nor even the need for drastic clampdowns such as the magic-cum-treason
trials instituted by his predecessors. Between them, the emperor and his
chief allies seem to have put together a plausible-enough ideological justi-
fication of Theodosius’s regime, with a working coalition of elite allies of
sufficient power, to stifle potential opposition.
This was no mean feat, but it was not, of course, a real solution to
the Gothic problem. The empire had no history of tolerating even the
partial autonomy of militarily significant immigrant groups on its soil, and
the peace of  clearly was not meant to mark the start of a new trend.
Subsequent groups of immigrants, not least further Goths who came to
the Danube in , were treated with nothing like the same generosity.
There can be no doubt, in other words, that the peace of  reflected a
momentary balance of power between the empire and one particular set
of Goths, created by two Roman defeats in short order, rather than the
dawn of a new era in the underlying aims and ideologies of Roman foreign
relations. As Themistius’s closing remarks in Oration  make so very clear,
most elite Roman opinion expected – and sooner rather than later – that
normal service would be restored. Elite Romans were taught from an early
age that the correct, Divinely ordained pattern of affairs consisted of total
barbarian subservience to Roman imperial might. This world-view was not
abandoned overnight in , and no one, I suspect, would have been more
surprised than the Emperor Theodosius and his orator that the tolerated
Gothic autonomy inaugurated – temporarily in their view – in  was
destined to have massive historical significance.
That it did so was due in part to the career of Alaric. Thirteen years after
peace was signed, as the story traditionally goes, the treaty Goths broke out
into open revolt under his leadership and spent the next two years in a Greek
Odyssey, which took them as far south as Athens, round the Peloponnesus,
and back north again to Epirus. After a brief rest, Alaric then moved his
followers into Italy in /, before returning to the Balkans until , when
the Goths headed west again, definitively this time, spending  to  in
Italy once more, before taking off for Gaul, where they finally settled down
in Aquitaine in the late s. And from the Aquitanian settlement there
evolved, in due course, an independent Gothic kingdom, which formed
one of the major successor states to the western Roman empire. In such
accounts, a direct line runs from Theodosius’s Gothic peace settlement to

 On ‘normal’ Roman immigration policies, and their continued application after , see Stallknecht
; Heather : –, –.
 peter heather

the end of the western empire. This linkage was never doubted in the past,
but new expectations that barbarian identity will always have been fluid
have fuelled demands in recent years that the correspondence between the
Goths who made peace in  with Alaric’s rebellious following of , and
hence the link with the end of the empire, should be proved. Can it?
In simplistic terms, the answer has to be no. No Roman source lists in
detail the sources of manpower drawn upon by Alaric in , or describes
exactly how he mobilized support. On the other hand, we are talking about
the middle of the first millennium here, so this is hardly surprising, and
it is important not to use demands for an inappropriate level of certainty
as an excuse for denying reasonable probability. In my view, a good case
can be established that in  Alaric did indeed lead a major revolt on
the part of the treaty Goths of . The argument is not that all those
settled under treaty necessarily participated in the revolt, or that others
from outside didn’t join in, but rather that there was sufficient overlap in
manpower between those Goths settled under the agreement of  and
Alaric’s followers to allow the basic point to stand.
The first plank in the argument is the fact that Alaric’s following is
described precisely as the  Goths in revolt by our two earliest, least
problematic, entirely contemporary, and independent Roman commenta-
tors on his rebellion: Claudian in the west and Synesius in Constantinople.
To discredit their testimony, convincing reasons need to be found for both –
writing in separate halves of the empire, for different audiences and for
different purposes – substantially to have misrepresented the action. None
has yet been offered. This basic observation – powerful enough in itself –
can be strengthened. The testimony of Synesius and Claudian has been
put aside sometimes in recent years on the basis of a passage in Zosimus
which reports that Alaric originally revolted because Theodosius had only
given him the command of some barbarian auxiliaries on the Eugenius
campaign rather than a proper Roman command (..). From this it has
been supposed that his ambitions and hence his revolt in  did not orig-
inally encompass the mass of Goths settled in the Balkans under the treaty
of . I see three major problems with this method of argument.

 On supposed fluidity of barbarian identities, see for instance, the essays and tone in Gillett .
The following paragraphs respond largely to the comments of Kulikowski in that volume, which
have heavily influenced the discussion of Halsall . I myself consider the emphasis on fluidity
to be capturing only one element of the total picture: see Heather .
 Claudian, De Bell. Get.  ff.,  ff. dating to ; Synesius, De Regno – (with Heather )
dating to . Neither Kulikowski () nor Halsall (: –) offers any explanation of the
fundamental distortion they suppose these authors to be incorporating.
 Liar in winter: Themistius and Theodosius 

First, what was originally Eunapius’s contemporary account of Alaric

has become generally and demonstrably mangled at the hands of the sixth-
century Zosimus. Putting the contemporary Claudian and Synesius to one
side because of three lines (literally) of the much later Zosimus whose
account is anyway problematic, with no further argument about why
they should both have distorted the action in the same way, is therefore
simply bad methodology. Second, rewriting Alaric as having essentially
Roman ambitions runs into the problem that, just four years after Alaric’s
revolt began, an East Roman general of barbarian origins – Gainas – took
the opportunity of the revolt of some Gothic auxiliary troops to lever
himself into power in Constantinople. Rewriting Alaric on the basis of
Zosimus makes him into an analogous figure, as those who want to do so
acknowledge. This generates the further problem, however, that Synesius
had no problem in describing Gainas and his activities with an entirely
reasonable degree of accuracy. Why should he have so misrepresented
Alaric when, though certainly hostile to Gainas, he could and did describe
his activities relatively straightforwardly? Third, we can be certain that,
from the beginning, Alaric’s following amounted to a major military force,
surely , warriors plus, since it was able to face down, already in , a
full Roman field army. If we don’t accept what Claudian and Synesius tell
us, therefore, that he was leading the treaty Goths in rebellion, we also have
to find a large alternative source of military manpower for Alaric. This is
not straightforward given that Stilicho had both eastern and western field
armies under his command at this point.
The second plank of argument, quite simply, is that it is entirely plausible
that the Goths of  had maintained sufficient continuity of political
identity over the intervening period actually to mount such a revolt. We
are only talking thirteen years. Another generation will have come to
maturity in the intervening period, but many of the adults active in 
 The basic approach with slightly different outcomes of Liebeschuetz ; Kulikowski ; Halsall
: –. Amongst other problems Zosimus conflates Stilicho’s two campaigns against Alaric
(of  and ) and wipes out ten years of the history of Alaric’s Goths in making the join between
his two main sources here: Eunapius and Olympiodorus (at Zosimus ..; cf. Heather : ).
To say that Zosimus had no real grasp of Alaric’s career, therefore, is an understatement.
 The activities of Gainas are well covered, if certainly with hostility, in Synesius, De Providentia; cf.
Cameron, Long, and Sherry .
 To my mind, this is why Liebeschuetz () cannot be correct in viewing Alaric as leading no
more than a regiment or two of Gothic auxiliaries in . Halsall (: –) tries to deal with
this problem by continuing to deny the overlap with the Goths of  while accepting that Alaric’s
armed following must have been large, mostly Gothic and from the Balkans. Having accepted these
points, he is most of the way to the conclusion that Alaric led the  Goths in revolt. He resists
this conclusion because he doesn’t believe there was a peace deal in  which licensed continued
Gothic autonomy, but see notes  and .
 peter heather

will still have been alive in . And while woefully ignorant of many of
its details, the whole point, overall, of the  treaty – for supporters and
critics alike – was, as we have just seen, that it allowed an unprecedented
degree of autonomy to continue among the treaty Goths. Although guilty
of rebellion and the death of an emperor, the Goths had not been broken
up and distributed very widely across the empire, which is why Themistius
had to work so hard to sell the peace to the Senate of Constantinople. This,
of course, makes it entirely plausible that the same Goths could have acted
in concert again, just thirteen years later, in .
Third, I would also argue, although this certainly requires a greater
degree of argumentation, that the compromises involved in the treaty left
unresolved two big issues in Goth–Roman relations, and it was precisely
these two issues which came to a head in Alaric’s revolt. First, the Romans
had recognized no overall Gothic leader in the peace of . This was
very much in line with established Roman policies for limiting the politi-
cal cohesion of groups they perceived as potential threats: standard policy
towards Alamannic over-kings, for instance, in the fourth century. It was
also facilitated by developments within the confederations of the Tervingi
and Greuthungi themselves. In both, the decision to move into Roman
territory had been accompanied by political turmoil at the top, and the
removal of established leaderships, whether by death in battle or political
overthrow. But not recognizing a single Gothic leader did not prevent
struggles continuing for the overall leadership of the treaty Goths. These
were demonstrably under way already between  and , and continued
after . Aiming for such a position, undisputed leader of both Tervingi
and Greuthungi, was clearly part of Fritigern’s manoeuvring before Hadri-
anople, and there is reasonable evidence in my view both that the same kind
of ambitious jockeying for position continued after  and that Alaric was
its eventual beneficiary. The best example of such jockeying is provided by
the famous quarrel over policy between Fravitta and Eriulph which erupted
at a banquet Theodosius held explicitly for the leaders of the Goths in .
Both led factions and both held different views over the right ordering of
Goth–Roman relations. There is thus every reason to see their rivalry as
a continuation of the type of attempted self-elevation that Fritigern had
engaged in.

 See further Heather .

 Neither the original leaderships (Athanaric, the dynasty of Ermenaric), nor their successors (Frit-
igern, Alatheus, and Saphrax) survived the struggles of –: in more detail Heather : –.
 Eunapius, fr. , dated by the summary of it at Zosimus .. Further discussion Heather :
–. That Theodosius should have held such a banquet undermines the contention of Halsall
 Liar in winter: Themistius and Theodosius 

As to Alaric himself, it is certainly possible that his position evolved

in the course of  and . Although in my view more probably a
mere confusion, Alaric may originally have had ambitions for a more
Roman career, as the passage from Zosimus might suggest. Such switches
of career trajectory were open to leading Goths, for example Fravitta, who
became a purely Roman general after murdering Eriulph at the banquet
(but Eunapius tells us that Theodosius’s guards had to intervene to prevent
Eriulph’s followers from killing him there and then, so I take it Fravitta had
no choice). But even if the thought of a Roman career was originally there,
Alaric did not take this option and there is good evidence that his eventual
leadership of the Goths involved at least one further round of competitive
politicking. His later career and that of his successor Athaulf were dogged
by the interventions of a Roman general of Gothic origins by the name of
Sarus, who waged a one-man war to undermine any peace deal they were
in the process of negotiating with the west Roman state, into whose service
Sarus had moved. What’s so very interesting here is that Sarus’s brother
Sergeric organized the coup in which Athaulf and his immediate family
were killed and made himself – briefly – ruler of Alaric’s Goths. This shows
that Sarus came from a family grand enough to compete for the overall
leadership of the Goths, and his unrelenting hostility indicates that Alaric’s
rise was probably responsible for his departure for Roman service. Seeing
Alaric as the beneficiary of continuing power struggles among the treaty
Goths takes a little bit of arguing, then, but not that much, and helps
explain one aspect of his subsequent agenda. Alaric constantly demanded
Roman recognition of his position in a break with the type of engineered
political chaos that the treaty of  had attempted to maintain among the
Goths’ leadership.
Alaric’s broader political success among the Goths, moreover, was intim-
ately linked to the line he took on what I would argue was the second issue
unresolved in : the military obligations owed by the semi-autonomous
Goths to the Roman state. It was normal Roman policy in the peace
agreements it imposed upon outsiders to extract drafts of young males as
(: –) that Alaric couldn’t have been the leader of the Goths of  in revolt, because there
is no evidence that their socio-political hierarchies had continued in being after that date (Halsall
does not discuss the incident).
 In more detail, Heather : –. Neither Kulikowski () nor Halsall () discusses the
Sarus–Sergeric linkage and its broader potential significance.
 In my view, as I’ve argued fully elsewhere, extracting Roman recognition of his leadership was
precisely the significance of the generalship Alaric periodically demanded of compliant Roman
regimes – along probably with the financial package for his followers that came with it. But note
that the generalship was an optional extra which he was ultimately willing to drop to make a deal:
Heather : ch. .
 peter heather

recruits for its armies. This may well have happened in , creating Gothic
auxiliary units in the regular Roman army, two of which are listed in the
Notitia Dignitatum. But, as had previously been the case with the Tervingi
from , the treaty seems also to have stipulated that the Goths would
provide irregular military service in the form of larger, autonomously led
contingents, for specific campaigns. Contingents from the Tervingi fought
on four separate occasions for Rome against Persia between  and 
and similar contingents were turned out from the treaty Goths by the
east Roman Emperor Theodosius I for his two civil wars against western
usurpers: Maximus and Eugenius.
There is patchy but compelling evidence that this military service was
resented by the Goths. On each of the campaigns against the usurpers,
Gothic participation was accompanied by revolts of some kind among
the treaty Goths. Theodosius’s decision to seek assistance on the second
occasion also prompted a vicious quarrel among the Gothic leadership,
which was linked to different views of how they should respond to the
request. The fate of the Gothic forces on the second expedition also
shows precisely why this participation should have been such a problem.
At the battle of the Frigidus in September , the Goths found themselves
in the front line on the first day and suffered heavy casualties. Orosius
even comments that the battle saw two victories for Theodosius: one
over Eugenius but a second over the Goths. Given that the Goths’ semi-
autonomy was tolerated by the Roman state only because they hadn’t been
properly defeated, there was a real danger that such casualties would change
the balance of power sufficiently to allow the Romans to rewrite the terms
of the treaty. It shouldn’t seem in the least bit surprising, therefore, that
pretty much as soon as they got home (sometime in winter /) the treaty

 Kulikowski (; largely followed by Halsall : –) denies large-scale Gothic military
service by the treaty Goths of  in the years between the treaty and Alaric’s revolt in , but this
involves too much special pleading to be convincing. For the Maximus campaign, Pacatus, Pan.
Lat. ..– strongly implies that the main Gothic contingent was recruited only for the campaign
(especially his explicit comment that it would have been dangerous to leave the Goths behind:
contra Kulikowski and Halsall) while Eunapius, fr.  and Zosimus .. note Maximus’s attempts
to undermine the recruited Goths’ loyalties, which again makes most sense if this was something
unusual. For the Eugenius campaign, a range of sources note the participation of large numbers
of Goths (Zosimus .; John of Antioch, fr. ; Orosius ..) and Theodosius’s banquet for
the Gothic leaders (see note ) was held precisely when Theodosius was mulling over his answer
to Eugenius’s envoys (Zosimus .). In my view, the banquet was probably a first move towards
securing Gothic participation for this second civil war.
 Maximus revolt: Eunapius, fr. ; Zosimus .., –. Banquet: see previous note. Alaric of course
led the revolt after the Eugenius campaign. The arguments of Kulikowski  and Halsall :
– comment neither on the Maximus revolt nor on the significance suggested by the exact
chronology of the banquet quarrel.
 Liar in winter: Themistius and Theodosius 

Goths rose in revolt under a leader committed to rewriting the terms of

.
Much of what we would like to know about the treaty of , and the
pattern of Goth–Roman relations it dictated, is beyond recovery. But, as
with many diplomatic agreements, it would appear to have been a working
compromise for a particular moment, which left certain larger issues of
contention to be resolved later. And working from what we know, it is
entirely plausible to suppose that Alaric’s revolt of  had the character our
two contemporary commentators ascribe to it. He was the leader of the bulk
of the  Goths in revolt, the treaty having precisely left them autonomous
enough to look to rewrite their terms of agreement by collective action,
and losses at the Frigidus having given them a real reason for doing so.
The available information about the treaty itself and these incidents which
punctuated its operation down to  strongly suggest the continuity of
Gothic military-cum-political agendas after .
To that extent, then, a line of development does run from the comprom-
ise peace that Themistius started to sell in January  and the deposition
of the last western Emperor, Romulus Augustulus, in late summer . It
was not, however, a very direct line. For the settlement of  to develop
into a Gothic successor state based in southwest Gaul, a series of exogenous
interventions were required, not least a series of further outside invasions.
Radagaisus’s assault upon Italy in / was necessary, for instance, to
provide a large body of further recruits which Alaric could draw on when
he himself moved definitively to Italy in . It was this extra military
power which eventually made it impossible for the Roman state not to
accept the fundamental revisions to the old treaty of  encapsulated in
the Goths’ new treaty and subsequent settlement in Aquitaine in the later
s. Equally if not more important, the further intrusions of Burgundians,
Anglo-Saxons, Franks, but first and foremost Vandals, Alans, and Sueves
were required. These both destroyed important elements of the west Roman
field army, and, above all, removed key tax-producing provinces from
central imperial control, resulting – the aggregate effect of their thoroughly
separate activities – in a massive decline in the power of the western
Roman state. All of this was crucial to the creation of an independent
Gothic kingdom on the basis of the descendants of Alaric’s followers. It

 Orosius .. (casualties confirmed at Zosimus .). Neither Kulikowski () nor Halsall
(: –) discusses this backdrop to Alaric’s revolt.
 In recent times, we have seen one successful example of this kind of diplomatic strategy in the Good
Friday agreement over Northern Ireland, and one so far unsuccessful example in the Oslo accords
dealing with the Middle East.
 peter heather

was not so much the case that these Goths carved out a kingdom from the
still living Roman body politic, but, rather, that they expanded the regions
under their control to create a kingdom when west Roman imperial power
was declining at the hands of others.
In that sense, I suppose, it would be wrong – even if you decided to
take an entirely Roman point of view – to be overly critical of the Emperor
Theodosius and his publicist Themistius. The emperor was probably gam-
bling on being able to overturn in due course the temporary power deficit
which had made a generous treaty for the Goths unavoidable in the early
s. But then additional factors intervened. To that extent, it was both
unexpected and entirely unforeseeable that the treaty of  would have
a major part to play in a sequence of events which led to the extinction
of the western Roman empire. In this, the treaty thus plays a similar role
in the historical revolution which it instituted to Jerome’s initial salvos in
the triumph of the Christian Latin Chronicle tradition. There, too, other
factors – equally unforeseeable in  – had to come into play to make its
victory over classical Roman historiography quite as total and irreversible
as it would eventually prove to be. And, in fact, the Goths were implicated
in at least two of them.
For one thing, classical Latin historiography was generally triumphal-
ist in tone. There could certainly be ups and downs, better and worse
emperors, but its subject matter was basically that empire and the victories
which kept it in being. West Rome’s total extinction in the fifth century,
as opposed to the occasional setback, thus made it a deeply problematic
genre to compose in. Christian chronography, based on the alternative
certainty of an unfolding story of long-term providential salvation, pro-
vided a much more attractive option. Just as important, the creation of
the Gothic kingdom – one of a series of warring successor states to the
western empire – also instituted a massive socio-cultural revolution in the
form of a thoroughgoing militarization of non-religious elite life. This was
to have major consequences for the power of state structures, undermin-
ing consent to taxation and introducing new types of military obligation
among landowning elites right across the former Roman west. In changing
upper-class career structures in this way, however, it also made sophisti-
cated literacies much less central to elite life, as landowners abandoned
bureaus for the battlefield. Despite continuing to pay lip service to classi-
cal cultural forms, therefore, successor state elites were no longer willing

 For this view of the end of the empire in more detail, see Heather .
 Liar in winter: Themistius and Theodosius 

in practice to pay for grammarians to educate their children in the old

styles.
The Goths and other outsiders who changed the politics of western
Europe so dramatically in the fifth century thus also destroyed the subject
matter, the authors, and even the potential audience for large-scale classi-
cizing histories. Arguably, this made the victory of Jerome’s new historical
form much more total in the Latin west than it would ever otherwise
have been, even if the rise of the Christian Chronicle was always going
to be a major cultural phenomenon in its own right. But if it were not
enough that both politics and culture were inching – if unconsciously –
towards major turning points in Constantinople in , dramatic events
with equally unforeseeable consequences were unfolding simultaneously in
the religious sphere. Early in , the Emperor Theodosius called a coun-
cil of the bishops of his dominions, which would also change the world
forever.
 There is a huge bibliography of relevant materials, but see, e.g., Heather  and Wickham ,
esp. chs. – on the evolution of the western successor states and their dominant elites.
 chapter 11

Moments of truth: Gregory of Nazianzus

and Theodosius I
Neil McLynn

One of the earliest surviving verdicts on the emperor Theodosius was

pronounced in (probably) , by the recent bishop of Constantinople,
Gregory of Nazianzus, some two-thirds of the way through his vast poetic
apologia, De vita sua. Emperors were never to be judged casually, least of all
when they were potentially within earshot – as Theodosius must have been,
of a poem intended for the Christian elite of Constantinople. Yet Gregory
begins in a resoundingly minor key: Theodosius was “not a bad man, in
respect of faith in God” (DVS ), at least “as far as the simpler sorts [can]
grasp” this (); and he was “exceedingly overcome by the Trinity” ().
Defeat is not usually expected of rulers, but Theodosius’s submission was
acceptable as “the principle for all hearts which operate on a solid basis”
(–). This modest praise is then opened to question, as Gregory begins
to weigh the emperor on a balance that is calibrated with elaborate (and
surely deliberate) obscurity. Theodosius was not so great in fervor of spirit
as to equate the present to what was past by using the opportunity to
heal completely the misfortunes inflicted by opportunists past (–).
Or rather (and only gradually do we realize that Gregory has here played
his favorite trick of rehearsing conventional wisdom in order to pick it
apart), Theodosius was equal to the task in fervor, but not in – and here
Gregory pauses, wondering aloud whether what the emperor lacked was
“bravery” or “brazenness” (–). He throws the question to his readers –
“you tell me!” – but quickly moves to suggest that the relevant quality, the
explanation for the emperor’s failure to make sweeping changes, should
 This is a significant juncture: a new section is signaled at DVS . For the date of the poem, see
Jungck : .
 Translators conventionally take Theodosius himself as the subject of the kratsai, mastering or
surpassing the simpler souls, but this requires an irregular accusative for the object (contra, e.g.,
DVS ); it seems instead that an auxiliary verb has been suppressed (cf. Or. .: Âson dìoÔn –m•
ginÛskein). The verb is used of holding orthodox doctrines at, e.g., Carm. .. .
 Tuilier, Bady, and Bernardi (:  and n. ) gratuitously emend the manuscript spl†gcnwn
to Spanän, to suit preconceptions about Spanish orthodox piety.


 neil m c lynn

perhaps be called “foresight” (); for Gregory’s Theodosius thought

that persuasion, not compulsion, was legitimate and would count for more
both for his own side and for those who were to be brought to God: for
anything unwilling that is mastered by force is like an arrow held on a drawn
bow, or a stream confined in a narrow channel, and when circumstances
allow would defy the force that restrained it; whereas the willing is secure
for all time, bound by the unbreakable chains of desire (–). This,
Gregory suggested, had been the reasoning behind Theodosius’s policy thus
far of restraining fear and attracting all men gently, “making his wishes into
a written law of persuasion” (–).
These twenty lines prompt reflection, not merely for what they might
imply about the policies of an emperor usually regarded as introducing a
new level of official commitment to the enforcement of Christian ortho-
doxy, but also as one of very few third-person assessments of a living
emperor to survive from late antiquity. My contention will be that Gre-
gory’s comments should be taken seriously, both as a reflection of and as
a contribution to a contemporary debate. On any reading, his meditation
upon Theodosius’s motives was unusual. Historians usually reserved their
judgments for safely dead emperors, consigning the living to the “broader
brush” of panegyric. But although panegyrists addressing an emperor
would regularly look inside their subject’s mind, divining the reasons
behind his behavior – the following year, for example, Themistius would
dilate upon Theodosius’s reasons for appointing Saturninus as consul –
there is no parallel for the way that Gregory here talks across Theodosius,
inviting argument over policies that might (he suggests) be misunderstood
and therefore wrongly criticized. He neither glories in his intimacy with
the emperor, as a eulogist might do, nor assumes the ominous authority of
a prophet. Instead he offers a dispassionate, carefully measured display of
parrhesia.
 For the significance of this term in the context, cf. below, p. .
 Tuilier, Bady, and Bernardi (:  and n. ) follow Jungck (: ) and emend the manuscript
›ggrafon to Šgrafon on the basis of allegedly parallel comments on Julian at Or. .; but here
he inverts the conceit. Gregory uses the present tense of Theodosius’s policy, implying that it was
still operative in : translators have converted it to the past (White : ) on the basis of
received belief about what the emperor did; cf. Errington : , reading these lines as anxious
commentary on Theodosius’s policy in November .
 For Theodosius as Nicene enforcer, see memorably Matthews : –. On judging emperors, see
Pan. Lat.  () .: existimare quidem de principe nemini fas est.
 Amm. Marc. ..: see further Mark Vessey in this volume, ch. , at n. .
 Them. Or. .a–d.
 Ausonius thus presented himself as Gratian’s familiaris notitiae secretus interpres (Grat. Actio .);
cf. Ambrose on Theodosius, De obitu Theod. –, dilexi virum. For Lucifer’s combative rhetoric, see
Corti .
 Moments of truth: Gregory of Nazianzus and Theodosius I 

This chapter will argue that this passage is a valuable source on Theodos-
ius and his impact on the Christian politics of the eastern empire; I shall
also suggest that it helps illuminate the purposes of Gregory’s verse auto-
biography. Broader examination of the seven-month period during which
emperor and bishop interacted at Constantinople, and which provided the
basis for Gregory’s judgment, will also reveal something of the workings of
the Theodosian church and state. The result will be to query the standard
view that Theodosius arrived with a firm set of goals and a determination
to implement these. In his first few months at Constantinople, Theodos-
ius indeed expelled the incumbent homoean bishop from the cathedral,
legislated impressively on behalf of Nicene orthodoxy, and convened a
council that would confirm this same creed. All this, however, does not
make him either the leader or the instrument of a Nicene party. Even after
he had done all this, there was still room for debate about his motives and
intentions.

the great entrance

In Gregory’s poem the immediate function of his verdict on Theodosius is
to announce the emperor’s arrival in Constantinople and to introduce his
own first interview with him. This encounter is presented as properly over-
whelming – the point Gregory makes of passing over the “honor” that the
emperor did him (–) bears out Themistius’s picture of Theodosius’s
genial charm. The emperor’s words are reported directly, a rarity in the
poem: “Through me, God gives the temple to you and to your pains” (–
). Gregory adds commentary and context, emphasizing the passionate
opposition of the city to this preferment and the likely futility of using
force to impose it. He returns again, significantly, to the emperor’s words,
and his own nervously thrilled reaction; then he turns to Christ, with a
prayer that links his pains then with the “evils” of his present situation.
This sets the stage for a justly famous scene. “The moment was come”:
soldiers had already stealthily occupied the cathedral, and the whole city
was marching out in opposition, bombarding Gregory with their curses and
the authorities with their prayers (–). Sandwiched uncomfortably

 In what follows I develop some arguments earlier presented in McLynn .

 Them. Or. .c.
 Compare Mamertinus’s savoring of Julian’s brief greeting at Pan. Lat.  () –.
 The “great and terrible fervor” (z”siv) at DVS  reprises the discussion of the emperor’s own
fervor at –.
 For the venue, see Mayer ; cf. McLynn : –.
 neil m c lynn

between a “general” and his army, Gregory shuffled into the church sick,
broken, and hardly breathing, gazing upwards (–). But then some-
thing “worth telling” occurred, an apparent miracle that Gregory intro-
duces with what seems at first disproportionate length (–). Dawn
had broken, but the city was still cloaked by clouds. The ceremony began
in inauspicious gloom. But at the very moment when Gregory and Theo-
dosius entered the sanctuary together and the congregation chanted their
acclamation (a liturgical innovation that had presumably been planned
beforehand), the sun broke through, to illuminate the scene in sudden
glory. Common prayer swelled into a thunderously unanimous petition
that Gregory, who had taken his seat on the presbyters’ bench, be installed
immediately by the emperor on the empty episcopal throne (–).
His frail plea for “moderation,” conveyed through a presbyter, was even-
tually heard, and the emperor departed with an expression of satisfaction
(–).
The episode is conventionally understood within a context supplied by
the fifth-century historian Socrates. He describes how the emperor had
fallen ill at Thessalonica but had then recovered and received baptism from
the bishop Acholius; he then made his entrance into Constantinople “not
many days” after his recovery, “on November th, in the fifth consulate of
Gratian, and the first of his own” (HE ..–). But Socrates’s Theodosius
does not bring Gregory to the cathedral. During the two intervening
generations the story had been sanitized: to minimize the rancor caused by
Gregory’s ultimate resignation from the see, Socrates has him waiting long
enough only to “express his joy at the emperor’s arrival,” before cutting short
his term in Constantinople and returning home to Cappadocia (HE ..–
). The historian therefore leaves the emperor to devise his own solution.
True to Socrates’s own irenic instincts, Theodosius “began to consider how
he could make peace, achieve unity, and enlarge the churches.” This
led him to approach Demophilus, who headed the “Arian” (homoean)
party favored by the previous regime, with the inquiry “whether he was
willing to believe in the synod of Nicea, unite the people, and establish
peace.” When Demophilus refused, he was ordered to “leave the houses
of prayer” (HE ..–). Socrates seems here to be working with a source
 Errington (: ), and McLynn (: ) assume that the “general” was Theodosius; on
balance this seems unlikely.
 Gregory had witnessed the effectiveness of such methods of receiving an emperor in church: Or.
., with McLynn : –.
 There being no other bishops present, we must suppose that Theodosius was being invited to
conduct the installation himself: he had reason to appreciate Gregory’s retrieving a delicate situation.
 For Socrates’s perspectives on the episode, see Urbainczyk : .
 Moments of truth: Gregory of Nazianzus and Theodosius I 

that was at least sympathetic to Demophilus, who despite the historian’s

carping commentary emerges with credit. He preaches the gospel of conflict
avoidance to his congregation and sets them an example by “holding his
assemblies thenceforth outside the city gates.” And so the Arians, after
forty years in control of the churches, “went out from the city”; their
departure is dated precisely to “Gratian’s fifth consulate, and the first of
Theodosius Augustus, on November th” (HE ..–).
Gregory’s account leaves no space for any negotiations between Theodo-
sius and Demophilus. That he himself was not the first-choice candidate
considerably dims the luster of his own appointment. But there is more to
his silence than wounded pride. The offer to Demophilus, despite the doc-
trinal condition imposed, raised awkward questions about Theodosius’s
ecclesiastical priorities. Had the bishop accepted, he would have posed
terrible dilemmas for Nicene leaders. Basil of Caesarea had long ago rec-
ognized Demophilus’s aptitude for blandly subversive coalition-building;
and his immediate associates included one of the great villains of contem-
porary Nicene demonology. Nor could the bishop of Rome easily have
done business with a man who had acted as jailer to a pope and extracted
from him an unprecedented doctrinal surrender; it had been expected at
Rome that a council would meet to fill the presumed vacancy. Relief
among Nicenes, at Constantinople and elsewhere, at Demophilus’s display
of integrity will have been balanced by anxiety at the creative ecumenism
that the emperor had displayed. Nor would they have received immediate
reassurance. Gregory’s entrance into the cathedral cannot be assumed to be
a direct consequence of Demophilus’s refusal.
Socrates supplies two dates, for Theodosius’s entrance to Constantinople
on  November (a date accepted by Sozomen, to be preferred to the

 Van Nuffelen (: –, –, ) suggests a “Theodosian” source here; but Socrates’s
commentary seems designed to undercut his narrative, as when he quotes Demophilus’s sermon
and then accuses him of misunderstanding his text.
 The forty years, significantly, are assigned not to the Nicenes in the wilderness but to the Arians; the
forty years of David’s reign ( Samuel :) were elsewhere applied to a homoean bishop: Auxentius,
Epistula de fide, vita et obitu Ulfilae , in Gryson : . Similarly, the date provided marks the
Arians’ expulsion rather than the Nicenes’ return.
 Errington (:  n. ) relates DVS – to the conversation.
 Basil, Ep. ; on the atrocities attributed to Lucius (cf. Soc. HE .., for his presence in Con-
stantinople), see Greg. Naz. Or. .. Philostorgius, HE . has Dorotheus of Antioch returning “to
Thrace” after being driven from his see in . For the earlier context, see Brennecke : –,
–.
 For Demophilus and Liberius, see the letter included in Hilary of Poitiers, Collectanea antiariana
parisina, ser. B VII . Damasus, Ep.  anticipates a council to choose a bishop for Constantinople,
without even considering Demophilus a potential candidate.
 neil m c lynn

alternative) and for Demophilus’s departure from the cathedral two days
later. This gives us our traditional sequence, three purposeful steps from
entry to expulsion to takeover. But there is no evidence for the date of
Gregory’s arrival. The confident modern consensus that gives Friday, 
November, the day after Demophilus’s departure, is derived only from
Gregory’s reference to the dawn: it has simply been assumed that this
was the very next morning. But we should allow the emperor and his
advisors time to recover from their surprise at the outcome of the interview
with Demophilus. They will have felt no need for precipitate action. The
one source to provide a date, in fact, puts the “restitution” of the church
to the orthodox in December. In inviting Gregory to preside at the
cathedral, moreover, the emperor was merely putting him on probation;
no commitment was implied.
This allows us to reconsider the impact of the scene inside the church,
as described by Gregory. The son et lumière that he presents to us was also
presented to Theodosius. To bring the emperor into what was probably
the grandest church he had ever visited, just at the moment when the first
rays of daylight streamed through the windows, was a nicely contrived
stroke; the effect was perhaps all the greater for being slightly delayed by
the clouds. Inside, Gregory’s people supplied the necessary drama. Their
acclamations demanding Gregory’s installation by the emperor were of
crucial importance. The tensions generated matched those in the streets,
so that the absent homoeans were not missed, even if (as is likely) the
cathedral was anything but full. Nor did Gregory wriggle awkwardly
into the urbane Demophilus’s shoes. Instead, he began a dialogue that
provided a counterpoint to the howling demonstration outside. Unlike
the enraged homoeans, moreover, the congregation showed themselves
willing to listen, even to the most frail and least authoritarian of voices.
This in turn spared Theodosius the difficulty of either yielding or refus-
ing. His praise to Gregory for restraining the congregation’s enthusi-
asm was no doubt sincere: the candidate had accepted the terms of his
probation.

 Soz. HE ... Descriptio consulum s.a.  has  November; for Marcellinus Comes, cf. below,
n. .
 Rauschen : , correcting Tillemont’s assumption that the takeover took place the same day.
 Marcellinus Comes, Chron. s.a. : eam [sc. ecclesiam orthodoxorum] . . . nostris catholicis orthodoxus
restituit imperator mense Decembrio. Croke (: ) accepts the date as “presumably” correct.
 DVS –, for homoean taunts that Gregory’s cathedral congregation could only “fill the gates.”
 Gregory was occupying the throne when he delivered Or.  shortly afterward; this suggests that
his main achievement at the initial service was to spare the emperor embarrassment.
 Moments of truth: Gregory of Nazianzus and Theodosius I 

laying down the law

Our tendency to assume that Theodosius took single-minded executive ini-
tiative over the church at Constantinople owes much to the contemporary
evidence that seems to show him exercising single-minded legislative ini-
tiative on behalf of Nicene orthodoxy. Two laws in particular, which would
be paired at the beginning of Justinian’s code as authoritative expressions
of imperial commitment to orthodoxy (CJ ..–), are easily read as mani-
festos. However, questions emerge when these two laws are examined in
their contemporary perspective, especially in the light of two texts produced
by Gregory.
The first law, the earliest surviving expression of Theodosius’s commit-
ment to Christianity, is the famous edict issued to the people of Con-
stantinople from Thessalonica in February , some nine months before
he met Gregory. Theodosius announced his wish that all his subjects
should follow the religion that Saint Peter had brought to the Romans and
that Bishops Damasus of Rome and Peter of Alexandria followed: “that
is, that in accordance with apostolic teaching and evangelical doctrine we
should believe in one godhead of Father and Son and Holy Spirit under a
like majesty and a holy trinity”; he further commanded that while all who
followed these teachings were to embrace the name of catholic Christians,
any dissenters were to be judged insane, to incur the “infamy” of heretical
dogma, and to see their meeting places denied the name of “churches”:
“they are to be smitten first with divine vengeance, and afterwards also by
punishment on our initiative, which we shall have taken up on the basis of
the judgment of heaven” (CTh ..).
Soon after the publication of the Theodosian Code, the lawyer-
historian Sozomen exploited this text to explain the emperor’s overture
to Demophilus, which now becomes rather less than an offer of sup-
port (HE ..). Demophilus is simply commanded to bring his people
into “harmony” with the Nicene masters (HE ..); Gregory has already
been established in the narrative as the city’s authoritative pastor (HE
..). But Sozomen’s explanatory scheme involves impossible chronolog-
ical distortions; moreover, the modifications constructed by his modern
counterparts are also fragile, for it has proved strikingly difficult to attach
any consequences to the law. Once posted in a city, imperial edicts invited a
 For discussion and a guide to the considerable bibliography, see Barceló and Gottlieb .
 Sozomen creates a direct connection between the legislation and Theodosius’s arrival in Con-
stantinople “not long afterwards” (HE ..), explaining Theodosius’s actions as a straightforward
implementation of the terms of the law.
 neil m c lynn

response from the populace; yet despite the notoriously passionate interest
of the people of Constantinople in theological issues, there is no indication
that any officials either acted, or faced calls to act, when Theodosius’s pro-
nouncement arrived. We do not have the complete text of the edict but
can be confident that the commissioners responsible for the code extracted
all legally relevant portions. They did not find much. Theodosius, no
doubt under pressure from Nicene lobbyists to show his hand, deferred
instead to “the judgment of heaven”: all who read his edict would under-
stand that action against the “Arian” establishment in Constantinople was
being postponed indefinitely. This was the most conditional of all pos-
sible commitments. It fits nicely with the “written law of persuasion” with
which Gregory credited Theodosius.
Our key evidence, however, comes from the speeches that Gregory deliv-
ered during this period. These are much concerned with what he considered
the dangerous zeal of some among his congregation at the Anastasia; he
continually deplores their overreadiness to engage in combat on behalf of
their faith. Hence the significance of his sole allusion to the terms of
the edict. This occurs in one of the bravura performances where he rev-
eled in the paradoxes of his position at Constantinople, as the imported
pastor of a motley flock (Or. ). He summons opponents, to dismiss

 For the publication of edicts, see Matthews : –.

 Ritter (: –) thus interpreted the law as a programmatic statement without legal force.
Errington (b: –) similarly rationalizes Sozomen’s version by interpreting the law as an
announcement of Theodosius’s intention to make Constantinople his capital, advising the inhab-
itants (and in particular the clergy: cf. following note) of the behavior expected from them; this
ingenious thesis matches neither the terms of the edict itself nor Theodosius’s actions when he
arrived. Cf. Ayres : , “a prelude to more precise action in the future.”
 Their efforts apparently yielded the separate fragment excerpted and included in the code as CTh
..: those who ignorantly confused or negligently violated the divine law committed sacrilege.
Errington (b: ) infers that the edict contained provisions against clergy; but “sacrilege” is
a generic term applied in contemporary legislation to improper petitions (CTh .., of )
or improper destruction of vineyards (CTh .., ); its usage here matches the figurative (and
tautological) quality of Theodosius’s anti-Manichaean law of , which holds offenders against an
earlier enactment “tamquam in ipsius depictae legis iniuriam veluti sacrilegii reos” (CTh ..).
 Errington (b: ; : ) renders the crucial words post etiam as “but later,” blurring their
restrictiveness. Escribano Paño (: –) argues that the infamia decreed against recusants had
the legal effect of depriving them of civil rights; but the context (haeretici dogmatis infamiam) does
not suggest technical usage.
 Humfress (: –) discusses the case briefly but fails to apply her own crucial qualification:
“Theodosius’ legislation had to be played out in actual disputes.”
 Note the emphasis on Theodosius’s “wishes” as the basis of his legislation at DVS ; the edict
begins, Cunctos populos . . . volumus.
 The most vivid passage is Or. .–; cf. Or. .–, .–.
 The context has puzzled commentators. Bernardi (: –) inferred from the address to Arians
that it was a pamphlet intended only for written circulation; McGuckin (a: –) interprets
it as a quasi-forensic reply to the Arians after an attempt to prosecute him, “making it clear to his
 Moments of truth: Gregory of Nazianzus and Theodosius I 

them with ruthless broadsides. The speech is composed of three distinct

movements. The first, beginning with a thunderous opening – “Where
are they, those to whom size is all?” – fashions Goliaths at whom he
can fire his slingshots. The Arians of Constantinople are matched against
him, their enormities against his modesty. He presents a battle with
dinosaurs, the lumbering survivors of an extinct regime. When Gregory
demands that each play their king, he trumps “their” Ahab with “his”
Josiah (); but then he launches immediately into a catalog of atrocities
committed under Valens, to contrast his opponents’ brutal recklessness
with his own moderation (–). In unmasking the murderous savagery
of the previous establishment, he presents an implicit manifesto for a
new one; however, his “Josiah” remains discreetly offstage. Moreover, as
Gregory shifts to the case being made against him personally (), we sense
that the Arian monster is being used as a mouthpiece for accusations com-
ing from closer to home. His accuser disdains his foreign origins and his
shabby clothes, so that the whole middle section of the oration becomes
a new version of the ancient quarrel between urban chic and rural virtue,
between the empty refinements of Constantinople and Gregory’s apostolic
simplicity (–). Gregory’s sarcastic zest (and his talent for dramatizing
his own squabbles) has obscured not only the bathos of this shift in the
charge-sheet, from his enemies’ homicides to his own fashion sense, but
also the point, which emerges by comparison with a parallel passage in his
autobiographical poem, that he is now dealing with accusations that come
not (or not only) from the homoeans at their cathedral but from inside
his own congregation. He overwhelms his critics with a Bible quiz –
a quick-fire burst encompassing Samuel’s Armathaim, Amos’s sycamores,
and Abraham’s nomadism (–) – which sets the scene for the third move-
ment, in which he interrogates his own behavior since his arrival in the city,
endowed (he permits himself a Pauline boast) “with no small authority”
(). Which “insatiable” exemplars had he followed? What aspect of “the
current situation” had he provoked to jealousy? He explicitly rubs home
the contrast with the “precedents” established by the Arians, so lovingly

hearers that he has gathered evidence” in order to deter further attacks. Both require an improbably
elaborate form of indirect dialogue between Gregory and his opponents.
 DVS –, locating resentment squarely among “my own.”
 Gautier (: ) relates this claim to a new authority conferred by Theodosius’s edict; but
Gregory refers explicitly to the circumstances of his arrival, and must mean his own rank as
(Cappadocian) bishop.
 The former expression recalls the provisions made against his own congregation’s ˆplhst©a at Or.
., , ; he had used kair»v as a point of comparison at Or. .,  (cf. , on the murder of
Eusebius of Samosata).
 neil m c lynn

detailed in his opening paragraphs: great care is taken to ensure that his
position should command assent.
He next turns to his imagined Arians, speaking now in his people’s
name: “What churches have we contested with you? What money?” The
implied answer is none; and here, for once, Gregory does not need to calm
the hotheads in his audience. Had any such contestation been thought
possible (as the letter of Theodosius’s law implied, with its denial of the
name of churches to their conventicles), Gregory would here have had
to apply the brakes. Instead, the next question refers explicitly to the
emperor’s command: “What disregarding of a royal decree did we jealously
resent?” (). The reference must be to the edict. But Gregory chooses
to pass over the treason that his heretics were committing in defying it
and remaining in their cathedral; instead, he invites his people to take
pride in not making an issue of this defiance. Their virtue can only have
been born of necessity. For had it been possible to act on the basis of the
edict, and to call upon the city prefect to dispossess the incumbents of the
cathedral, Gregory’s more zealous constituents would eagerly have done
so. Always happiest when scoring against his own party, Gregory plunges
on: which magistrates had they lobbied, against their enemies? Whose acts
of recklessness had they brought to court? Having faced much criticism
for his refusal to use the courts, he now enjoys his vindication. Here,
then, Gregory astutely exploits the perceived weightlessness of Theodosius’s
command to reinforce his own preferred position of defiantly aggressive
passivity. “And what about me?”: he shifts back to the singular and to
his central concern. Collective restraint becomes the basis for his own
personal decision to embrace the persecutors, his refusal to pursue those
who had thrown stones – a policy of “restraint” that had, as he notes, been
called “madness” ().
So here we see Theodosius’s legislation being recycled into ecclesiastical
polemic, to make a very subsidiary point against its ostensible targets,
the Arian heretics, and more significantly to provide a useful benchmark
against which Gregory could justify to his own more excitable followers
his own performance, or rather his refusal to perform. Moreover, Gregory
 For exact parallels to this sense of zhlotupe±n in Gregory, of claiming something for oneself and
resenting its mistreatment, see Or. .; ., . The most accessible recent translation, by P. Gallay
in Moreschini : , reverses Gregory’s meaning: “Quel décret impérial avons-nous méprisé,
pour que nous provoquions l’animosité?”
 Moreschini (: –) denies this, arguing that Gregory could not have relegated so revolutionary
a law to so low a profile.
 For Gregory’s elaborate defence of his conduct, see Epp. –.
 I owe this perspective on Gregory’s posturing to Veronika Grimm.
 Moments of truth: Gregory of Nazianzus and Theodosius I 

most definitely does not speak here as the city’s bishop-in-waiting. His
gestures of confidence in the future prospects for his community, of wolves
transformed into sheep or shepherds (), cannot be translated into an offer
to potential defectors. The text instead offers a valuable impression of
the atmosphere at Constantinople, the pressures operating on Christians of
different persuasions as they waited for Theodosius. Gregory’s performance
also illustrates the qualities that enabled him to rise so successfully to
the challenge of replacing Demophilus. The posture of argumentative
opposition that he strikes likewise shows that he was not at this point
mounting a direct challenge to the bishop. Even so, it might still conceivably
be argued that Gregory was at this point too marginal a figure to understand
the true significance of the edict. Was he, in fact, correct to minimize its
import in this way?
Our second law would suggest, on first reading, that Theodosius had
been committed to the extirpation of heresy from the outset. The emperor
addressed a letter to the praetorian prefect Eutropius from Constantinople
on  January , some six weeks after he had handed over the cathedral
to Gregory. The language is certainly uncompromising: “Let no place be
available to the heretics for their mysteries, let no opportunity be avail-
able for exercising the madness of a stubborn mind . . . Let the crowds of
the heretics be kept away from their illegal gatherings.” Theodosius then
provides a clear theological criterion by which orthodox “defenders of the
Nicene faith” are to be identified, involving due acknowledgment of the
holy spirit and of a single and indivisible divine ousia, the central planks of
Gregory’s own teachings. And the emperor now demands action against
those who refused these doctrines:
Let them be removed and barred completely from the premises of every church,
since we forbid all heretics to conduct unlawful meetings inside towns; and we
command that if a factional outburst does attempt anything, it should be driven
from the very walls of the city and the frenzy expelled, so that the catholic churches
all over the world might be restored to all the orthodox bishops, who hold the
Nicene faith. (CTh ..)

This measure, moreover, was clearly well publicized. When Ambrose of

Milan wrote to Theodosius toward the end of , he reminded him of
the “recent benefits” that he had heaped upon the church, quoting back at

 As McGuckin (a: ), who sees a threat to Demophilus’s clergy; cf. Bernardi : .
 Gautier (: ) argues for direct theological influence.
 For detailed discussion of the law, with full bibliography, see Escribano Paño : –.
 neil m c lynn

him his achievement of restoring the catholics to the churches. However,

no ancient source discusses the law directly. We must therefore infer its
scope and intention from its wording and context and once again, Gregory
will provide a useful perspective.
Much turns on Eutropius, the addressee of the law, who some fifty
years ago was established to have been praetorian prefect of Illyricum.
But Eutropius was, in fact, both more and less than that. He was less
because the vast prefecture of Illyricum, stretching from the Alps to the
Succi Pass in the Balkans, is unlikely ever to have been subject in its
entirety to Theodosius. The evidence here is famously equivocal. In July
 Gratian bracketed Illyricum with Italy when regulating the tariff for
the collatio lustralis, the tax on urban services; in  Theodosius was
issuing laws to the vicarius of Madeconia; in probably the same year,
two bishops from the diocese of Dacia took their case to Gratian, who
convened a council that met at Aquileia in September . In  Gratian
would explicitly demand that his prefect enforce a regulation “through all
Illyricum.” The fifth-century ecclesiastical author Sozomen is the sole
source to claim that the “Illyrians” as well as the east were assigned to
Theodosius; Orosius, however, has Gratian granting him the empire of
“the East and Thrace.” Although Sozomen is usually much the more
reliable, his notice here cannot safely be interpreted to mean a temporary
transfer of the whole prefecture. Certainty remains impossible, but the
most reasonable solution to the puzzle of the laws is to assume that there
was a temporary division of the large central prefecture of Illyricum, along

 Ambrose, ep. extra coll. .: quod catholicos ecclesiis reddidisti; cf. CTh ..: ut cunctis orthodoxis
episcopis catholicis ecclesiae . . . reddantur.
 Jones b: .
 I here restate, against the arguments of Burns (: –) and Errington (b: –), the position
advanced by Grumel (: –).
 CTh ...  CTh ...
 For the view that the council of Aquileia was organized following an interview between Palladius of
Ratiaria and Gratian in , see McLynn a: –; for an alternative chronology, see Errington
a: –. At the council and afterward, both Ambrose and Palladius treat the latter as a Westerner
(Acta conc. Aquil. , ; cf. Ambrose, Ep. extra coll. .): Palladius’s “Orientales episcopi” (Acta –)
must therefore be colleagues from the diocese of Oriens, not eastern Illyricum (as Errington :
–).
 CTh ...  Soz. HE ..; Orosius, Hist. adv. paganos .. (cf. Zosimus .).
 As Errington b: –. Sozomen is, to say the least, confused: he has Gratian returning to
territory that his father had “left” to him and his brother, after “bestowing” Illyricum and the east in
a single (presumably permanent) package. The most likely explanation is that he was misled by his
knowledge of Theodosius’s activities at Thessalonica: he records the emperor’s delight at learning
there that “all the Illyrians” did not share the Arian heresy (HE ..). Sozomen would be familiar
with the (eastern) prefecture of Illyricum, based on Thessalonica since  (Justinian, Novell. .;
cf. Theodoret, HE ..).
 Moments of truth: Gregory of Nazianzus and Theodosius I 

the line that would become an enduring boundary after . Theodosius’s
military operations would cover the two dioceses of Macedonia and Dacia,
so he required civil authority over these to ensure the smooth provisioning
of his armies; but there is no reason to suppose that Gratian, who showed
little inclination otherwise to help his colleague, made any further territorial
adjustment in his favor.
In terms of the scope of his authority, Eutropius was therefore one
of the least powerful fourth-century praetorian prefects on record. His
involvement in legislation, however, seems wholly disproportionate: he is
addressed in some thirty items preserved in the codes, comprehensively
outscoring his immediate colleagues in both east and west. These laws,
moreover, seem to have been calculated to boost the new emperor’s claims
to be a renovator of the empire. The reason for Eutropius’s legislative
prominence is that he was far more than merely a regional administra-
tor: he was to be assigned instead to Theodosius’s right hand. He was first
attested in January , when the emperor was wintering at the administra-
tive center of Thessalonica; he perhaps accompanied him on campaign
during the summer of  and in all probability then moved with him
to Constantinople that November. The sequence of laws addressed to
him continues unabated through the following summer. Although Con-
stantinople was in the diocese of Thrace and would normally be subject to
the jurisdiction of the prefect of Oriens, Theodosius’s appointee Neoterius
seems to have followed standard fourth-century practice and taken up resi-
dence at Antioch. Only after the emperor’s installation at Constantinople

 So Grumel ; Errington (:  [cf. b: –]) argues that the recreation of a separate
prefecture of Illyricum was a response to military emergency on the Danube; rather, the appointment
of the nonagenarian prefect Julius Ausonius should be seen as a reassertion of what had been the
norm before the reign of Valentinian I.
 Burns (: ) has Theodosius campaigning westwards from Sirmium, into Pannonia, based solely
on the dubious identification by Várady (:  n. ) between Vicus Augusti, where Theodosius
issued CTh .. in August , and the Pannonian “mansio Augusti” mentioned in a peculiar
entry in the Antonine Itinerary (Cuntz, Itineraria Romana :, .). Burns also argues (–)
that Theodosius exercised political control over the Sirmium mint – from the match between gold
ingots stamped at Sirmium and Theodosian issues from Naissus and Thessalonica – but, even if
these ingots are attributed to him rather than to Gratian, both emperors were based there in the
immediate aftermath of January .
 For the “Mini-code” of June , issued at Thessalonica, see Honoré : –. Cf. Matthews
: –.
 Eutropius’s movements in summer  are somewhat mysterious: the court that Symmachus,
Ep. . has him leaving Rome to “revisit” is presumably Gratian’s, since Palladius, the bearer, also
carried a commendation to the western courtier Syagrius: Ep. ..
 Norman (: ) argues that Neotarius’s direct patronage fueled the ambitions of the consularis
Carterius at Antioch (ambitions that foundered when he traveled to Constantinople to try to
implement them: Libanius, Or. .).
 neil m c lynn

did perspectives change sufficiently to create what became the definitive

Theodosian pattern, so that Neoterius’s successor, Florus, already serving
at court as magister officiorum, simply remained there; it seems a reasonable
surmise that he inherited staff from Eutropius’s office. Florus overlapped
in office with Eutropius for two months in , while Theodosius was
campaigning (presumably with one of the two in attendance); this brief
period might perhaps be treated as a rare example of a “joint prefecture.”
Eutropius’s situation is directly relevant to our understanding of the
January law. Despite its universalizing rhetoric, this was not, as has been
shrewdly pointed out, a command requiring all officials everywhere to
bestir themselves against heretics. The fact that four months later three
dozen bishops who explicitly refused to accept the theology prescribed in it
were still operating freely within easy reach of Constantinople demonstrates
this. But neither was it intended to provide a basis for purging the cities
of Eutropius’s Illyrican prefecture. The bishops of the region, at least,
seem blissfully unaware that this had ever been the emperor’s intention.
Theodosius issued his command in Constantinople, and Eutropius proba-
bly published it there: there is not the slightest hint that it was ever applied
in Illyricum or anywhere else.
The circumstances that elicited the letter to Eutropius are suggested
in the second sentence of the preserved extract, which proclaimed that
“if anything has been obtained by any sort of private rescript which has
been fraudulently elicited by this kind of people, it has no validity.” This
has been interpreted to suggest that Arian congregations in Illyricum had
been equipping themselves with documents to protect their property, prior
to an assault that they could already foresee. However, we should seek
a more straightforwardly reactive context than this doubly preemptive
scenario. One is available at Constantinople itself, where the many well-
connected homoeans could find opportunities to stake creative claims upon
 Theodosius issued legislation from Constantinople on  July and from Heraclea on  July (issuing
his first attested law to Florus as prefect there on  July); he was still at Hadrianople on  September
and had returned to Constantinople by  September (when Eutropius received his last attested
law). One infers that one prefect accompanied him, the other remained at Constantinople.
 Errington a: –.
 See below, at n. . The “Macedonianists,” who rejected the full consubstantiality of the spirit, were
not named in the January law but would fail the test of orthodoxy set out there.
 As Errington : –.
 The Acta of the council of Aquileia do not support the argument that proceedings there were
governed by the January law (Errington a: ): Ambrose offers Palladius a choice between
condemning Arius or proving him correct, without attaching any legal consequences if he preferred
the latter (Acta ). In seeking imperial enforcement of the council’s verdict, Ambrose makes no
reference to existing legislation: Ep. extra coll. .–.
 Errington a: – and n. ; cf. Errington : .
 Moments of truth: Gregory of Nazianzus and Theodosius I 

the ecclesiastical portfolio, as the new Nicene establishment struggled to

take stock of their windfall at the cathedral. The rescripts deplored by
Theodosius would be the most convenient instrument for doing this.
Some two decades later this would still be a confident and defiant com-
munity, engaged in acts of calculated provocation in the city streets against
their Nicene supplanters. The present law, for all its solemnity, would
not have cramped such activities, and it perhaps even helped create the
framework that enabled the homoeans’ regular processions to their extra-
mural church. Theodosius had reaffirmed the status quo created by the
appropriation of the cathedral: no urban property was to be flaunted by
dissidents as “ecclesiastical” (as Gregory had recently flaunted his house-
church of the “Anastasia”), but in the streets anything short of “factional
outbreak” (eruptio factiosa) remained permissible. This too, then, could
be treated as a “written law of persuasion”; Theodosius was learning the
symbolic effectiveness of such declarations of faith.
Here again Gregory helps provide contemporary context, for at about the
same time as Theodosius’s law, he would harangue the emperor on the need
for legislation against heretics. His sermon “On the Word of the Gospel:
‘When Jesus Finished These Words’” (Or. ) is, as the title implies, a
discussion of points arising from the day’s Gospel text; the text makes
it clear that the occasion was an ordinary Sunday, making this the only
conventional “sermon” among all Gregory’s surviving works. But Theo-
dosius was apparently present, to receive some home truths from his favored
churchman during the final movement of what is again a tripartite piece.
In the first and second acts Gregory had again indulged his preference
for imaginary dialogue. In the first, the boldest and most elaborate, he
takes on the injustice of the legal framework for marriage. Rejecting
the established nomothesia, the legalized discrimination that held women
to much stricter standards than men, he squares off against a latter-day
“Pharisee” (–), overwhelming the latter’s captiousness with a hymn to
the concentrated power, like water from a pipe, of which celibacy was
capable (–). His text for the day then invited him to turn – with
 For controversy over accounting procedures, see Gregory, DVS –.
 For an example of such a rescript (this time elicited by ultra-Nicenes), see Coll. Avell. a, with useful
commentary by Escribano Paño (: –).
 Socrates, HE . describes the torchlit processions suppressed by John Chrysostom.
 Other laws relating to Constantinople were addressed to praetorian prefects there: note CTh ..,
discussed in Cameron, Long, and Sherry : .
 Bernardi : –; Gómez Villegas : –.
 McGuckin (a: –) argues ingeniously but implausibly that it was “a discourse preached
before the emperor as part of a chancery meeting for legislative consultation.”
 Arjava : –.
 neil m c lynn

obvious relish – to the subject of eunuchs. He wanted to say something

“manly” on the subject, to forbid any false pride in enforced chastity ().
And he has a further commandment: no theological whoring (–). Here
Gregory points his finger at individuals present in his church, the staff
of the imperial bedchamber. The subject was evidently a congenial one.
Even after having declared it closed, he begins again: “Why have you made
impiety yours? Why do you all betake yourselves to the bad, so that being
called a eunuch is the same as being called a heretic? Join the men . . .”
When he stopped to ask the audience whether he should continue, he was
clearly confident of their response; but in resuming he switched to pane-
gyrical mode (). There were three kinds of eunuchs (); by the time he
has reached the third he is pointing instead at a role model of Christian
commitment, one of those “who have made themselves eunuchs for the
kingdom of heaven’s sake,” who despite lacking an orthodox mother, father,
priest, or bishop has been his own teacher and has performed the neces-
sary surgery himself (). This launches Gregory into the third and final
argument of his speech, for surgery was indeed necessary; physical passions
needed to be excised, as did also their spiritual equivalents, Arianism or
Sabellianism ().
Abruptly, Gregory has again become a legislator, prescribing these rules
both to laymen and to priests – and also “to those entrusted with govern-
ment.” Authority’s responsibility for dealing with murder, adultery, and
theft is acknowledged but subordinated to the task of “legislating piety.”
Gregory’s words carried less weight on the Trinity’s behalf than an impe-
rial command, “if you put a bridle upon those who are wicked, if you
assist those being persecuted, if you repress the murderers, if you pre-
vent murder.” By murder, Gregory explains, he meant killing not just
the body but also the soul – “for every sin is a soul’s death” (). So here
Gregory was suddenly addressing the emperor, in front of his congregation,
demanding legislation. But one is struck by how casual and perfunctory
the demand is. Gregory is neither building a case to the emperor for a
policy nor is he selling an imperial policy to the audience, as Themistius
was doing in speeches that probably still echoed in the ears of many of
the senators in Gregory’s audience. No sooner has he delivered his epi-
gram than he stops abruptly and moves to the solemnities of the closing
prayer ().

 It is not impossible that Themistius himself (with whom Gregory had corresponded over a decade
previously: Greg. Naz. Epp. , ) accompanied the emperor to church, to be found among the
bearded and cloaked “philosophers” whom Gregory invokes at Or. ..
 Moments of truth: Gregory of Nazianzus and Theodosius I 

It says much for Gregory’s virtuosity that he could paint as embattled

victims of persecution a congregation that included the emperor himself.
Even inside the cathedral of Constantinople, his was a voice crying from
the desert. But the sense of danger was not entirely artificial. Unrepentant
homoeans were present in the church, in their capacity as imperial atten-
dants; they remained a clear majority in the streets: in the circumstances,
anything better than an empty cathedral seemed achievement enough for
Gregory. The successive arguments from which Gregory constructed his
sermon might therefore be taken to indicate the room for maneuver that
was available to him – and also to Theodosius. The iniquities of the mar-
riage laws, for example, were a perennial favorite with Christian moralists,
but few were ever given the chance to confront an emperor with the
contradictions. For all Gregory’s vehemence, however, he carefully picks
his fight with those who exploited the laws rather than with those responsi-
ble for maintaining them. Theodosius is spared any awkward demands for
reform. With the eunuchs, too, despite their political significance (they are
prime suspects for any homoean success in eliciting favorable rescripts),
he confines himself to name-calling and never questions their right to a
place either in his own church or in the emperor’s service. It might be,
then, that the demand for legislation against heretics should be taken to
indicate the relative straightforwardness with which this could be imple-
mented. This was a demand to which the emperor could comfortably nod
his assent. Such an interpretation would suit the reading of the January law
proposed above. Theodosius was discovering, by a process of experiment,
that legislation against heresy was a relatively painless way of asserting
himself as the defender of “his” church.
On this view Theodosius and Gregory made an effective partnership,
with the bishop interpreting the emperor’s pronouncements shrewdly and
accurately. This should, in turn, encourage us to take his later comments
on the emperor seriously. But there remains a problem. In the year that
separated this sermon before Theodosius and the composition of De vita
sua, Gregory had found himself at odds with a large majority of the bishops
summoned to Constantinople by the emperor, and he had been driven
to resign his see amid much controversy. We must therefore examine
Theodosius’s role in the council and in Gregory’s resignation.

 Cf. above, n. .  For the “double standard” in Christian rhetoric, see Arjava : –.
 Theodosius would eventually dismiss some eunuchs who supported Eunomius: Philostorgius,
HE .. The purge is dated to / by Vaggione (: –).
 The singling out of “Photinian” heretics in CTh .. matches the rhetoric of Gregory’s sermon,
which balances the two opposing heresies of Arius and Photinus’s alleged master Sabellius.
 neil m c lynn

the council of constantinople

In modern studies of the council of Constantinople, Theodosius looms
large. He called the bishops to his capital; although there is no direct
evidence for his participation, he is credited with setting the agenda and
deciding the council’s goals. And before it had finished, he had accepted
Gregory’s resignation and had himself selected his successor – the first time
that a ruler had ever been so intimately involved in a senior ecclesiastical
appointment.
This version is derived from respectable ancient sources, principally
the fifth-century ecclesiastical historians. Socrates shows the emperor act-
ing immediately after Gregory’s departure and Demophilus’s expulsion to
“summon a synod of the bishops of his own faith, in order that he might
establish the Nicene Creed, and appoint a bishop of Constantinople” (HE
..); he also makes him personally responsible for inviting a contingent
of thirty-six “Macedonianists,” “hoping to unite these with his own party”
(..): the emperor, with his bishops, “used his utmost exertions to bring
Eleusius and his adherents into union” (..). Later, when the council had
compiled its canons, “the emperor too voted in favour of these” (..).
Sozomen adds details about the appointment of Nectarius, tracing this
directly to Theodosius’s personal choice of Gregory. “The emperor,” he
says, “deeply admiring Gregory’s life and eloquence, judged him worthy
of election to this bishopric” (HE ..); when Gregory declared himself
unavailable, “the emperor urged that diligent investigations be instituted,
so that the finest and best individual might be found, to whom the high-
priesthood of the great and royal city should be entrusted” (..). There
follows a detailed account of how Theodosius ordered a shortlist to be
brought to him and how he chose the praetor Nectarius from this, persist-
ing in his choice even when it was discovered that the candidate was not
yet baptized (.). Sozomen says more than Socrates about the emperor’s
confirmation of the council’s canons, exploiting his knowledge of the
Theodosian code to cite the law that established a list of named bishops as
guarantors of orthodoxy (..); the emperor, he adds, “having seen and
met all these bishops praised them, and a good report prevailed, that they
governed their respective churches piously” (..).

 Ritter : –; Hanson :  (on Theodosius’s “watching brief”); Gautier : , –;
Ayres : –; Errington a:  and : – (on Theodosius’s “long-term structural
agenda”).
 Errington a: –.
 Moments of truth: Gregory of Nazianzus and Theodosius I 

These fifth-century sources are in turn connected to fourth-century

texts. Sozomen’s law, for example, a letter to the proconsul of Asia, marks
a new level of imperial involvement in ecclesiastical administration. The
logos prosphonetikos that precedes the canons of the council, addressed
by the bishops collectively to the emperor to announce their decisions,
explicitly recalls the “letter of your piety” that had brought the authors
to Constantinople, and does so again when requesting another imperial
letter, “so that just as you have honored the church with your letter of
summons, you might also seal the end of our resolutions”; the canons
are then introduced as the work of “bishops who have come together
from different provinces according to the summons of the pious emperor
Theodosius.” The bishops at the council, then, affirmed their imperial
sponsorship in the strongest possible terms; other texts would also dwell on
the imperial summons. Moreover, in a letter the following year, the same
bishops would emphasize in a letter to a Roman council that Nectarius had
been consecrated “under the eyes” of Theodosius (Theodoret, HE ..).
However, Gregory’s De vita sua presents a significantly different perspec-
tive. Here the emperor remains in the background. “The authority” is duly
held responsible for summoning the bishops (by “I know not what words
from God” [DVS ] – a creative play, perhaps, on the sacral character
of documents emanating from the imperial chancery) and so creating “the
beginning of sorrows” for Gregory (); once they had convened, how-
ever, so far is the emperor from controlling proceedings that Gregory can
claim that anarchy reigned: discussions were run “by all, the same as to
say by none: for mass rule is no rule” (–). Gregory blames especially
the amphidoxoi, bishops whose beliefs were determined by “what pleases
authority” (), but he never suggests that authority had made its plea-
sure known. Only at the very end, when Gregory decides to resign from
the throne to which he had been appointed at the start of the council, does
he introduce Theodosius, when he runs to him for an interview ().
He dwells lovingly on his own comportment. “Did I bend? Did I bow?
Did I clasp his hand?” (). Scholars have conventionally sidestepped the
problem of reconciling this perspective with those in our other texts; but
Gregory’s presentation deserves to be taken seriously.
It has recently been argued that there is, in fact, a subdued polemic in
Gregory’s poem against what he considered government interference in the
theological discussions. The argument is ingenious but requires a forced
 CTh ..: see Errington b: –, –; cf. a: –.  Benesevic : –.
 John Chrysostom, De S. Meletio  (PG , ) mentions “letters from the emperor” twice.
 Gautier .
 neil m c lynn

reading of Gregory’s admittedly allusive Greek. The “proclamation in the

midst” () by whose specious orthodoxy he claims that some other-
wise right-thinking bishops were deceived has been identified as the law of
 January. This was the orthodox “child” of very different parents () –
on this interpretation the devious politicians who had helped the emperor
draft the law, imposing a theology of which Gregory heartily approved,
were now at work to impose a fatal compromise. If correct, this would
put imperial measures at the very heart of the council’s proceedings. Not
only, however, does Gregory never hint elsewhere that the emperor was sur-
rounded by theologically wicked advisers (although he occasionally criti-
cizes the imperial entourage), but here he proceeds directly to “the manifold
mob of Christ-pedlars” (), who from the context must be churchmen:
these provided the parentage for any deceptively acceptable pronounce-
ment. Moreover, Gregory immediately points a contrast between his ene-
mies’ “brazenness” and his own “foresight” (), which echoes exactly the
terms of the previous argument about Theodosius: he here aligns his own
approach with the emperor’s, showing two thoughtful men criticized by
the bullies.
We should therefore reexamine the imperial interventions attested in the
other sources to consider whether these justify the conventional interpre-
tation. The thirty-six Macedonianist bishops would seem to present the
clearest case, since Socrates presents the invitation as the emperor’s personal
initiative, and scholars have duly presented it as an expression of Theo-
dosius’s personal commitment to consensus. However, Socrates’s account
requires that all three parties suffered delusions: the emperor thought that
the two groups of bishops could be reconciled, the Nicenes thought that
the Macedonianists could be persuaded to subscribe to Nicea, and they in
turn believed that the creedal basis was negotiable. But this assumes that
the three parties were all equally committed to a negotiated solution, a dan-
gerously naı̈ve view given the contemporary evidence for the manipulation
of conciliar machinery. The most economical explanation in this case is
an initiative originating with the Nicene bishops once the council was in
session, which played to Theodosius’s much-advertised desire for consen-
sus. In issuing the invitation, the emperor was co-opted into an enterprise
that he indeed may have imagined as a means of reconciliation, but which

 Gautier (: ) strenuously absolves Theodosius himself.

 Ritter : –; Errington a: ; McGuckin a: –.
 Van Nuffelen (: –) nicely shows the implausibility of Socrates’s account; his conclusion
that the episode is fiction is nevertheless overstated. Socrates seems here to be working from the
documents produced by the Macedonianists after their return to their sees (HE ..): for his
interest in this sect and their activities, cf. HE .., .., ., ..
 Moments of truth: Gregory of Nazianzus and Theodosius I 

became in practice a formidable demonstration of the council’s resolve.

The historians record only an ultimatum. The Macedonianists were sim-
ply invited to sign the Nicene creed, and were reminded that their leaders
had done so in Rome fifteen years earlier. This looks suspiciously like an
ambush, confining the Macedonianists to an invidious choice between
tame submission and an appearance of intransigence. It was hardly in the
emperor’s interest to force such a choice upon them.
Only one contemporary source shows the emperor participating in the
bishops’ proceedings. When the influential coalition-builder Meletius of
Antioch died during the council, a lavish funeral was organized. Sozomen
reports that Theodosius himself organized the solemn cortege that carried
the body across Anatolia, city by city (suspending the usual prohibition
on bringing dead bodies into the cities), back to Antioch. The unprece-
dentedly lavish send-off from Constantinople is also noted by Gregory, but
again without any hint at the emperor’s involvement (DVS –). The
difference can be explained by reference to our fullest picture of the occa-
sion, a speech of Gregory of Nyssa, one in a series of memorial addresses
delivered as the body lay in state in Holy Apostles.
The enormity of the loss is vividly evoked: Meletius had been a bride-
groom to both Constantinople and Antioch, so two cities had been wid-
owed at a stroke; the tragedy was both a shipwreck and a decapitation.
Much of Gregory’s eloquence is presented as instruction to those who
would accompany the body on how to give their news at Antioch: how
to report the “glorious tale” of the instant cult that had grown up around
Meletius’s body at Constantinople, as “the people in their myriads, so
densely crowded together as to look like a sea of heads, became all one
continuous body, and like some watery flood surged around the procession
bearing his remains.” The eagerness of the whole people, Meletius’s lodging
in “the company of Apostles,” even the snatching by pious relic-hunters of
the wrappings that swathed his face – all excite the orator’s enthusiasm. He
then continues in distinctly more measured tones: “Let there be added to
your narration an emperor plunged into gloom over this misfortune, and
starting from his throne; and a whole city joining the procession for the
saint” (In Meletium ).
Gregory thus acknowledges Theodosius’s participation, but with only
a brief nod: one would never guess from this that the emperor was
 Sozomen, HE ...
 Greg. Nyss. Oratio funebris in Meletium episcopum, ed. A. Spira (: –). For the context, see
Ritter : –.
 Theodosius, we might infer, was not in the audience. His “starting from his throne” (In Meletium
) is presumably designed to recall such grieving biblical prototypes as the king of Nineveh at
Jonah :.
 neil m c lynn

himself sponsoring the journey on which Gregory was sending his pupils.
We might infer, given the orator’s political shrewdness (his next two great
funeral speeches at Constantinople would be for the emperor’s daughter
and wife), that Theodosius had refrained from making any overt exhibition
of personal grief. This, in turn, betokens good sense on the emperor’s part.
He cannot have been unaware of the tensions concerning the succession
at Antioch, and any demonstrative gestures were liable to be translated
into commitment to the faction that controlled the body. The formidable
apparatus of the state transport system helped ensure that Meletius arrived
home as a saint; but at Constantinople the emperor did not throw his
political weight behind the dead bishop’s partisans in their struggle to
secure the succession.
The emperor seems also to have kept a safe distance from the bitter
dispute that erupted after Meletius’s death between these partisans and
Gregory, who promoted the candidacy of Meletius’s old rival Paulinus.
Decisive in resolving this confrontation was the arrival of bishops from
Macedonia and Egypt, “suddenly summoned” (DVS ) by an invitation
that must have come formally from Theodosius, although this time Gregory
does not say so. This is significant. Since their arrival triggered his own
resignation, he had every reason to establish responsibility and no cause
to disguise any imperial initiative. Here again, the emperor was probably
following advice, most likely from Gregory himself, as chairman of the
council. He could not have foreseen that the newcomers would prefer a
cheap scalp (his own) to a settlement of the Antiochene schism on their
own terms; nor is there any reason to suppose that the emperor engineered
or particularly welcomed this outcome.
From this perspective, Theodosius’s sudden prominence at the end
of Gregory’s account assumes fresh significance. Gregory has already
announced his resignation, pronounced his farewell to the bishops, and
received the dubious honor of their ready assent: he could easily have
ended his narrative there. Instead he introduces another topic: “But how
was it with Authority?” (DVS ). He answers himself with a string of
further questions, covering all that he did not do: bend or bow or clasp
 McGuckin (a: ) states without argument that “the emperor seems to have acted decisively,
and on his own counsel.” Cf. Gautier : ; Ayres :  (“according to Gregory these new
arrivals came at the behest of Theodosius”); Ritter : –. Errington (a:  n. ) argues
that all the invitations had been issued together, and that the newcomers had simply been delayed,
but this distorts Gregory’s language.
 Note that Ambrose, writing to Theodosius, would attribute the invitation of Acholius to the bishops
(Ep. extra coll. .).
 Errington (: ), argues that “Theodosius must have been grateful” for the opportunity to
remove someone who “had failed his first test” as bishop of Constantinople; Gautier (: –)
has Theodosius actively plotting Gregory’s removal.
 Moments of truth: Gregory of Nazianzus and Theodosius I 

the emperor’s hand, deliver suppliant speeches, send eminent friends to

intercede, or bribe his way to security of tenure (–). Such recourses
he left to others, to the time-serving “serial turncoats” ().
In these seven questions, clearly intended to resonate, Gregory conjures
with the mystique of access to the imperial ear. He can count on his read-
ers’ imaginative engagement with this scene: an imperial audience was the
ultimate late-antique moment of truth. And Gregory continues by describ-
ing his conduct at the interview, emphasizing the many witnesses present
(), whose testimony would be available to readers at Constantinople.
Again he rehearses, this time in a speech addressed to the emperor, the
many favors he could have sought but did not: gold or altar cloths or offi-
cial posts for relatives. His sole petition was instead to be permitted to “yield
a little to envy.” This gambit has attracted little comment. The emphasis
that Gregory places on it should require us to take it seriously: had there
been even any suspicion that he had merely been attempting, and failing, to
secure an imperial vote of confidence, as has been suggested, he would have
had to counter this, as he does other potential counterarguments elsewhere
in the poem. Instead he reports his solemn proposal (and the dramatic
gestures with which he illustrated this [–]) that Theodosius should
intervene to restore harmony to a faction-ridden church, while he himself
be allowed to continue “to suffer for the sake of the world.” If (as there is
no reason to doubt) this was the gist of his actual speech, it will have served
to redefine the situation; by publicly applauding these words (), the
emperor was validating Gregory’s own construction of his resignation, and
so allowing himself to be drawn into the factional politics of the council.
Not without reason, then, did Gregory make this explicit endorsement of
his own position the climax of his enormous poem a year after the event.
He makes so much of the interview, not because he was overwhelmed by
the glamour of the purple, but because he was able to harness it to his own
cause. His connection with Theodosius provided the basis for his claim
to the moral high ground against the bishops who had dominated the
council.
It is even possible that Gregory’s initiative pushed Theodosius into more
direct involvement with the council’s proceedings than he would otherwise
have accepted. Having publicly pledged support to Gregory’s peace-making
mission, the emperor could not easily remain detached from the appoint-
ment of his successor, an appointment that inevitably reanimated the same
“envious” factions from which he had now agreed to demand “loving

 Errington (: ) portrays, without further discussion, Gregory’s “surprise and horror” when
Theodosius accepted his resignation.
 neil m c lynn

harmony.” The arrangement described by Sozomen, whereby Theodosius

reserved the right to select a name from the lists of candidates submitted
by different bishops, might be seen as a mechanism to prevent any rep-
etition of the partisan “screeching” that had so disfigured the discussion
of the candidates for Antioch (DVS –). The result, of course, was
the choice of an unbaptized layman. Yet for all the scorn that Gregory
heaps on Nectarius in a series of bitter poems and ironical letters, he never
once suggests that he was either the tool of an ecclesiastical policy or an
imperially controlled puppet.
Theodosius did, of course, have an ecclesiastical agenda. Of course he
wanted “orthodoxy”; of course he wanted unity. And without doubt he also
wanted the eastern bishops to look to him, and him alone, as the referee of
their disputes. But this did not constrain him within any narrow ideological
framework. Gregory’s poem instead suggests that he succeeded in remain-
ing beyond the reach of any doctrinal party, for Gregory finds neither cause
to demonize him (or to treat him with the indignation he showed toward
even friends who failed to live up to the standards he had set for them)
nor grounds to take his support for granted. The claims he makes upon
him are instead carefully limited and cleverly argued. His dispassionate
summing-up is therefore to be taken as a measure of Theodosius’s success
in eluding capture. The surprising range of churchmen who looked to the
emperor, and would continue to look to him, for a sympathetic hearing –
not just the ultra-Nicene schismatics who in / elicited a letter to a
praetorian prefect where Theodosius trumpeted their cause just as loudly
as he had the mainstream Nicenes’ in , but also dispossessed homoeans
and hopeful Eunomians – should already have suggested the inadequacy
of any stereotype of narrow, “Spanish” orthodoxy. Theodosius (unlike so
many previous emperors) emerges Teflon-coated from the toxic quarrels of
his churchmen; not the least value for historians of Gregory’s account of
their dealings is to help explain how he achieved this.
In addition, Gregory’s presentation of the emperor can usefully be set
beside the “lies” that Themistius was peddling about him to much the
same audience at much the same time. Peter Heather’s suggestions that
Themistius was involved in a comprehensive rebranding exercise assume
 On Gregory’s invective against Nectarius, see McGuckin b.
 For the granting of the Luciferian petition, see above, n. ; the hopes of the homeans when
invited to Constantinople “praecepto imperiali” in  are reflected in Auxentius, Epistula de fide,
vita et obitu Ulfilae  (in Gryson ); Sozomen HE .. has the empress intervening to prevent
Theodosius from inviting Eunomius to an interview.
 Even the Eunomian historian Philostorgius, whose Nero was Gratian, would find some redeeming
features in Theodosius (HE .): cf. Marasco : .
 Moments of truth: Gregory of Nazianzus and Theodosius I 

further significance when we compare the panegyrist’s homage to imperial

pacifism to Gregory’s use of the same repertoire: Themistius’s Theodosius,
like Gregory’s, was distinguished by his gentleness; he “raised a trophy to
Persuasion.” Just as the Themistian echo lends force to Gregory’s appro-
priation of the emperor, so Gregory’s overtly apologetic portrait lends
force to Heather’s point about the competition that Themistius faced in
developing a quasi-philosophical image of the emperor that remained as
provisional as the Gothic policy of which it was an indirect reflection.
Themistius and Gregory thus each seized his opportunity to reshape a
compliant emperor in an image congenial to himself, but neither could
expect his version to remain authoritative. Hence, perhaps, the careful
distancing by which Gregory scores his points in , his deliberate prob-
lematizing of the emperor’s style of Nicene commitment. Themistius had
identified himself much more closely with the regime, and his protestations
in / about the authenticity of Theodosius’s philosophical commitment
sound correspondingly shrill.
Nor were these by any means the only games in town. While Gregory
and Themistius, and their respective rivals, were battling over these facets
of the emperor’s public image, Jerome was skirmishing on the fringes of
the new regime, marking its beginning with his end and hinting, as Mark
Vessey explains in a paper that should be read in conjunction with this and
with Peter Heather’s, that more might be forthcoming. Such maneuvers
provide the cultural context within which the emperor and his entourage
made themselves at home in their new capital and suggest why they found
the environment congenial. Theodosius’s bland responsiveness, in turn,
helps explain why Gregory, when thrown overboard at the council of ,
should choose to blame his fellow bishops and not his emperor. And
Theodosius would subsequently find many other Christian impresarios
(including several of Gregory’s close allies) to offer him his cues and enhance
his initiatives. Through such interactions would be constructed the cycles of
Christian ceremony that Brian Croke will discuss in the following chapter.

 Them. Or. .a (gentleness); .c (trophy).

 Or. .–, –. For commentary, see Heather and Moncur : –.
 chapter 12

Reinventing Constantinople: Theodosius I’s

imprint on the imperial city
Brian Croke

Constantinople was established and embellished as a new Roman imperial

city by the emperor Constantine I in the s and s. He provided “New
Rome,” as he called it, with a solid defensive wall stretching from the Sea
of Marmara to the Golden Horn, inside which he created a splendid and
spacious new metropolis with an oval-shaped forum, an imperial palace,
baths, and porticoed thoroughfares plus a local senate. His efforts to
consolidate and promote his new city met with mixed success and over
the succeeding decades its status and fortunes advanced only haltingly.
Rather than become a new imperial home as Constantine had intended,
Constantinople was treated by his successors more as a transit camp as
they progressed back and forth from Gaul and Italy to the eastern limits
of the empire in Mesopotamia and its northern limits on the Danube.
Between the death of Constantine in  and the accession of Theodosius I
in  emperors spent an average of less than one month per year in
the city. Arguably the longest single stretch was from September  to
May , the duration of the unsuccessful usurpation of Procopius, who
was proclaimed emperor at Constantinople by capitalizing on his family
connection with the city’s founder and on local eagerness for a resident
ruler. It is one of the many fourth-century episodes that John Matthews
was the first to penetrate and elucidate effectively.
Procopius’s usurpation nearly terminated the reign of Valens. This
explains why the victor remained suspicious of Constantinople and unfor-
giving. Valens went on to pass most of his fourteen imperial years at
Antioch, but in  he threatened to plough Constantinople into the
ground on his return from trouncing the Goths . Instead, Valens died in
battle with the Goths on  August and the city was spared his wrath. By
 Fundamental are Dagron , Mango , and Berger . On the development of the senate in
particular: Heather  and ; Vanderspoel : –; Errington : –.
 Based on calculations available in Dagron : –.
 Matthews : –; Lenski : –.  Socrates, Hist. eccl. ...


 brian croke

contrast, his successor Theodosius I spent the vast majority of his reign
in and around Constantinople from the day he first entered the imperial
capital on  November  to August , when he left it for what turned
out to be the last time. In doing so he set the pattern for generations
of Byzantine emperors who succeeded him. This elementary fact is not
so evident, however, in most modern accounts of Theodosius, where the
prevailing picture is that of an emperor forever on the move between east
and west, in between which his legislation marks a decisive advancement
for Christians and defeat for pagans everywhere. Modern accounts tend
to concentrate on his years at Milan (–), highlighted by his contests
with bishop Ambrose, plus his military encounters with barbarians and
with the usurpers Maximus (–) and Eugenius (–). One of their
overriding historiographical preoccupations is determining whether or not
Theodosius deserves the soubriquet “Theodosius the Great.”
Except for an original and illuminating chapter of John Matthews’s West-
ern Aristocracies (: –) focused on the emperor’s western courtiers
in an eastern environment and their influence on religious life, too lit-
tle attention has been paid to Theodosian Constantinople and the key
role played by Theodosius in transforming the city into an imperial and
Christian capital. What has obscured the centrality of Constantinople to
the reign of Theodosius is an undue dependence on balancing the hostile
contemporary account of Eunapius with the fifth-century church histo-
rians (Philostorgius, Socrates, Sozomen, Theodoret) who wrote from the
perspective of the established Christian empire in which Theodosius is
assigned a key part. Other factors are reliance on purely western sources
such as Orosius and Rufinus, as well as the habit of treating “Theodosian
Constantinople” as a single period from  to  (as do Janin 
and Bassett , for example), thereby failing to distinguish properly
the very different contributions to the city’s growth of Theodosius I, his
son Arcadius, and his grandson Theodosius II. Evaluating the impact of
Theodosius I on Constantinople requires devoting greater attention than
 Theodosius was emperor for sixteen years (less two days), or  months. Of these, he can definitely
be assigned to Constantinople for a total of  months ( percent) or an average of over six months
per year – but probably more if documentation permitted greater precision.
 Typically, Williams and Friell .
 The quest began with Stein : – and Jones a: , then led to dedicated chapters in
Lippold : – and Leppin : –. Note also Williams and Friell :  and Ernesti
: –.
 Recent notable exceptions are Leppin : – (but focused almost entirely on Theodosius’s
building activity), Errington : – (focused on administrative and legal aspects); and previ-
ously Dagron : –.
 Cf. Ernesti : –.  Brown : –; Leppin : –; Errington b.
 Reinventing Constantinople 

previously given to the ceremonial and monumental aspects of the city

from . Doing so sharpens our appreciation of how pivotal the city was
to his whole reign. It also highlights how sudden and significant these years
were in shaping the urban contours of Byzantine court and civic life for
centuries to come.

theodosius meets constantinople

Theodosius was an experienced general and the son of a general. He was
proclaimed at Sirmium in January  but had scarcely had time to assem-
ble his own imperial court when he was forced to set out against the
Goths for the campaign season of that year. It was a hard fought and
not overwhelmingly successful operation for the Romans. With hostilities
ended, Theodosius and the imperial retinue wintered at Thessalonica, dur-
ing which the emperor fell dangerously ill and was baptized. Mindful of
Valens’s recent attitude and fearful that the new emperor from Spain might
choose to make Rome-oriented Thessalonica his permanent capital, the
senate of Constantinople sent an urgent delegation to persuade Theodos-
ius to come to them instead. The senate’s principal advocate was its most
distinguished member, the philosopher and rhetorician Themistius, who
was an experienced petitioner and confidant of emperors. The argument
he put to Theodosius was that “the great city” should “receive her protector
as soon as possible and meet him before the rest of the east, and . . . that all
the gifts which your forefathers decreed might remain secure for her.” Urg-
ing Theodosius to consider increasing the senate’s influence and prestige
as his key task, Themistius concluded: “If you, o divine eminence, should
dedicate such prizes of victory to the great senate, then truly shall your city
be a second Rome.”
Despite a further unsuccessful encounter with the Goths in  Theodos-
ius made his triumphal entrance into the “second Rome” on  November
. It was probably the first time he had ever seen the city and he would
have been welcomed in style, although to the locals he was a virtual “blank
page.” The imperial adventus would have involved a formal reception
outside the city walls that was then repeated inside. The prefect of the city,
Restitutus, would have led the local delegation, followed by acclamations

 Errington a; Leppin : –.  Errington a: ; : –.
 Them. Or. . (b–a, Maisano –) = Heather and Moncur : –.
 McCormick : ; Errington b: .
 The striking phrase of Leppin (: ); cf. Errington a: .
 brian croke

and formal panegyrics by the city’s most renowned orators. Themistius

must have been foremost among the group, although no such oration
survives. Likewise, the praetorian prefect of the east, Neoterius, appears to
have been there too.
Theodosius had spent most of his adult life on campaign every summer,
living with his soldiers and wintering in nearby camps and towns. Now
he was taking possession of a relatively new and marvellous metropolis
but as an already baptized Christian. Martial emperors may not yet have
spent sufficient time at Constantinople to establish a ceremonial, cultural,
and administrative dominance over the city, but it was always an imperial
capital. As early as January , however, it appears that Theodosius had
deliberately shifted the focus of imperial life and purpose from military
exploits where he had become increasingly unsuccessful to civic and urban
development, beginning with the palace overlooking the Bosporus which
had always needed to be maintained. The emperor, his army and courtiers
could require the palace at short notice, especially a new emperor created
locally, as Procopius was in . When Theodosius had first settled into
the palace, he evidently found that some of the staff had outlived their
usefulness so had them pensioned off, but he was later forced to protect
them from local financial obligations.
Theodosius arrived in the city accompanied by his family, including
his wife, Flaccilla, and his infant son Arcadius; a large retinue of imperial
guards; and a group of essential courtiers such as Florus, the magister
officiorum who was to succeed Neoterius as praetorian prefect of the east
within a few weeks. His entourage also included a significant posse of
relatives and family friends, in fact “an entire clan moved in with him to
dominate the court life of Constantinople,” as Matthews put it. Once
there he would soon have encountered the remnants of earlier imperial
households, including the surviving wives of former emperors, namely,
Domnica, the spouse of Valens; Charito, the spouse of Jovian; and perhaps
even Procopius’s wife, Artemisia, who had been reduced to undignified and
sightless penury. Presumably their imperial children were still alive too.
 Restitutus (PLRE I, s.v. “Restitutus” []) is most likely to have been prefect at the time since
his successor, Pancratius (first attested as prefect on  July ), was still comes rei privatae on
 September  (PLRE I, s.v. “Pancratius” []) and possibly arrived in Constantinople with
Theodosius.
 PLRE I, s.v. “Flavius Neoterius.”
 Proclaimed by Themistius (Or. ) but explained by Heather (above, pp. –).
 In  Libanius hoped that Theodosius and his court might still come to the home of his predecessor
at Antioch (Or. .–). Presumably the imperial palace there was still being kept ready.
 CTh .. ( February ).  PLRE I, s.v. “Florus” ().  Matthews : .
 Jo. Chrys. ad vid.  (PG .), with PLRE I, s.vv. “Charito,” “Domnica,” “Artemisia.”
 Reinventing Constantinople 

Carosa and Anastasia were the daughters of Valens and Domnica, while
Varronianus, the son of Jovian and Charito, had been an infant consul in
 and was still only a young man in . Charito and Varronianus lived in
fear of their lives, perhaps from those who saw him as a potential usurper,
like Procopius, especially since he had once been a designated imperial heir
(nobilissimus). Otherwise, the city’s memory of court and imperial life
was selective and patchy.

constantinople meets theodosius

The Constantinople that Theodosius first encountered was bustling and
boisterous. It was also disjointed and discordant. Not long before Theo-
dosius set foot in the city, Gregory of Nazianzus, who had been there for
over a year guiding the small and alienated Nicene community, lauded
the city for having “walls and theater and racecourses and palaces, and
beautiful great porticoes and that marvellous work the underground and
overhead river [aqueduct of Valens] and the splendid and admired column
[of Constantine] and the crowded marketplace and a restless people and a
famous senate of highborn men.” At the same time the city was a cultural
and intellectual center that attracted distinguished teachers and ambitious
students such as Jerome. The populace, estimated at around , at
this time, was fed by the fertile fields of Egypt with grain transported to
Constantinople every year, stored in large warehouses, and carefully dis-
tributed, so Gregory could greet the advancing grain fleet in  as a most
pleasant sight. Just before that, the Goths who approached the city were
awestruck by its size and beauty, the housing blocks they could see jutting
out above the walls, and the evidently dense population. In fact, shortly
after his own arrival Theodosius welcomed the Gothic king, Athanaric, and
just a few days later he was leading the king’s funeral procession through

 PLRE I, s.vv. “Anastasia” (), “Carosa.”

 Philostorgius, Hist. eccl. ., also hinted at by Themistius addressing Theodosius in  (Or. .,
c [Maisano –] = Heather and Moncur : ).
 Them. Or. ..
 Jerome was attracted to Constantinople by Gregory of Nazianzus (J. Kelly : –), who
considered the city to be “distinguished by the eminence of its rhetorical and philosophical teachers”
(Or. .) when his friend and fellow-student Basil went there, just as it was adorned earlier by
pagan luminaries such as Maximus (Eun. VS  [Wright –]) and Libanius (Eun. VS 
[Wright –]).
 Them. Or. .. Eunapius, on the other hand, considered this a symptom of the pampered
rapaciousness of Constantine’s city (Eun. VS  [Wright –]).
 Amm. Marc. Res gestae ...
 brian croke

the city. On encountering Constantinople Athanaric exclaimed: “I have

seen what I have often heard of though I did not believe it.”
Within days of Theodosius’s entry Gregory noted the crowded urban
environment: “The market places were full, the colonnades, streets, every
place, two and three storey houses were full of people leaning out, men,
women, children, the very aged.” It was a similar scene evoked not long
after by Gregory of Nyssa in his snapshot of the funeral procession at
Constantinople in mid- for Bishop Meletius of Antioch. He describes
how “the people in their myriads, so densely crowded together as to look
like a sea of heads, became all one continuous body, and like some watery
flood surged around the procession bearing his remains.” As the procession
jostled along into the darkness, you could see “how the streams of fire,
from the succession of lamps, flowed along the unbroken track of light,
and extended so far that the eye could not reach them.” Then Meletius’s
body was transported overland to Antioch but Theodosius had requested,
contrary to custom, that his casket be solemnly received in every city,
like the arrival of a governor or emperor, and accompanied by singing of
psalms just as the traditional civic adventus involved laudatory choirs.
Almost instantly the Spanish emperor had created a new ceremonial model
by fusing secular and religious practice at Constantinople.
The hippodrome and theater were major attractions for the people of
late-fourth-century Constantinople. News of any imperial victory, however
marginal, was celebrated in the hippodrome. The announcement of impe-
rial military success on  November , for instance, involved races, and
the victory spectacle was repeated in . Doubtless one of Theodosius’s
first public displays was a triumph in the hippodrome. Other celebratory
occasions were more measured. The birthday of Constantine’s city was
commemorated on  May each year with a colorful cavalcade through the
main street (Mese) and around the hippodrome. The central feature was
a statue of Constantine carefully mounted in a chariot, which progressed
with attendants carrying candles and dressed in short white cloaks. They
entered the hippodrome through the starting gates and carried the statue to
the Stama, opposite the normally empty imperial box, where the emperor’s
representation could be acknowledged and could then preside symbolically

 Jord. Get. .  Greg. Naz. Poems ..– = White : .
 Greg. Nyss. In Meletium ( Spira).  Sozomen, Hist. eccl. .., with Geyssen : –.
 Marcell. . (MGH.AA, XI ), Chron. Const. . (= MGH.AA, IX ) with McCormick
: –.
 Chron.Const. . (= MGH.AA, IX ).
 Reinventing Constantinople 

over the anniversary races. Yet, such displays of civic solidarity and unity
at Constantinople masked the fact that the locals were still scarred by fear
of the Goths. War widows were still in mourning.
Above all, the Constantinople that greeted Theodosius was deeply fis-
sured by competing religious allegiances, whose passionate tone is captured
in Gregory of Nyssa’s contemporary observation that when asking the price
of some goods or even the price of bread one is likely to get into an argu-
ment about whether the son is “begotten or unbegotten,” or is of the same
substance as the father or not. This religious ferment had been exac-
erbated by the law the emperor had issued at Thessalonica the previous
winter in which he prescribed the Orthodox Catholic religion as being sup-
port for the doctrinal tenets of the bishops of Rome and Alexandria. The
Arians, or homoeans, used to enjoying imperial support under Constantius
and Valens, were dominant, and they held the city’s main churches, those
of the recently consecrated Holy Apostles and Saint Eirene in particular.
Then there were the Apollinarians as well as the followers of Macedo-
nius, who held their own church, the adherents of the strict Novatian,
and Eunomius’s supporters, known as “Anomoeans,” who gathered with
him in various welcoming mansions. The orthodox Nicene congregation
nurtured by Gregory of Nazianzus was centered on the chapel of Anastasia
in the portico of Domninus. The community was not huge and the chapel
was just a large reception room in the mansion of Constantine’s praeto-
rian prefect, Ablabius. It was around this time that the Western pilgrim
Egeria passed through the city, where she reported that “when I had arrived
there, I went through all the churches – that of the Apostles and all the
martyr-memorials, of which there are very many.”
Faced with this plethora of church practice and belief that had produced
a range of competing congregations centered on particular churches and a
deeply factionalized community, Theodosius’s instinct was to bring them all
together and let them find their common ground. Theodosius was himself
prepared to listen to what the local religious leaders and disputants had to
say, which encouraged palace officials to believe they could dissuade the
emperor from his orthodox position. Further, his doctrinally resolute wife,
Flaccilla, insisted that he keep away from Eunomius in case the emperor
was unduly swayed by him. No less disconcerting to the empress was
 Jo. Mal. Chron. (Dindorf .–.); Paras.  (Preger .–) and  (Preger .–); Patria
. (Preger .–), with Bauer : –.
 Jo. Chrys. Ad uiduam – (PG .–).
 Greg. Nyss. De deitate filii et spiritus sancti (PG .–).  CTh .. ( February ).
 Socrates, Hist. eccl. ..; Sozomen, Hist. eccl. ...  Van Dam : –.
 Peregrinatio Aetheriae, .  Sozomen, Hist. eccl. ...
 brian croke

perhaps his preparedness to allow the homoean bishop Demophilus to

repent his heresy. The homoeans and Demophilus were ousted from
their preeminent position in November , but it took a highly charged
demonstration of brute force to install Gregory as his replacement in the
Church of the Holy Apostles. By May  all the bishops of the east had
assembled at Constantinople at Theodosius’s invitation, and in the Church
of Saint Eirene they produced an immediate settlement, although further
synods had to be held in  and  before a new doctrinal equilibrium
emerged, culminating in a formal hearing before the emperor in the palace
in which he considered the written depositions of the bishops of each
party before opting for the Nicenes. Even then the situation was delicate,
and a backlash always lingered just below the surface. The false rumor of
Theodosius’s defeat by Maximus in  was enough to incite Arians to
burn down the house of the bishop Nectarius. One of the momentous
decisions of the council of  (canon ), however, was the pronouncement
that Constantinople should hereafter take ecclesiastical precedence after
Rome because it was the “New Rome.”
With the emperor and his court established at “New Rome,” at least for
the time being, the orthodox community consolidated its position despite
the religious indifference of most imperial officials and palace staff as well
as the need to dispense with the services of those who insisted on adhering
to their support for Eunomius. They achieved this by a doctrinal clearing
of the urban liturgical space (churches, squares, streets) and by expanding
their control over it. The key churches were now in orthodox hands and
heretics were relegated beyond the city and their ministers were even
flushed out of hiding places. Then the local orthodox liturgical forms and
calendar expanded to fill the space but were linked to the life and fortunes
of the city as a whole. By the time Theodosius arrived in Constantinople,
the city was used to night vigils and processions through the streets as
well as communal singing and chanting on the great feasts, but they were
dominated by the homoeans. Now the orthodox appropriated them.
The so-called stational liturgy, developed at Jerusalem and imported to
Rome, took on a distinctive imperial character at Constantinople as it
centered around local events and new local spaces such as the Forum of
 As argued by McLynn (above, pp. –).  Greg. Naz. Poems ..– = White : .
 Socrates, Hist. eccl. ..–. Bishops such as Amphilocius of Iconium clearly had privileged access
to the imperial palace under Theodosius (Sozomen, Hist. eccl. ..–; Theodoret, Hist. eccl. .).
 Sozomen, Hist. eccl. ...  Errington : .
 A particular complaint of Gregory of Nazianzus (Or. ., cf. .–).
 Philostorgius, Hist. eccl. ..  CTh .., cf. .. ( December ).
 Ibid. .. ( January ).  Baldovin : , .
 Reinventing Constantinople 

Constantine. So too, many key festal days were already part of the annual
rhythm of urban life. Apart from Easter, there was Epiphany on  January,
and on  December  Theodosius witnessed the first celebration of
Christmas at Constantinople. Certain saints’ days were also now part of
the local calendar: Cyprian, for example, and even Athanasius, as well as the
local Constantinopolitan martyrs Acacius ( May) and Mocius ( May).
Other local anniversaries had also begun to be celebrated: Constantine
and Helena on  May, the dedication of the Church of Saint Menas
on  September, the transfer of relics of Andrew, Luke, and Timothy
to Constantinople and their deposition in the imperial Church of the
Holy Apostles, and the military martyr Theodore on the first Saturday
in Lent. Indeed, the only extant sermon of Bishop Nectarius (possibly
mid-s) was delivered on the feast of Theodore and mentions that the
annual commemoration is now well established at Constantinople. To
these feasts was added immediately a commemoration on  August of the
local Council of . All these new annual feasts still survived in the
eleventh-century Byzantine liturgical calendar, and the number of such
local feasts quickly grew under Theodosius.

imperial ritual and ceremonial

The presence of the emperor and his court also refocused imperial ritual and
ceremonial in and around the palace itself, where to date it had only been
witnessed spasmodically. Immediately, the established rite of the statue of
Constantine being honored in the hippodrome on Constantinople’s foun-
dation anniversary on  May was halted. It could now be replaced with
a real flesh-and-blood emperor keen to stamp his own claim on the city.
Major imperial milestones called for special celebrations. Imperial births
had been a relatively rare event in recent decades. At Constantinople, how-
ever, Theodosius fathered five children born into the purple: Honorius

 Ibid. –, , .  Greg. Naz. Or.  (Christmas); – (Epiphany ).
 Ibid. . (Cyprian, on  October ); Or. . (Athanasius, on  May ); Typicon CP, Mateos
–, vol. : .– (Acacius), .– (Mocius).
 SEC .–.
 Typicon CP,  September (Mateos –, vol. : .–). For the church itself: Janin : –.
 Details in Burgess : –.
 Typicon CP (Mateos –, vol. : .–), also on  February (Mateos –, vol. : .–).
The annual liturgy was celebrated in the church of Saint Theodore in the quarter ta Sphorakiou
(Janin : –).
 Nectarius, De festo S. Theodori (PG .D).
 Typicon CP,  August (Mateos –, vol. : .– [H]).
 Paras.  (Preger .), Patria . (Preger .).
 brian croke

(), Pulcheria (), Gratian (), Galla Placidia (), and John
(). In later Byzantine times such imperial births triggered several days
of ceremony inside the imperial palace and in the hippodrome. Having
assembled fifty members of each faction on the fourth day after the birth,
the imperial chamberlain (praepositus) would then address them as follows:
“Our sacred emperor requests that, in accord with standing traditional
custom and ancient practice you assemble tomorrow and proclaim the
name of the infant born in the purple.” It is possible that this “ancient
practice” was initiated or took its essential Byzantine shape in the s
and s. In any event, each year from the elevation of Arcadius in 
to  the imperial birthdays of both Theodosius (born  January) and
Arcadius (date unknown) were celebrated. In  and , after Honorius
(born  September) had also become emperor, there were three imperial
birthdays, although the only one Honorius actually celebrated at Con-
stantinople was on Friday,  September . Theodosius decreed in 
the annual celebration of both the emperor’s birthday and his dies imperii,
the anniversary of his accession on  January  (plus that of Arcadius
on the same day after , and Honorius on  January in  and ).
He was later forced to declare that Sunday was sacred and a day the hippo-
drome should be closed, but not if the emperor’s birthday fell on a Sunday.
Theodosius was obviously anticipating his next birthday on Sunday,  Jan-
uary . To judge from earlier and later examples, each of these events
was accompanied by games and celebrations in the hippodrome and in
other public spaces. In later Byzantine times the centerpiece of the memo-
rial day was a palatial banquet, which may already have been the case in
Theodosius’s time. Imperial anniversaries also gave rise to orations such
as those Themistius delivered on  January  (Or. ) and  January 
(Or. ). The sheer regularity of these celebrations in the s and early
s rapidly consolidated the accompanying ritual. Thus was invented
what became the Byzantine ceremonial for imperial births and their annual
celebration.
Proclaiming a new emperor was one of the most significant imperial
events. In  at the tribunal at the Hebdomon, on the fifth anniver-
sary of his father’s proclamation on  January, Arcadius was crowned

 PLRE I, s.v. “Fl. Honorius” ().  PLRE I, s.v. “Pulcheria.”

 Not in PLRE I, but see Rebenich .  PLRE II, s.v. “Aelia Galla Placidia” ().
 Not in PLRE I; cf. Rebenich : .  Const. Porph. de caer. . (Reiske .–.).
 CTh .. ( August ).  Ibid. .. ( April ).
 Const. Porph. de caer. . (Reiske .–.).
 Following the date of Errington : –.  Moffatt .  Kent : .
 Reinventing Constantinople 

Augustus. Ten years later, Theodosius’s next eldest son Honorius was
proclaimed there on  January. In addition, by the time the fifteen-
month-old Honorius entered his consulship in , he already held the
rare but official title of nobilissimus puer, which meant he was an emperor-
designate. The title must have been conferred on him sometime in  at a
palace ceremony similar to that recorded for the title in later times. The
empress Flaccilla was also elevated with even greater pomp to the no less
rare title of Augusta in , so she could have her own coinage and honorary
statues, on both of which she was represented with the accoutrement of
an emperor. In recent decades the traditional Roman ritual of proclama-
tion had been witnessed at widely scattered points including Paris (Julian),
on the Persian frontier (Jovian), at Nicaea (Valentinian), Constantinople
(Valens, Procopius), Amiens (Gratian), and Sirmium (Valentinian II, Theo-
dosius). Suddenly it was exclusively concentrated at the “second Rome.”
Arcadius’s proclamation in  was the first in a long line of Byzantine
emperors stretching for centuries ahead and the Byzantine empresses after
Flaccilla adopted her first name “Aelia.” Both the ceremonial and the mean-
ing of proclamation developed over time but its customary Byzantine core
was essentially established in the time of Theodosius.
Every fifth year, generally beginning on his dies imperii, an emperor
would commemorate his anniversary with games, statues, and donatives
to the soldiery of specially minted commemorative coins. Vows (vota)
were discharged for the previous five years and renewed for the next five.
Theodosius began his fifth year on  January  by crowning Arcadius
as emperor. Thereafter the two emperors, father and son, celebrated their
imperial anniversary on the same day. This duplex celebration not only
amplified the imperial dignity and authority at Constantinople; it also
constituted a significant dynastic statement. Beginning on  January 
Theodosius marked his tenth year and Arcadius his fifth “with exhibitions
and games.” This was the very occasion that gave rise to the celebrated
silver dish, or missorium, of Theodosius, in which he and Arcadius are
 Socrates, Hist. eccl ..; Sozomen, Hist. eccl. ..; Philostorgius, Hist. eccl. .; Cons. Const. .
(MGH.AA, IX ), Marcell. . (MGH.AA, XI ).
 Socrates, Hist. eccl. .; Sozomen Hist. eccl. ..; Philostorgius, Hist. eccl. .; Lib. Ep. ;
Marcell.  (MGH.AA, XI ).
 References in PLRE I, s.v. “Fl. Honorius” (), and CLRE, – (consulship of ). For the later
investiture ceremony: Const. Porph. De caer.  (Vogt : –) =  (Reiske .–.).
 Holum : –. For the later ceremonial: Const. Porph. de caer. . (Vogt :–) = .
(Reiske).
 Const. Porph. De caer. .– (Vogt : –) = .– (Reiske), with Bauer : –.
 Cons. Const. . (MGH.AA, IX ); Marcell. . (MGH.AA, XI ). Numismatic data suggests
that the celebrations may not actually have been synchronized (Kent : ).
 brian croke

depicted in the tribunal of the imperial palace handing the codicil of

appointment to some official. On the same day in  Theodosius marked
his fifteenth and Arcadius his tenth regnal year. At least, that was the normal
pattern.
An imperial marriage was also a splendid and colorful occasion full of
light and striking scents, song and merriment. We catch a furtive glance
in  at the celebrations for the union of Theodosius’s favourite niece
Serena, “clad in scarlet,” and his promising young general Stilicho. The
city was bedecked with flowers and the Bosporus glittered with the torches
carried in procession, as the empress Galla was standing in as mother of the
bride “ordering the bridal veil beneath a weight of jewels.” All the more
resplendent and memorable would have been the day when Galla herself
had married Theodosius the previous year, or when the emperor Arcadius
was married in May . Once more the ritual for Byzantine imperial
marriages was evidently accelerated in the late fourth century.
Imperial funerals were no less public and ceremonial. Gregory of Nyssa
was chosen to present an encomium in honor of the young princess Pul-
cheria, who died in July . A few weeks later her mother, Flaccilla Augusta,
also passed away, and again Gregory was pressed into service. In the course
of his eulogistic set-piece, he depicts the funeral cortege of the empress pass-
ing through the streets of Constantinople. Both “strangers and residents”
wailed loudly at the sight of her coffin draped in purple and glittering gold
and carried on the empress’s litter. As it passed by “people of every rank
and age rush out, they marvel at this sight visible to all. Enthusiastically
following on foot in a great throng and giving vent to grief.” It was a
repeat of the crowded scene of concentrated grief witnessed at the funeral
liturgy for young Pulcheria. “I have seen a sea of men crammed together”;
so Gregory depicts the scene, “the full temple, the vestibule, the open
expanse before it, people in mourning, the nearby streets, public areas, the
side streets and houses. Wherever one looks there are crowds of people as
if the entire world had run together for this tragedy.” Similar outpouring
of communal sorrow will also have accompanied the funerals of the young
prince Gratian in  and newborn John in . Theodosius himself died
 Ernesti : –; Leader-Newby : –. The detailed iconographic argument seeking
to date the missorium to  and identify the emperor as Theodosius II, which was developed
by Meischner , is fatally flawed. Theodosius II celebrated his vicennalia in . To claim that
emperors were cavalier about dating their vota celebrations (Meischner : –) is itself cavalier.
 Claudian, Cons. Stil. .–.  Chron. Pasch.  (.–.).
 Const. Porph. De caer. . (Reiske .–).
 Greg. Nyss. In Flaccillam ( Spira), with Holum : –.
 Greg. Nyss. In Pulcheriam ( Spira).
 Reinventing Constantinople 

in Milan in January , but his body was transported to Constantinople

and laid to rest there alongside both his wives, Flaccilla and Galla, who
had died in childbirth as he was setting out on campaign in August .
Exactly where his predeceased children were buried is not clear.
By then Theodosius had systematically fixed and promoted the city as
the imperial capital by bringing into it the remains of as many of his
predecessors as possible. Constantine had set up his own mausoleum and
was buried there. Constantius joined him in the mausoleum in .
Julian’s status as an imperial apostate could even be turned to advantage by
a Christian emperor. Accordingly, Theodosius had Julian’s body brought
from Tarsus, where he had been buried in , and at the same time
he evidently transferred from Rome the body of Julian’s wife, Helena.
They were reunited in the early s in a round porphyry sarcophagus
located in a portico just outside the mausoleum of Constantine at the
Church of the Holy Apostles. Jovian’s body had already been transferred
from the place where he died in Asia Minor. Now he was reburied in the
mausoleum portico, and his wife, Charito, was buried beside him when
she died sometime in the reign of Theodosius. Since Valens’s body was
never found, he required no burial, but it is not known what happened
to his wife Domnica, who also probably died in Theodosius’s time. The
western emperor Valentinian, brother of Valens, had died in Pannonia in
, and his body had been brought to Constantinople in . Yet it was
left to Theodosius in  to arrange for his sarcophagus to be transferred to
the imperial mausoleum stoa accompanied by due ceremonial. The ashes
of Valentinian’s son Gratian had been acquired for Milan by Ambrose,
who treasured them, but Theodosius transported to Constantinople the
body of Gratian’s first wife, Constantia, daughter of Constantius II. On 
December  her body was solemnly placed in an imperial sarcophagus at
the Church of the Holy Apostles.
Theodosius had gone to considerable lengths to bring together into a
single precinct in the imperial capital every imperial corpse he could find.
By the early s he had retrospectively turned the Mausoleum and the
Church of the Holy Apostles into a truly imperial resting place and concep-
tually linked the two buildings. This sudden concentration of imperial
inheritance could only strengthen the authority and dynastic prestige of

 Grierson : , Johnson, a: .  Amm. Marc. Res gestae ...
 Grierson : – and G. Kelly : –, contra Woods .  Grierson : –.
 Marcell. . (MGH.AA, XI ); Cons. Const. . (MGH.AA, IX ), with Johnson b: –.
 Cons. Const. . (MGH.AA, IX ); Chron. Pasch.  (.–), with Grierson : .
 Grierson : .
 brian croke

Theodosius, who had erected a solid imperial phalanx protecting the city’s
Apostolic shrine. As John Chrysostom observed, “At Constantinople those
who wear the crown think themselves fortunate to be buried not near the
apostles but outside on the threshold of the basilica [of the Holy Apostles].”
He went on to explain that “from now on the emperors are the doormen of
sinners and in their eyes and the eyes of their descendants that is no shame
but an honor for their ashes.” By the early sixth century the mausoleum
populated by Theodosius’s “doormen” was full and another mausoleum was
built by Justinian, which became thereafter the resting place of successive
Byzantine emperors.
Other imperial occasions gave rise to urban ceremonial and Theodosius
legislated on  February  to specify some of them, essentially reinforcing
the prescript of his predecessors that certain formal public announcements
should be dignified and free:
whenever any of our auspicious achievements are announced, if wars should cease,
if victories should arise, if the honor of the bestowal of royal vestments should be
added to the calendar [that is, an imperial consulship], if the announcement of the
tranquillity of peace that has been concluded is to be spread abroad, if by chance
we display the imperial countenances [sacros vultus] to the eager multitudes.
Among other events, this advice covered victory announcements at Theo-
dosian Constantinople in , , , , and , as well as the estab-
lishment of peace with the Goths in , which was cast as a triumph,
and probably that with the Persians in /, when peace was agreed to.
In fact, Constantinople celebrated more triumphs and victories in the s
than it had in all its previous life. Also included were imperial consul-
ships in  (Theodosius),  (Arcadius),  (Honorius),  (Valentinian
II),  (Theodosius),  (Valentinian II),  (Arcadius),  (Theodos-
ius), and  (Arcadius and Honorius). Claudian expressed the hope that
Theodosius would continue to enjoy such festal days.
The traditional imperial ceremonial surrounding births, marriages,
installations, anniversaries, triumphs, and deaths began to take on a dis-
tinctly liturgical flavor at Constantinople in the era of Theodosius, although
this has usually been considered a later development. Candlelit proces-
sions along the city’s porticoed backbone, accompanied by hymns, began
to spread. The monumental and physical layout of the city facilitated the

 Jo. Chrys. Contra iudaeos et gentiles  (PG .).  CTh ...

 Them. Or. ., c (Maisano –) = Heather and Moncur :  with n. .
 Greatrex .  McCormick : .  Claud. Carm. min. .–.
 Diefenbach .
 Reinventing Constantinople 

development of a close relationship with the population, which continued

to grow under Arcadius and later emperors. It exposed the emperor to
his officials and people in new ways, while the senate and imperial court
were also now in a close political and social relationship that was partly
expressed in ceremonial engagement. When Theodosius finally set foot in
Rome in  and spent several months there, the locals were struck by
the emperor’s public visibility, as highlighted by Pacatus: “You frequently
emerge [from the palace] and you show yourself to the waiting people,
and being willing not only to let yourself be seen, but to be approached
readily.” Theodosius was transposing to Rome the emerging imperial
ritual of Constantinople by openly presiding at civic events and meet-
ing with people in public. At Rome seeing the emperor out on the city
streets surrounded by clergy and courtiers was a novelty. Those who could
remember were used to more closeted emperors.
Theodosius’s blending of secular and religious ceremonial was especially
evident in his regular acquisition of the relics of saints and martyrs, begin-
ning with the body of the exiled former bishop of Constantinople, Paul, in
. The spiritual and ceremonial emphasis came to be placed on the actual
arrival of the relics, the traditional adventus, which symbolized God’s mercy
for the city and its divine destiny. The emperor’s direct role in cradling
the holy relics placed him at the center of this unifying new ceremonial,
which brought together as common suppliants the court and clergy, aris-
tocracy and general populace. The transfer of Paul involved Theodosius
carefully carrying the former bishop’s skull in procession through the city,
depositing it in the church previously occupied by the Macedonians, and
formally changing the name of the church to “Saint Paul.” Theodosius
was also responsible for bringing to Constantinople the remains of the
African martyrs Terentius and Africanus with solemn deposition in the
church of Euphemia, and in  the head of John the Baptist reached
the city. Theodosius received the skull of John encased in its precious relic
box, then carried it himself folded in his purple imperial cloak all the way
out to the Hebdomon, where he deposited it in his newly built church.
The removal of John’s head to the church at the Hebdomon enhanced its
significance as a location where emperors were legitimated and presented,
 pan. Theod. . (trans. Nixon). This is not mere panegyrical exaggeration. Theodosius’s accessibil-
ity was also noted by Rufinus (HE .) and Gregory of Nazianzus, who tells us that the emperor
“treated me with respect from our first meeting both in the way he spoke and in listening most
kindly” (Poems ..– = White : ).
 Brown : –.
 Socrates, Hist. eccl. ..–; Sozomen, Hist. eccl. ... For the church itself: Janin : –.
 Nicephorus Callistus, Hist. eccl. . (PG .A).  Sozomen, Hist. eccl. ...
 brian croke

originally in the military ceremony of being raised on a shield. The

new presence of John formed part of a wider process at Constantinople
under Theodosius that saw the replacement of the traditional military
ideology with a sanctified civilian one. Reinforcing this process, the
emperor deliberately employed the Hebdomon as the departure point for
the campaign against Eugenius in . There he prayed for victory in the
new church, fortified by the prophecy of the holy man John of Lycopolis
whose prediction Theodosius had first sought out, both in  and .
On hearing of the death of the monk Isaac on the outskirts of the city on
 May , Theodosius had his body brought to Saint Eirene for a vigil.
The next day the emperor, patriarch Nectarius, along with all the clergy
and people, bore Isaac’s remains to his grave, accompanied by psalms and
hymns.
These processions served to consolidate further the status and dynastic
prestige of Theodosius and his family. There were many more of them
in the time of Arcadius and Theodosius II, and they are often thought to
have originated then, but they followed the model for relic receptions
established by Theodosius I. All of these processions and receptions at Con-
stantinople were infused with color and pageantry and they contributed to
the emperor’s concern for improving the standards of formal public dress
and deportment. Senior civil and military officials were to progress through
the city in fine carriages, while a fierce fine or even banishment from the
senate could result for any senator not complying with the new Theodosian
public dress code. Each new type of procession became an exemplar for
the next and thus the ritual and practice of Byzantine imperial ceremonial
began to be codified. An integral part of the ritual was the involvement
of the patriarch and, frequently, the patriarchal churches of Hagia Sophia
and Saint Eirene.
Except for the first few months, Nectarius was patriarch for the whole of
Theodosius’s reign. This continuous partnership of emperor and patriarch
from  to  was crucial. Nectarius was a distinguished senator who
was surprisingly nominated, then chosen for the position by the emperor

 Diefenbach : –.  MacCormack : .

 Theodoret, Hist. eccl. .; Aug. civ. Dei .; Cassian, Conf. .., Inst. .; Sozomen, Hist.
eccl. .; Ruf. Hist. eccl. .; Pall. HL .. John had a well-established reputation and was
consulted by officials from everywhere (Sozomen, Hist. eccl. .; Sulpicius Severus, Dial. .;
Augustine, De cura pro mortuis gerenda ).
 Vita Isaacii  (AA.SS. Mai VII, ). There is uncertainty about the date and identity of Isaac, but
the information of the vita may well be correct (cf. Dagron : , –).
 Mergiali-Sahas : .  Klein : –.
 The various relevant laws are discussed in Errington : –.  Berger : –.
 Reinventing Constantinople 

himself. He formed a link between the ecclesial and civil aristocracy.

Through the s and s he also helped define the emperor’s central
role in the local liturgy and in other religious occasions such as relic
processions. Moreover, some of these original Theodosian events and
processions came to be commemorated annually thereafter. So, in the
eleventh century the Byzantines still celebrated, with an annual liturgy,
events such as Theodosius’s transfer of Paul’s relics on  November and
the death of Flaccilla on  September. All these events provide a clear
indication of the significance of Theodosius’s years at Constantinople for
the formation of the Byzantine liturgical calendar.

monumental and decorative settings

Constantinople may already have been close and crowded when Theo-
dosius arrived in , but he continually expanded it during his reign
and improved its street plan. The prolonged location of the court there
attracted an increased population of officials and soldiers, teachers and stu-
dents, petitioners and craftsmen, both permanent and transient, which put
new pressure on the city’s basic infrastructure of shelter, water, and food. So,
by  the city prefect Aurelian had to be reminded that the senior military
figures receiving public rations (annonae civicae) in exchange for build-
ing houses must meet their obligations. Contributing to the necessary
restoration of the city’s port and water supply was another obligation that
could not be avoided. Theodosius’s restoration consisted of extending
the aqueduct of Valens further across the city and building a new harbor
on the Marmara side of the city at the mouth of the Lycus river, which
expanded capacity to unload the extra grain ships and other regular cargoes
required for provisioning the city. Recent excavations for the new Mar-
maray rail tunnel have unearthed the original Theodosian harbor, which
lasted for over a millennium and has already led to some spectacular finds

 Bauer ; Berger . In Milan, on the other hand, Bishop Ambrose refused to allow the
emperor onto the altar in the wake of the imperial sanction for the massacre at Thessalonika in
 (Theodoret, Hist. eccl. ..–). This episode later became a talismanic one for the Byzantine
tradition (e.g., Const. Porph. De caer. . [Reiske .–]) and helped clarify the boundaries of
imperial sacrality, as explained in Dagron : –, –, –.
 Typicon CP,  November (Mateos –, vol. : .–). Other feasts of Paul were on  August
(Mateos –, vol. : .–) and  September (Mateos –, vol.  .–).
 SEC .–.  Berger : –; : .
 CTh. .. ( April ).  Ibid. .. ( January ).
 Berger : . The “Theodosian aqueduct” was extensive enough to require management by
five praetors (CTh ..:  December ).
 Mango : ; : .
 brian croke

of later Byzantine ships and other objects. Nearby he constructed new

grain stores, the horrea Theodosiaca, which Byzantine emperors continued
to inspect with elaborate ceremony each year. These stores enabled him
to increase the imperial donation of free public grain, which was publi-
cized on his equestrian statue erected in the Milion. Theodosius must
also have been responsible for some modification and expansion to the
imperial palace but there is no record of such work – similarly with the
palace at the Hebdomon and possibly other palaces around the Bosporus.
At the same time new personal palaces were built for both Flaccilla, near
the Church of the Holy Apostles, and Galla, near the imperial palace.
Arcadius may also have constructed his own palace after , but again
there is no documented confirmation. Theodosius was also responsible
for a new hippodrome near the mansion of Eleutherios, and in  the
Arcadian baths situated just north of the imperial palace were opened with
great celebration for the city.
Imperial building sponsored or executed by Theodosius reflected the
emerging ideology and the new modes and expectations of court ceremo-
nial. One of the earliest (/) may have been the “Gothic column,”
which still stands in the garden of the Seraglio just beyond the Topkapi
Palace. However, the most important and impressive of Theodosius’s
constructions was the forum which bore his name, although it was also
known as the Forum Tauri, perhaps best explained as being originally the
property of the distinguished former prefect Taurus. It was located west
of the forum of Constantine, further along the Mese but before the Capi-
tol. In recent decades the forum has been the subject of regular research,
mainly inspired by new excavation which has unearthed various different
pieces of the forum structure and related buildings. Work on the forum
must have commenced not long after Theodosius arrived in Constantino-
ple in . The porticoes and entrance archways at each end were clearly
 Rose and Aydingun .  Const. Porph. De caer. . (Reiske .–.).
 Paras.  (Preger .–.).  Chron. Pasch.  (.–) with Janin : .
 Chron. Pasch.  (.–) with Janin : –.  Patria . (Preger .).
 Marcell. . (MGH.AA, XI ).  Peschlow : ; Bardill : .
 PLRE I, s.v. “Flavius Taurus” (). He was the father of two even more distinguished sons, Aurelian
(PLRE I, s.v. “Aurelianus” []) and Eutychianus (PLRE I, “Flavius Eutychianus” []). Taurus
presumably retired to Constantinople after Julian’s death in , when he was able to be recalled
from exile for having supported Constantius against Julian in . The other plausible explanation
of the name is that it was originally a market modeled on that at Rome known as the “forum of
the bull (tauri).”
 Janin : –; Faedo ; Bauer : –. A useful and well-illustrated starting point for
both context and the disparate excavated fragments related to the forum is Barsanti . Recent
discussion began with R. Naumann (), who reconstructed the forum’s large entrance arch in
light of the bases and columns unearthed in excavations between  and .
 Reinventing Constantinople 

grand as was the basilica erected along one side which was finished by the
mid-s. Questions about the exact size of the forum (larger or smaller)
and whether the uncovered arch is the western or eastern entrance to the
forum are still not answered satisfactorily. The forum was decorated with
equestrian statues of both Arcadius and Honorius which may have been
installed at the time the column was completed or in , when there was
a formal dedication of a new equestrian statue of Theodosius mounted on
a pedestal.
The building and iconography of the forum anchored and echoed the
emperor’s dynastic ambitions, linking back to Trajan and forward to his
imperial sons. The carved spiral column that dominated the forum was
constructed at the same time, with its decoration depicting the imperial
victory over the Goths in , when Theodosius and Arcadius celebrated a
triumph at Constantinople. Some fragments of the column were incor-
porated in the baths of Sultan Bayazit II and in recent times other remnants
of the column have been uncovered nearby. The column was a deliber-
ate imitation of that of Trajan at Rome, and the connection with Trajan
was always an integral part of Theodosian propaganda. It was accentuated
in the forum by promoting decorative motifs of Hercules, a patron god
of the Spanish emperors Hadrian and Trajan, which have been identified
on the entrance arch columns. The inscription on the equestrian statue
addresses Theodosius as a “second light bringing sun,” a phrase designed
to evoke Constantine’s solar associations and promote Theodosius as the
second founder of Constantinople to rival Constantine. The forum cre-
ated a distinctive space unique to Theodosius and new opportunities for
expanding the public ritual of the city. It became a new ceremonial stage
for the emperor, a place where he could be seen in public and in context
and where the voices of the people could shout appropriate acclamations.
Apart from becoming a fixed station on regular processional routes, it was
now the central locale for certain events such as the reception of barbarian

 Faedo : .

 Mango (: ) estimated the size of the forum as  meters by  meters, being equivalent to
that of Trajan’s forum at Rome, but Berger (: –) has argued strongly for a smaller forum
( meters by  meters) and that the surviving part of the arch of Theodosius was the western, not
the eastern, entrance, as R. Naumann () had proposed in his reconstruction.
 Chron. Pasch.  (.–). Statues: Paras.  (Preger .–), Patria . (Preger .–), .
(Preger .–.), . (Preger .–). Arcadius later added an equestrian Theodosius II to
the collection: Patria . (Preger .–.)
 Zosimus ..; Cons. Const. . (MGH.AA, IX ); Oros. ..; Marcell. . (MGH.AA, XI
). For the erection of the column: Theoph. Chron. AM  (de Boor .–).
 Sande , which presents and analyzes five fragments discovered in  and their iconography.
 Barsanti : –; Faedo .  Anth.Plan. . with Leppin : .
 brian croke

envoys and the emperor himself when he returned from the west. The
Persian envoys in  were probably received there too.
In addition to those statues in the Forum Theodosii, there were many
others erected by or for Theodosius throughout the city: at the Chalke, the
entrance vestibule of the imperial palace, the Great Church, and the
hippodrome, while at the Basilica cistern could be found a bronze statue
of the seated emperor. There was also the silver one on the column in
the Augusteon. In addition, a bronze equestrian one was located at the
Milion, where Theodosius also set up statues of Hadrian and Trajan,
which were possibly erected elsewhere in the second century and reappro-
priated by Theodosius to form a dynastic set. Statues could even have their
own sacred function in representing the sacred presence of the emperor.
In fact, Theodosius announced in  that asylum could be claimed by
fleeing to the “statues of the emperors.” As for churches, while Theo-
dosius promoted their use and integrated them more fully into the public
and imperial life of the city, he was a limited church builder. Certainly
he was responsible for the Church of John the Baptist at the Hebdomon
which was evidently proposed to him by Rufinus; as well as that of
the “Holy Notaries” (Marcian and Martyrius, executed ) and the
Church of Saint Mark near the Forum of Theodosius. Also attributed
to him, probably incorrectly, is the Church of the Virgin on the property
of Eugenius.
All this imperial construction – columns, forum, statues, churches –
consolidated and amplified the emperor’s status within the city and reflected
his imperial power, patronage, and dynastic intent in a traditional way.
It also complemented and enhanced the ceremonial by exhibiting the
 Details in Bauer : –.
 Marcell. . (MGH.AA, XI ); Oros. ..; Socrates, Hist. eccl. .; Cons. Const. .
(MGH.AA, IX ); Chron. Pasch. ..
 Paras.  (Preger .–); Patria . (Preger .–) with Bauer : .
 Paras.  (Preger .).
 Paras.  (Preger .–); Patria . (Preger .–), a (Preger .–).
 Paras.  (.–).
 Marcell. . (MGH.AA, XI ); Paras.  (Preger .–).  Paras  (.–.).
 Cedrenus, Hist. comp. vol. , .–. Theodosius is the most obvious candidate for this particular
statue set.
 CTh .. ( July ).
 Patria . (Preger .–); Chron. Pasch.  (.–); Sozomen, Hist. eccl. ..; Theodore
Anagnostes, Hist. eccl.  with Berger : –, –.
 Patria . (Preger .–) with Janin : –.
 Patria . (Preger .–) with Janin : .
 Patria . (Preger .–). He is also often thought to be responsible for the reroofing of Hagia
Sophia, although this may be the work of his grandson Theodosius II (Dagron : –
[translation of Descriptio] and , ).
 Reinventing Constantinople 

imperial countenance in a fashion that bound it to the closely pressing

populace of the city. It was now a shared ritual in a shared space. At the
same time, Theodosius and his court officials along with local dignitaries
and aristocrats were increasingly attracted by the new forms of power
and patronage embedded in the local holy men, along with the relics
of saints and martyrs. There was a sudden growth in monasteries and
martyria at Constantinople, encouraged and promoted by Theodosius.
The emperor himself was responsible for bringing the monk Dios from
Antioch and settling him in a local monastery. Prior to that, the city’s
first monk, Isaac, was provided for by Theodosius’s general Saturninus
on his own property, which was near the city wall at Helenianae and
neighbored that of his fellow general Victor. Only recently they had both
been leading Roman troops against the Goths. Theodosius used to visit
Isaac there, as well as his successor in the monastery Dalmatius, a former
palace guardsman of Theodosius.
The way was paved for development of martyria by the emperor himself,
who decreed in  that burials should be outside the city and that any
sarcophagi resting above ground should be relocated there. In  he
further pronounced that in the case of a local saint or relics a martyrion
could be built over the grave. This is what happened in the case of
Eusebia’s shrine of the forty martyrs and probably that of the Persian
trio put to death by Julian in , namely, Manuel, Sabel, and Ismael.
Theodosius’s praetorian prefect Rufinus built a martyrion of Peter and
Paul on his estate at Drys near Chalcedon, incorporating relics he had
brought from the west in a solemn depositio. The city prefect Aurelian,
however, built a shrine in anticipation of receiving the relics of the martyr
Stephen but they never eventuated for him. Instead he had to content
himself with the temporary relocation of the body of Isaac there in .
Noble women, no less than men, were also patrons of holy persons and
objects in Theodosius’s time. Martyria were constructed in the Lycus valley
by Alexandria and Gregoria, otherwise unknown but possibly the wives of

 The city’s martyria are listed in Maraval : –.

 Patria .a (Preger .–) with Janin :  and : –.
 Vita Isaacii – (AA.SS.Mai VII, –), with Janin : .
 ACO .. [], . with Saradi : .  CTh .. ( July ) with Dagron : .
 Ibid. .. ( February ).  Sozomen, Hist. eccl. ..–.
 Patria . (Preger .–).
 Callinicus, Vita Hypatii .– (Bartelink ). The monastery and its location became known as
“Rufinianae” (Janin : –).
 Vita Isaacii  (AA.SS. Mai VII, ).
 brian croke

leading imperial courtiers. Another monastery was founded at the time

by a western woman named Domnica who came to Constantinople via
Alexandria and was baptized by Nectarius.
The single event that gave rise to a new burst of imperial building activity
was the victory over the usurper Maximus and his son Victor in , which
was duly celebrated in Constantinople. Indeed, the day became firmly
fixed in the Byzantine calendar thereafter. Although Theodosius did not
return to the east immediately and remained away from Constantinople
for nearly four years (August –July ), the city still had a resident
emperor, the teenage Arcadius. So the ceremonial routine of the city and
court continued uninterrupted, and the task of immortalizing Theodosius’s
new success proceeded apace in the usual fashion. Outside the city was
built the magnificent Golden Gate, a ceremonial portal for the victorious
emperor as he approached the capital once more and it may well have been
completed in time for his return to Constantinople in . By then, too,
there was erected another famous memorial that is still standing in situ, the
obelisk in the hippodrome. The obelisk rests on a base in two parts and
depicts scenes of the emperor and his sons presiding at the games. Again,
it was probably set up in  by the prefect Proculus when Theodosius was
still in the west. The obelisk has been the subject of intensive research
in recent times, but uncertainty remains about the identity of the imperial
figures depicted, although there is general agreement that its iconography
is essentially dynastic, expressing the concord of east and west. Other
construction was also inspired by Theodosius’s victory in the west, namely,
a possible column in the Forum Strategii.

 Patria . (Preger .–.) with their historicity doubted by Janin (: , , and –).
 SEC .– with Janin (: ), who is inclined to dismiss the possibility on the, at least
questionable, grounds that it is difficult to imagine women founding monasteries at Constantinople
so soon after men, that is, Isaac and Dios.
 Procopius records its continued celebration in the s (Wars ..).
 The gate was previously considered most likely to commemorate the victory of Theodosius II
over the usurper John in  and to be built out of the Theodosian city wall constructed in .
However, more recent research (Bardill ) has clearly established that it was built separately
as a stand-alone monument before the wall was built and is therefore dated to the time of
Theodosius I.
 Marcell. . (MGH.AA, XI ). An earlier date () was proposed by Rebenich (b).
 Ritzerfeld () emphasizes the dynastic continuity in the monument while its presentational
function is addressed by Safran (; with access to earlier literature and excellent photographs).
Effenberger () mounted an elaborate technical case for claiming that the present base was not
the original but was added later to compensate for a broken bottom section of the column in order
to maintain the original column height from the ground. However, this has not proved convincing
(Safran ; Speck ; Kiilerich  contra Ritzerfeld : ).
 Mango : –.
 Reinventing Constantinople 

conclusion
Theodosius was drawn into Constantinople in  by the city’s senate,
facilitated by the persuasive intervention of Themistius. The local aristoc-
racy of office wanted to have the advantage of imperial proximity and the
patronage they thought they should have been enjoying all along. Once
there he reclaimed the city for the orthodox cause to which he was already
committed by baptism and turned his imperial attention from military
campaigns to protecting, improving, and promoting the lives of his impe-
rial subjects. By January  the senate and people of Constantinople could
be confident that Theodosius was now there to stay, and Themistius could
express strong satisfaction with the emperor’s investment in the city and the
way it was growing. Indeed, it was now difficult to say whose contribution
was the greater: how Constantine created a new city out of Byzantion or
how Theodosius had created a new city out of Constantinople. Looking
ahead, Themistius concludes, Theodosius would be seen to have created
a whole “third city.” Moreover, he had also now begun to entrench
Constantinople’s role as the “New Rome” by developing the imperial mau-
soleum precinct and relocating the remains of previous emperors there.
The succession was guaranteed now, too, as Themistius proclaimed.
Theodosius’s eldest son Arcadius was already crowned Augustus, and soon
after his wife, Flaccilla, was crowned Augusta. There would be no more
uncertainty about succession as there had been on the deaths of Julian,
Jovian, and Valens, nor would Constantinople ever experience an imperial
vacuum again for several generations.
By the time Theodosius left Constantinople to engage with Maximus
in /, he had already resided longer in the city than even Constantine.
By then he had set about creating his own monumental forum and other
trophies to decorate the city and promote himself and his family as an
established imperial dynasty. The settled presence of the imperial court
at Constantinople for most of the reign of Theodosius I enabled the
already large city’s rapid development as a genuine “New Rome.” Lack
of military success against the Goths in  and , combined with the
peace settlement with them in  as well as that with the Persians in ,
produced an emperor focused on court and capital. Instead of training his
troops and marching out with them each summer, as his predecessors had
been obliged or chose to do, Theodosius remained in his capital. The city

 Them. Or. ., b–b (Maisano –).  Ibid. ., b–b (Maisano –).
 brian croke

became his battlefield, and he even legislated to keep his sons and future
emperors anchored in the city and away from the military front.
In these years rather than later, as usually supposed, the imperial and
topographical core of Byzantine public liturgy and life was first fashioned.
This involved the creation of new ceremonial spaces in the city and the
development of a distinctive public ritual that resulted from a fusion of
stational liturgy from Jerusalem, traditional Roman imperial ceremonial, a
demilitarization of formal rites, and, as Matthews has shown, the orthodox
piety of Theodosius’s family and the many western officials and relatives
who accompanied him to Constantinople. It was Theodosius I who turned
the emperor’s role into one of managing and mediating the imperial power
through church and consistorium, senate and bureaucracy, court and urban
ceremonial. These are the very elements that appear now to be such a
solid and prominent feature of the reigns of his son Arcadius and grandson
Theodosius II. However, they merely followed and expanded the model
of urban development and decoration established by Theodosius I, who
had spent most of his reign inside the walls of Constantinople. Following
the final committal of Theodosius to his imperial sarcophagus in the
Church of the Holy Apostles on  November  the Byzantines continued
to remember on the same day each year the man responsible for the
reinvention of their city as a Christian imperial capital.

 Jo. Lyd. De mag. . (no such law is extant).

 As in the seminal study of Millar (), but with no mention of the role of Theodosius I or
Arcadius in the process. For this perspective, Holum  still provides the best model.
 SEC .–; Typicon CP,  November (Mateos –, vol. : .–). On the second anniver-
sary,  November , the new patriarch, John Chrysostom, could encapsulate Theodosius as
someone honored “not because he was emperor, but because he was pious; not because he was
clothed in the purple but because he had put on Christ” (Jo. Chrys. Adversus Catharos. Homilia
VI [PG .]).
 chapter 13

Reinventing history: Jerome’s Chronicle and the

writing of the post-Roman West
Mark Vessey

From Gibbon to Jones and beyond, late Roman historiography observes

a period at the year . The pause signals not just the disaster of Hadri-
anople, reckoned as the “beginning of evil” (Rufinus) or the “beginning of
the end” (Seeck) for Rome’s empire, but also the termination of Ammianus’s
Res gestae. For Jones, following closely in Gibbon’s footsteps, the quality of
Ammianus’s information merely postponed the moment when the modern
historian found himself obliged to rely on “the very inferior narrative of
Zosimus, eked out by the three Greek ecclesiastical historians, Socrates,
Sozomen and Socrates” and “supplemented by some Latin historians, who
lived nearer to the events which they describe, but are wretchedly meagre
in content.” Aside from the ecclesiastical histories of Rufinus, Sulpicius
Severus, and Orosius, the only Latin historical source worth mentioning
at this juncture was “the last two chapters of the Epitome de Caesaribus,”
amounting to a few pages on the reigns of Gratian and Theodosius. And
yet all was not lost. Disappointing as his strictly historiographical sources
for the Theodosian era might be, the modern historian had other stores
to fall back on. “The codes,” wrote Jones, “are rich in laws for the whole
period” and “the contemporary literature is also abundant.”
In the four decades since Jones’s Later Roman Empire appeared, the “law”
and “literature” of the Theodosian era have been laid ever more heavily
under contribution by late Roman historians. No single scholar’s work
is more emblematic of that development than John Matthews’s. As The
Roman Empire of Ammianus paused to take full measure of “the last of the
great Roman historians” and Laying Down the Law offered new prolegom-
ena, a century after Mommsen’s, for historians’ use of the first of the great
 Gibbon , vol. : – (with Matthews : ); Jones a, vol. : ; Rufinus, HE .;
Seeck –, vol. : . Cf. Matthews : –; Lenski ; G. Kelly : , .
 Jones a, vol. : . See also ibid. vol. : vi–vii.
 Matthews : , from the last sentence of an essay concluding a book on Ancient Writers: Greece
and Rome that opened with Homer and whose sequel for European Writers: Medieval and Renaissance
would begin with Prudentius, author of the “Theodosian history” of the Contra Symmachum.


 mark vessey

imperial law-codes, so in these and other studies – most commandingly

in Western Aristocracies and Imperial Court – Matthews has shown how a
sense of the “public” Roman history that we are used to reconstructing
from the evidence of ancient historical, legal, and documentary texts can
be enhanced, even transformed, through a close reading of those ostensibly
more “private” archives whose contents Jones, like others, lumped together
as literature: letters, poetry, even certain kinds of confessional and devo-
tional writing. In doing so, moreover, he has never failed to convey his own
strong sense of historiographical and historical limits.
The lower chronological terminus of Western Aristocracies and Imperial
Court, AD 364–425 is carefully chosen for its resonance and ambiguity.
Olympiodorus of Thebes, a Greek “man of affairs” who was at Rome in
 for the installation of Valentinian III, ended his annalistic narrative
of the western empire since  at that point, rounding out a story not
of decline and fall but of “the decline and recovery of the west, a divided
empire united by eastern initiative.” This historiographical plot, Matthews
suggests, would have been especially satisfying “for the eastern readers to
whom it was primarily addressed.” The suppression of the Italian usurper
Johannes and elevation of the five-year-old Valentinian as western Augustus
had been brought about by a force sent from the east by Theodosius II,
the emperor to whom Olympiodorus dedicated his history. Yet from a
certain, western point of view Olympiodorus’s emphasis on the restoration
of the Theodosian dynasty was likely to have seemed exaggerated. To any
western senator of the time, Matthews argues, “it would have been perfectly
clear . . . that the survival of Roman life in the west depended precisely
upon the efforts, and existence, of himself and his colleagues.” While
Olympiodorus could represent an “idea [of] imperial unity founded upon
eastern initiative” that was to remain a fundamental feature of Byzantine
policy, his presence at Rome in  – as the latest Greek reporter, after
Ammianus and Claudian, of an emperor’s brief sojourn in the ancient
capital – also served as a pointer for modern readers to a future in which
the west “would in practice usually have to be left to go its own way.” In
the epilogue to The Roman Empire of Ammianus, Olympiodorus is called
up again as spokesman for the “restored harmony of the Roman empire,”
this time to supply a link between the dynastic restoration of  and the
initiative for the Theodosian Code, represented as a further “assertion of
confidence in the power of man by effort and will to affect the circumstances

 Matthews : –, ; for the eastern ascendancy as a legacy of Theodosius I, and ultimately of
Constantine, see already Matthews : .
 Reinventing history 

that surround him.” Such acts bore out Ammianus’s conviction that Rome
was destined to “live as long as there [were] men” (..) and justified his
“classical” insistence on “connecting the eternity of Rome with human will
and effort.”
For all that, east and west “Rome” still went their respective ways in
the course of the fifth century. An important recent study has empha-
sized the linguistic dimension of that division and placed Olympiodorus’s
Greek history of the western empire on the line of parting. Even when
Olympiodorus is treated as a natural continuator of Ammianus, he cannot
conceal the absence of extant classical or classicizing Latin histories from
after the turn of the fourth century. In the conclusion to The Roman
Empire of Ammianus, as in Western Aristocracies and Laying Down the Law,
Matthews fixes an instant at which ancient, classical traditions – of his-
toriography, political life and culture, juristic thought and practice – can
be seen approaching a limit, even as they continued boldly to defy any
limitation at the time. The limit is marked chronologically as the end
of classical (late) antiquity, geopolitically by the sundering of east and
west.
The strictly – that is, textually – historiographical correlates of this long
transitional moment were aligned more than a century ago by Hermann
Peter in his monumental survey “The Historical Literature of the Roman
Imperial Period to Theodosius I.” Peter’s outlook on the later part of that
“literature” was unambiguously expressed in the preface to his accom-
panying edition of fragmentary Latin histories, where he explained that
he had drawn a line below Constantine’s reign, since after that, “apart
from a few authors still extant, who made epitomes of what others had
written, and brought the historical account down to their own time, all
history went over to the Christians and the Greeks (omnis historia ad
christianos Graecosque migravit).” The one exception he allowed himself in
the Historicorum Romanorum Reliquiae was the inclusion of a passage of
the sixth-century Roman history of Q. Aurelius Memmius Symmachus,
as preserved by Jordanes, “the sole fragment of this writer of their own

 Matthews : . The rapprochement between Olympiodorus’s history and the code is closer still
in Matthews : . In similar vein, Lenski (: ) calls Ammianus’s history “a handbook for
the future.”
 Millar : .
 Cf. Matthews : : “He claims a central place in a continuous tradition of Greek writing on
Roman affairs – a tradition notoriously lacking in western historiography.”
 To be distinguished from the more capacious “postclassical” late antiquity of Bowersock, Brown, and
Grabar (), which begins earlier, extends later, treats Christianity as a thing nearly indifferent,
and tends to favor the east after the fifth century.
 mark vessey

faith that the Christians were willing to see transmitted to posterity.”

Since Peter still took the scriptores historiae Augustae at their word, his
treatment of “pagan historical literature in the [later] fourth century” in
his survey of Roman imperial historiography falls neatly into three parts:
Ammianus Marcellinus, the Latin breviaria beginning with Sextus Aur-
elius Victor, and the Greeks. Tellingly, however, no Greek or pagan gets
the last word. Instead, Peter’s final chapter, dealing with the methods of
the breviarists, ends with an appendix on the genre of the “short” world
chronicle, devoted to Jerome’s version of Eusebius’s chronographical tabu-
lation, the Chronici canones. After summarizing Jerome’s prefatory account
of the work, reporting Mommsen’s analysis of the sources for its additions
to Eusebius, and giving a few specimens of Jerome’s handling of his most
important source for recent political history (presumed to be Eutropius),
Peter closes with a prospect of the Latin chronicle tradition of later cen-
turies: “The series of chronographical tables (Zeittafeln) stretching far into
the Middle Ages . . . takes its cue from Jerome alone. These works declare
themselves to be continuations. Long after the form of the Roman empire
had been shattered and replaced by others, they held fast with remarkable
tenacity to the idea that, even though it had fallen, the empire could never
be destroyed and Rome still ruled the world.” As the record of pagan his-
tory writing in Latin tailed off after the death of Theodosius I, so Jerome’s
Chronici canones, compiled at Constantinople within a year or two of that
emperor’s accession, could appear at the term of Peter’s survey as prologue
to the post-Roman historical consciousness of the west. Or as Piganiol
succinctly put it half a century later, listing Latin historiographical sources
in his volume on “The Christian Empire (–)” for a collaborative
Histoire romaine, “Viennent enfin les chroniques.” By either scholar’s
account, the coming of the (Christian) chronicles announced the end of
Roman history.
“For one writing as a Roman historian,” Matthews has remarked, “eccle-
siastical sources present the most difficult problem.” The difficulty he had
in mind was one of bibliographical reference, but the problem posed for
Roman history by ecclesiastical sources is clearly much bigger than that of
 Peter –, vol. : vi. This Symmachus was the grandson of the fourth-century defender of the
Altar of Victory. For the milieu, see Matthews : . Reynolds and Wilson (: –) stress
the continuity of “classical” historiographical interests.
 Peter , vol. : –.  Peter , vol. :  (emphasis added).
 Piganiol : ix. This Histoire romaine formed part of an Histoire générale; the next volume,
launching the Histoire du Moyen Age, was entitled Les destinées de l’Empire [sic] en Occident de 395 à
888.
 Matthews : .
 Reinventing history 

deciding when not to cite a text in the defective edition of Migne. (That
said, the character of Migne’s collection – as an articulate body of “patristic”
texts, more or less chronologically arranged – may be an aspect of the prob-
lem.) For the would-be historian of the western empire and its immediate
successor states, the enormous early influence of Jerome’s “Chronicle,”
along with certain related works by the same author, poses a special chal-
lenge. It is not just that the supply of classical or classicizing historiography
seemingly dries up shortly after Ammianus; from the s onward, Latin
“literature” in all kinds was so quickly and so largely accommodated to
Christian ideological norms, a doubt must arise as to how legitimately it
can be considered source material for any kind of “Roman” history – even
Roman literary history. It is no coincidence that many of our histories of
classical Latin literature reach their nec plus ultra in the fifth century, as
often as not with Sidonius Apollinaris. The gradual collapse of Roman
political institutions in Gaul and other western provinces supplies conve-
nient coordinates for the terminal narrative of these histories – without,
however, accounting for the actual forms taken by Latin literary activity of
the period. For while it may be granted that by the mid-fifth century in the
west “the whole social framework within which classical literature had been
written, read and criticized was unmistakably changed,” no one would
ascribe the transformation in question to political developments alone.
One of the more adventurous recent histories of Latin literature underlines
the problem by leaving space between chapters entitled “From Constantine
to the Sack of Rome (–)” and “From Honorius to Odoacer (–
)” for another, outside political chronology, “The Apogee of Christian
Culture.” “Apogee” may be too sudden. The authors brought under this
heading (Ambrose, Jerome, Augustine, Rufinus, and Sulpicius Severus)
were chief architects of a framework for Christian writerly, readerly, and
critical enterprise that would long outlast the Roman empire in the west.
With the partial exception of Ambrose, and beginning in any case with
Jerome, they were also notable historians.
Is there a relationship, beyond synchronicity, between the gradual extinc-
tion of Roman historiography in the west and the emergence of a new com-
mon consciousness among readers and writers of Latin such as would now

 Though less correct than Chronici canones, this form of the title is more familiar and will be used
here.
 Walsh (: ), noting how “[a] synthesis of classical and Christian culture began to be formed
in the west, which was distinct from that of the Greek east.” See also Liebeschuetz b: –
(“The Transformation of Literary Culture in the West under the Influence of Christianity”).
 Conte .  For Ambrose’s sense of Roman history, see Inglebert : –.
 mark vessey

justify our speaking, with whatever necessary qualifications, of a Christian

“literature”? This essay will argue that there is, and that it can be seen already
plotted in Jerome’s Chronicle. Without ignoring such evidence as exists for
the continuing practice of classicizing Roman historiography in Latin after
the turn of the fourth century, we shall sketch the cultural conditions that
would make it impossible, in the long run, even for works like that of
Q. Aurelius Memmius Symmachus to survive intact. The demonstration
will entail some abuse of chronology, though none worse than Gibbon’s in
the climactic chapter  of his Decline and Fall. Preparing to narrate the
Gothic sieges and sack of Rome in , and wanting to “produce an authen-
tic state of the city and its inhabitants,” Gibbon recalled Ammianus, whose
company he had lost some chapters earlier, after the battle of Hadrianople.
For all his faults, the old Greek soldier could be trusted to render a faithful
“picture of the manners of Rome,” having “prudently chose[n] the capital
of the empire . . . as the residence best adapted to the historian of his own
times.” To supplement Ammianus’s description, Gibbon looked to two
other eyewitnesses: Olympiodorus of Thebes, who (so far as we know) did
not visit the city until , and Jerome of Stridon, who left it for the last
time in . The latter will here supplant the former. As John Matthews has
relied upon Olympiodorus to carry the enterprise of a Rome-centered, clas-
sical historiography beyond the temporal and geographical limits indicated
by the death of Theodosius, we aim to show how Jerome, as a traveling
Latin “man of affairs” without portfolio, unconsciously anticipated that
moment of demise and, as a historian in his own fashion, inaugurated a
line of Christian writing that would partly define the post-Roman west.
Our destination will be the textualized Rome of Jerome’s Constantinopo-
litan Chronicle. To begin with, however, we shall return to the city itself,
around the time Ammianus was putting the final touches to his Res gestae,
to consider more generally the kinds of historiographical end that were
then and there being made.

endings (rome, ca. 389)

In the late spring of  a provincial orator took the road to Rome, has-
tening, he would say on arrival, “from the farthest recesses of Gaul, where
Ocean’s shore receives the setting sun and where the land gives out and
is united with its companion element.” The miniature travelogue had
 Gibbon , vol. : –.
 Pan. Lat. ..; trans. Nixon and Rodgers : . For the political circumstances, see Matthews
: –; McLynn a: –; and Sogno : –.
 Reinventing history 

political as well as poetic resonance. A year earlier, Theodosius had defeated

the usurper Magnus Maximus, much of whose support came from Gaul.
In the speech that he was now to deliver before the Roman senate, in Theo-
dosius’s presence, this Gallic orator would affirm his countrymen’s loyalty
both to the eastern Augustus and, however grudgingly, to his ineffectual
colleague in the west, Valentinian II. If Latinus Pacatus Drepanius had
been to Rome before, that did not stop him from acting the awestruck
provincial. Those invited to speak before the emperor or his high officials
were used to beginning tremulously; it was in the official script. Pacatus
for his part professed an insecurity he patently did not feel. He was, he
claimed, only too conscious of speaking like a rude Gaul before so many
masters of pure Roman eloquence. To complete his self-discomfiture, he
imagined a host of Catos, Ciceros, and Hortensiuses in the chamber that
day, listening to him in the persons of their natural descendants (., qui
me in posteris suis audiunt).
After the obligatory opening address to Theodosius, these legendary
names from the days of the Roman republic are the first personal references
in Pacatus’s speech. The terms in which he recalls “all those orators” (omnes
illos oratores) hint at a canon that he could rely upon his hearers to share.
We may conjecture that it was the canon of the Brutus, even if few in
the Curia that day had picked up Cicero’s dialogue since their youth. The
republican roll call was certainly calculated: no emperor could demur at it,
while an urban patriciate nostalgic for ancient privileges would relish the
compliment. Jerome’s letters to aristocratic Roman women, many of them
ostensibly written a few years earlier but perhaps only now beginning to
circulate widely, played the gamut of republican prosopography. A similar
archaizing and genealogical impulse appears in the letters of Symmachus,
who was presumably there to hear Pacatus speak. Noblesse oblige, but
newer nobility likes to be reminded of its duty. Urbane as they preeminently
were, Pacatus’s hosts would know how to reciprocate the respect that this
visitor from the world’s end was showing their reputed ancestors.
When in Rome, flatter the Roman sense of history. It is easy to see
what Pacatus was up to. One reason it is so easy is that someone (Pacatus
himself?) set his readers up to see it. A score or so of Latin “imperial” pan-
egyrics in prose have survived more or less intact from the period between

 Rebenich .
 Bruggisser (: ) characterizes book  of the correspondence as the portrait of “a family by
blood and of the spirit.” See ibid. – on Symmachus’s republican nostalgia, citing his reverence
for Cicero and Hortensius as masters of oratory. Eigler  is now the fullest description of this
mentality, which was not confined to senatorial milieux.
 mark vessey

the late third and early sixth centuries. No fewer than eleven of them
have been transmitted together in the collection known as the Panegyrici
Latini, which also includes Pliny’s panegyric to Trajan, based on a speech
delivered at Rome in the year . Pliny’s is the first speech in the order
preserved by the manuscript tradition. It is followed by Pacatus’s panegryic
to Theodosius, delivered almost three hundred years later. After that, head-
ing backward now instead of fast forward in time, the reader comes upon
a panegyric to Julian, given by Claudius Mamertinus at Constantinople in
, then one to Constantine, given by Nazarius at Rome in . Next is a
group of seven speeches, presented in nearly perfect reverse chronological
order of their original occasions (from  to ) and clearly signaled in
the manuscript as having once made up a collection by themselves. One
further speech (from ) completes the sequence of twelve as we have it.
Apart from Pliny, all the orators appear to have been natives of Gaul. The
eight earliest speeches were delivered in Gaul, seven of them at the imperial
capital of Trier, one at Autun in Lugdunensis Prima.
The century-long tradition of Gallo-Roman oratorical prowess exem-
plified by these later Latin panegyrics explains the confidence with which
Pacatus played the rustic provincial card at Rome. The rhetorical schools
of Gaul were the best in the west. Since the creation of a tetrarchic capital
at Trier, moreover, Gaul had boasted an auditorium for those wanting to
catch the emperor’s ear. Even the squeamishly metropolitan Symmachus
had braved the savage conditions of a frontier zone, in –, to deliver
golden tributes to Valentinian I and his son Gratian on behalf of the
Roman senate. However, it is only the pre-archetypal, “core” collection
of seven (plus one) panegyrics, all from the tetrarchic period itself, that
revolves around Trier. The later additions tell a different story: one of
peripatetic Gallo-Roman orators in other capitals and of an emperor who
might still, even when not in Trier, be almost anywhere but Rome. Nazarius
had praised Constantine in Rome in , but not to his face. Mamertinus
had praised Julian to his face in , but in Constantine’s new capital on
the Bosporus. Only Pacatus, of this elite Gallic company, could truly rub
shoulders with Pliny, having, like him, spoken in the sacred presence of
the emperor himself and in the Eternal City. It is usual to assume that

 Text (reprinted from R. A. B. Mynors’s Oxford edition), translation, and historical notes in Nixon
and Rodgers ; for new perspectives on the genre, see esp. Whitby ; Averil Cameron :
–.
 Sogno : –. See also MacCormack :  for Symmachus’s panegyrics at Trier as “an
attempt to put Rome on the map,” with Shanzer  on his rhetorical contest with Ausonius over
the limes.
 Reinventing history 

Pliny’s panegyric was prefixed to the collection because it was the model,
even the school model, for the genre, but that inference does not exhaust
the codicological interest of the placement. The non-Plinian items in the
Panegyrici Latini would not account even for  percent of the total number
of Latin speeches in the genre that must have been made between  and
. Nor, of course, did the genre cease with Pacatus. Whoever it was who
put the latter’s panegyric to Theodosius next after Pliny’s to Trajan, then
sent time’s arrow flying back toward Trier in the late s, contrived an
effect of premature closure that can still be felt.
Ironically, that original artificial closure is now more likely to be expe-
rienced as a loose ending. For most practical purposes, the modern canon
of the Panegyrici Latini consists of just eleven speeches, and in a different
order from that of the tradition. Pliny’s oration – because it is so long,
so much earlier than the others, and by a ranking “classical” author – has
been set aside. And because we like our late Roman history to run steadily
forward, the remaining panegyrics have been rearranged (in current edi-
tions with English or French translations) so as to begin in  and end a
century later. As a result of this scholarly tidying-up, Pacatus now speaks
last of all. Instead of closing prematurely at the place where he and Pliny
met, the collection ends with his explicit.
We have already seen how this latest Gallic orator began, skillfully
conciliating his Roman audience. His peroration is no less Rome-centered.
The city herself is apostrophized, in much the same republican persona as
had earlier been conjured from her senatorial ranks:
You watched [spectabas] this from your hills, Rome, and sublime on your seven
citadels, you were raised even higher with joy. You, who experienced the raging of a
Cinna, and Marius made cruel by exile, and Sulla, “fortunate” by your destruction,
and Caesar, merciful to the dead, you used to quake at every trumpet blast of civil
war; for in addition to the slaughter of soldiers perishing for you on both sides,
you had wept for the leading lights of your Senate, the heads of consuls were stuck
on pikes, Catos forced to die, Ciceros mutilated and Pompeys unburied . . . Now
you have seen [vidisti] a civil war ended with the slaughter of enemies, a peaceful
soldiery, the recovery of Italy; and your liberation; you have seen [vidisti], I repeat,
a civil war ended for which you can decree a triumph.
The visual idiom is emphatic. To this scene of Rome’s own supernal
spectatorship succeeds, in the form of a recusatio, the spectacle of Theo-
dosius’s triumphal entry as seen from the street. But of what then actually
took place at Rome (., quae Romae gesta sunt), others would more fitly

 Pan. Lat. ..–; trans. Nixon and Rodgers :  (modified).
 mark vessey

speak who could call the city and its joys their own (propria). The visitor,
meanwhile, would return to farthest Gaul, there to discourse of wonders
(miracula). For he was now a qualified eyewitness, if not of great deeds
then certainly of the presence of a great person in a great place on a his-
toric occasion. He was one who could say: “I have seen Rome; I have
seen Theodosius; and I have seen them both together” (., Romam vidi,
Theodosium vidi, et utrumque simul vidi). In the last breath of his speech,
as he turned again to address the emperor himself, Pacatus’s anticipatory
narrative of living, oral testimony likewise took a turn toward a longer,
“literary” future:
The inhabitants of distant cities will flock to me; every pen will receive from me
the story of your exploits in due order; from me poetry will get its themes; from
me history will derive its credibility. Although I myself have said nothing about
you which is worthy of being read, I shall compensate for this injury I have done
you, Emperor, if I furnish materials for those who are read.
A copyist added (at the author-compiler’s behest?), “Here ends the pane-
gyric of Latinus Pacatus Drepanius, spoken to our lord Theodosius in the
eternal city of Rome.”
As Pacatus was the only one of the canonical later Latin panegyrists to be
able to brag of seeing the emperor at Rome, so he is the only one to end his
speech with what sounds like a routine committal of the panegyrical matter
in hand to the pens of future poets and historians. Routine as it may have
been, that final gesture acquires a certain poignancy from the placing of
Pacatus’s speech at the end of the modern sequence of the Panegyrici Latini.
For we are sure to ask, with all the ignorance and wisdom of hindsight:
Who would these later poets and historians be?
The most obvious poetical contender is Claudian, who flourished soon
afterward at the court of Honorius in Ravenna and would often have
cause to commemorate the deeds of that emperor’s father, Theodosius. We
should, however, be careful of generic boundaries, especially when we see

 Pan. Lat. ..: Ad me longinquae convenient civitates, a me gestarum ordinem rerum stilus omnis
accipiet, a me argumentum poetica, a me fidem sumet historia. Compensabo tibi istam, imperator,
iniuriam si, cum de te ipse nil dixerim quod legendum sit, instruam qui legantur; trans. Nixon and
Rodgers :  (modified).
 Cf. the converse formula by which Ammianus, having recounted the death of Valens at the battle
of Hadrianople, entrusts the composition of contemporary history to practitioners of a “higher”
style, meaning panegyrists (..); for this interpretation see Paschoud ; also G. Kelly .
The Historia Augusta offers both a version of this disclaimer (Carus et al. .) and a variant in
which the present writer, himself a historian, furnishes materials to be worked up more expertly
by his successor (ibid. .–, cited below, p. ). For instances of the latter type of profession in
Olympiodorus and Cassius Dio, see Matthews :  n. .
 Reinventing history 

them start to shift. While he filled the role as a poet, Claudian was still
a panegyrist, speaking of and for the present moment. His meter and his
diction might be those of epic, but not his “argument,” not at any rate
in the sense of Pacatus’s phrase, which surely hinted at a poem in at least
twelve books. Despite the formal difference between verse-panegyric and
the more customary prose performances of the genre, Claudian’s discourses
on state occasions continued the larger series of panegyrici Latini. To meet
the terms of succession laid down in Pacatus’s epilogue, any later-coming
writer on Theodosius and his times would have to escape the “in-built
obsolescence” of the panegyrical genre.
Failing poets, what would future historians have to offer? Here our
hindsight is obstructed by the loss of works whose coverage and dates of
composition can now only be guessed, such as the respective Historiae of
Sulpicius Alexander and Renatus Profuturus Frigeridus, both excerpted by
Gregory of Tours in book  of his history of the Franks, or the Annales of
Nicomachus Flavianus (which was dedicated to Theodosius and so unlikely
to have included that emperor among its subjects). Like his friend and
ally Symmachus, Flavianus was probably one of Pacatus’s auditors at Rome
in . Presumably also present that day was Sextus Aurelius Victor, the
current (or immediate past) prefect of the city, who three decades earlier
had composed an imperial-biographical history of Rome from Augustus to
Constantius II. A few years after the death of Theodosius in , another
potted imperial history of similar type would be penned by someone from
the Roman senatorial milieu. This work used the Caesares of Aurelius Victor
as a source and was for a long time erroneously thought to be his too. It
covered the whole imperial period, from the battle of Actium to the death of
Theodosius. Its last chapter often runs parallel to passages in the panegyric
of . Of Latin historical narratives still extant or known to have existed,
only this pseudo-Aurelian work De vita et moribus imperatorum, otherwise
misleadingly known as the Epitome de Caesaribus, can now incontestably
stand for the class of writings that Pacatus once imagined as the natural
inheritors, outside the panegyrical genre, of his Theodosian materials.

 Alan Cameron :  (absence of Theodosian epic),  (continuity with tradition of Latin
panegyric).
 MacCormack : .
 For Alexander and Frigeridus, see the summary in Zecchini : – (“the last exponents of a
doomed noble tradition. Perhaps continuators of Ammianus . . .”). For Flavianus, see Birley :
–. But note the strong cautions of Burgess () and, already for the Annales, Bardon (–,
vol. : –).
 Trans. with an intro. and commentary by Bird (). The dates of his office-holding are disputed.
 mark vessey

The pseudo-Aurelian history ends with the death of Theodosius at Milan

on January , , and the conveyance of his body to Constantinople for
burial: Corpus eius eodem anno Constantinopolim translatum atque sepultum
est (.). The work’s latest editor tentatively dates its publication between
 and , a time of mounting crisis for the empire in the west. A tradi-
tion of serial imperial biography as old as Suetonius that included Marius
Maximus in the early third century and that had enjoyed a notable flores-
cence in the mid-fourth, where it is represented by the lost Kaisergeschichte,
by the Caesares of Aurelius Victor, and by the imperial sections of the
Roman histories ab urbe condita of Eutropius and Festus, comes to an end –
at Rome, which, we must suppose, had always been its most natural Sitz im
Leben – with pseudo-Aurelius, De vita et moribus imperatorum. On the
world scale of culminating historiographic bangs, this is but the faintest
whimper. As Shakespeare had Octavian say, at the other end of Roman
imperial time, “The breaking of so great a thing should make / A greater
crack.”
Even were the De vita et moribus more precisely datable, it is doubtful
that many scholarly conferences would have been held to mark its sixteenth
centenary and the “end” of Roman history in the west. For one thing, we
are too distracted by the prodigies of the s. One of those prodigies
has been mentioned already: Claudian, the Greek poet from Egypt who
marked his debut as a Latin versifier with a panegyric on the consulship
of Probinus and Olybrius, delivered at Rome in January of , only days
after the death of Theodosius in Milan. With Claudian, the history of
Rome ascends again to myth. We can put our finger on the place where it
happens, when the personified Roma betakes herself to her winged chariot,
attended by her slaves Shock and Awe (Impetus and Metus), to go and salute
Theodosius after his defeat of the usurper Eugenius at the River Frigidus,
and entreat him to confer the next year’s consulship on the sons of Probus
(Prob. f ). Like Pacatus six years earlier, Claudian was speaking – in his
case, reciting – at Rome, to a senatorial audience that would have been
gratified to think that the consular naming of years might still be in Rome’s
keeping, even if her annals no longer were. (The suicide of Nicomachus
Flavianus at the River Frigidus would surely have put an end to any last
hopes on that score.) Claudian’s Roma helped seal the fate of Rome as the
once real but henceforth only mystical center of the world: “She is . . . the
personification of the city of Rome, no more and no less. And how faithfully
 Festy : lvi. Cf. Liebeschuetz : , dating the last gasp of “classicising [i.e., pagan, imperial]
history” in the Latin-speaking world “around .”
 For the Kaisergeschichte and related problems, see now Burgess .
 Reinventing history 

she mirrors the sorry state of Rome in the late fourth century, a city of
memories, accorded the honour of a visit by its Emperor but three times in
 years!” While the impeccably well-spoken provincial Pacatus could
still defer to a Roma securely implanted on her seven hills, and to the
city’s aristocracy as proprietary witnesses of the imperial triumph staged in
their midst, Claudian, who had traveled further and whose Alexandrian
accent must have been distinctly audible to his senatorial audience, seems
instinctively to have known – unless, as Cameron suggests, his practice of
the Greek genre of city-panegyric enabled him more readily to see – that
Rome’s brightest future was as an infinitely mobile creature of the poetic
imagination.
The second of the marvels to divert the modern eye from pseudo-
Aurelius is the Historia Augusta, now also generally ascribed to this decade
and to a single impostor. As Claudian turns current events into myth, so
the scriptor threatens to reduce the entire process of Roman historiography
to parody. The final installment of his history loops back before its starting
point to review the fortunes of the Roman commonwealth from the origins
of the city (ab ortu urbis). As if to abridge the breviaria of Eutropius
and Festus, the already “totalizing” history of the corpus Aurelianum, and
(for all we know) the Annales of Nicomachus Flavianus too, “Vopiscus”
recapitulates the story from Romulus to the late third century in a few
declamatory sentences, then hurries to the high point of the reigns of the
emperors in hand: a set of games given at Rome, the star turns at which
had been immortalized in a painting on the Palatine “near the portico of
the stables” (HA, Carus et al. .). And so, with as much dignity as our
author will ever muster, to an epilogue:
And now, my friend, accept this gift of mine, which, as I have often said, I have
brought to the light of day, not because of its elegance of style but because of its
learned research [non eloquentiae causa sed curiositatis], chiefly with this purpose
in view, that if any gifted stylist should wish to reveal the deeds of the emperors,
he might not lack the material, having, as he will, my little books as ministers to
his eloquence. I pray you, then, to be content and to contend that in this work I
had the wish to write better than I had the power. (Ibid. .–)
Whatever rivalry or complicity there may have been among Latin-writing
Roman historians of the mid-to-late-fourth century, it had now become a
 Alan Cameron : . Seeck () records visits for Constantine (, , ), Constantius II
(), Gratian (), and Theodosius (, ); on “absent emperors” and the ways in which they
could still be made present in Rome in the fourth and fifth centuries, see now Humphries ;
: –.
 Trans. adapted from Magie ().
 mark vessey

contest not to be the (last) one to write a definitive Roman history! Like the
rest of the Historia Augusta, this terminal profession rings so false, we are at
risk of divining a secret truth within it. Despite the recent mild controversy
around them, the terms on which Ammianus ends his history – that other
historiographic prodigy of the s – are comparatively straightforward.
Among the synchronisms seen or proposed by Matthews in Western
Aristocracies is one that has Ammianus reading aloud his description of the
Roman adventus of Constantius II in  (..f.) “at a time very close
indeed to Theodosius’ visit” in . The epilogue to Ammianus’s Res
gestae (..) and that of Pacatus’s panegryic to Theodosius (., quoted
above) are in any case perfectly reciprocal. Each allows explicitly for the
genre practiced by the other. The last of the great Roman historians may
even have had in mind the special discretion recently exercised in the last,
as it proved, of the Panegyrici Latini. Anyone who told the story of Rome
in the future, he insinuated, would have to bite his tongue, suppress the
truth, and tell lies – as panegyrists (and poets) were known to do. Anyone,
that is, who was not a Christian.

beginnings (constantinople, ca. 381)

From the [twentieth] year of Constantine’s reign, down to the sixth consulship of
the emperor Valens and the second of Valentinian, his co-Augustus, the work is all
my own. There I am happy to stop, saving the remaining period under Gratian and
Theodosius for more expansive historical treatment [latioris historiae stilo]. This
I do, not because I would be afraid to write freely and truthfully of the living –
the fear of God drives out the fear of one’s fellow human beings – but because,
with barbarians still cavorting in our land, all things are in a state of uncertainty.
Unlike Ammianus and other late-fourth-century Roman historians who
“signed” their discourse at the end by alluding to its possible sequels in dif-
ferent hands, Jerome turns the conventional profession of historiographic
limits into a prefatory topos. The words quoted are the last of his intro-
duction to a Latin translation and continuation of Eusebius’s Chronici
canones, produced at Constantinople within a few years of the disaster at
Hadrianople. Eusebius (after Julius Africanus) had pioneered the genre
of Christian universal chronography two generations earlier. Jerome gave
it the currency it was to have in after-ages. Viennent enfin les chroniques,
beginning with Jerome.

 See above, n. .  Matthews : ; see also Matthews : .
 Jerome, Chron. pref. (ed. Helm) .
 Reinventing history 

Now that the historiographical impact of “Jerome’s Chronicle” is at last

being correctly appreciated, there is more reason not to overestimate
the complacency with which its deviser, as translator and literal auctor or
enlarger of another man’s work, first offered it for copying and reading
outside his own household. In the roughly chronological list that Jerome
was to publish of his written production down to /, the Chronicle
would appear in fifth place. By then he had a “life’s work” to advertise
and a definite shape to impart to it. A connoisseur of Roman (Varronian,
Suetonian) traditions of author-portraits (imagines) and collective biogra-
phy de viris inlustribus, Jerome raised his own monument long before he
was ready to be inhumed. Posterity would only embellish it. Jerome has
been reputed a saint, a founder of the Middle Ages, a father and doctor of
the Latin Church, and the “author” of the Vulgate Bible. At the beginning
of the s, he was a wandering priest in the entourage of an ousted bishop
from Antioch. Socially well connected, doubtless admired by some for
his rhetorical prowess, he was still on any reckoning a minor figure in
the crowd of litigants, lobbyists, and aspirants then swarming the eastern
capital, as Theodosius established his court there.
Scholarship with its long views can explain why Jerome produced the
Chronicle in Constantinople, ca. . Reading the preface, we may be struck
by how little preliminary sense the author himself could make of the project
at the time. The work’s auspices are vague, its dedicatees – friends of his
called Vincentius and Gallienus – mere shadows; had this been a presti-
gious commission, we should know more about the sponsors. Translation
is always difficult, Jerome reminds the reader: difficultas, difficile, and dif-
ficilius appear altogether five times in barely fifty lines. The task of the
translator is even harder when, as in the present case, he is working by
dictation and in a hurry. Biblical translation (which this was not) can be
especially demanding: consider the Septuagint and other Greek versions
of the Hebrew scriptures or a Latin Homer, for that matter. Quorsum ista?
These points were made lest anyone complain about the quality of the
work to follow – which consisted entirely of short notices in prose. Well,
there would always be envious critics! Only after exhausting his declama-
tory themes does the prefacer supply the information a typical reader might
actually be imagined to seek. Most of the work had been translated directly
from Eusebius; Jerome had made selective additions for the period from
the fall of Troy to Constantine’s vicennalia, where Eusebius had finished;
 Croke ; Burgess , ; Jeanjean ; Goetz .
 Jerome, Vir. inl., ed. Richardson : ; Rebenich : –.
 Vessey : –.  Rebenich a: –.
 mark vessey

the final section was all his own (totum meum). For the sequel, he could
not speak. No more, it seems, could he say directly why he had taken the
story as far as he had.
What kind of a beginning was this? And of what?
Jerome professed to be a Roman historian. Eusebius had written in Greek
for Greeks, skimping on Roman history as such. Jerome would make good
the deficit for Latin readers. But what did it mean to write “Roman”
history, in either language, then or at any time in the recent past? Jerome’s
answer to that question would necessarily depend on the histories he himself
was used to reading or (if there was any difference) on those he had to hand
when expanding upon Eusebius. As to his sources for the Chronicle, he is
disarmingly clear: for the period already covered by his Greek exemplar,
he had transcribed his new materials carefully “from Suetonius and other
distinguished historians,” de Tranquillo et ceteris inlustribus historicis.
We could excusably infer from this statement that Jerome had eked out
Eusebius’s notices for the earlier imperial era with data from Suetonius’s
Caesares, and that his other authorities were likely to have included a
string of “Suetonian” biographers from later periods, writers with access
to court and high senatorial circles – men who had had the good fortune
or sense, like Suetonius or Gibbon’s prudent “historian of his own times,”
to reside at Rome, where reliable documents and informants were most
readily available.
But the Caesares was not among Jerome’s immediate sources. (His actual
debts to Suetonius, we shall see, are to the latter’s compilation of lives of
famous literary figures, the De viris inlustribus.) Nor is it possible to name
any historian, apart from Suetonius, whose work he can be shown to have
consulted firsthand. For Rome’s remotest history and prehistory, Jerome
appears to have relied on a work similar to the (anonymous) Origo gentis
Romanae; for the sequel down to the end of the republic, on an epitome
of Livy; and for subsequent political history, on “a now-lost epitome of
imperial biographies starting with Augustus that enjoyed remarkable pop-
ularity in the second half of the fourth century, being used by every Latin
historian to survive from that period.” The last of these abridgments, the
fabled Kaisergeschichte, was probably extended several times in the course

 Chron. pref. (ed. Helm) : nonnulla . . . adieci, in Romana maxime historia, quam Eusebius huius
conditor libri non tam ignorasse ut eruditus, sed ut Graece scribens parum suis necessariam perstrinxisse
mihi videtur.
 Ibid.
 Burgess : . For Jerome’s sources in the Chronicle, see also Burgess , conveniently
summarized in Burgess : –.
 Reinventing history 

of the fourth century. R. W. Burgess has recently suggested that Jerome’s

lower terminus of  may reflect the limit of its (then) latest updating.
This conjecture makes even better sense if, as has also been argued, the
Kaisergeschichte by this time formed the end section of a unified corpus of
Roman historical texts that also comprised the other “sources” just listed
for Jerome’s Chronicle. We could then say that Jerome’s Roman history –
in the literal, bibliographical sense of his history book – was one that any
Latin reader of his time who cared about such things was bound sooner or
later to consult, if not possess. By the early s, such a history book was as
likely to come to hand in Constantinople as anywhere else in the Roman
world. The main service rendered by Jerome in his version of the Chron-
icle would have been to bring the contents of this standard Latin-Roman
history within the more comprehensive chronographical structure devised
by Eusebius.
Jerome, however, did more than merely Eusebianize the historia Romana
vulgata. He also gave his own inflection to the Roman sense of what counted
as a historical event by consistently presenting Latin writers, teachers, and
orators as persons whose actions, however petty, belonged in the same
historical space – literally, in the same portion of the tabular chronicle
page, the spatium historicum as it has been called – as the res gestae of those
more typically conspicuous actors in Rome’s history, its rulers and military
leaders. In a turn of phrase for which the tradition of the text in question
gives no warrant whatsoever, Peter once wrote of Jerome’s complementing
Eusebius “from Suetonius’ work ‘De viris in literis [sic] inlustribus.’”
The interpolation of Suetonius’s (presumed) title was Jerome’s in spirit
before Peter made it in the letter. Other chronographers, including latterly
Eusebius, had inserted the life dates of famous philosophers, poets, and
other writers into their synoptic histories. Jerome took the process a step
further. From the start, and with special emphasis as it advances into his
own, post-Eusebian time, his chronicle provides the abstract of a Latin
literary history of Rome.
Having allowed that Jerome’s chief non-Eusebian source for the Chron-
icle could have been a ready-made compendium of Roman histories, we
should not be too surprised to discover that his one overt reference to it,
under the name of “the Latin history,” occurs as he assigns the floruit of

 Burgess .  For Eusebius’s innovations, see now Grafton and Williams : –.
 Peter , vol. : .
 Eusebius, Chron. pref. (trans. Jerome, ed. Helm) : qui de inlustribus viris philosophi poetae principes
scriptoresque variorum operum extiterint . . .
 mark vessey

Homer to an epoch in the pre-Roman history of Latium. The dating of

Homer (long after Moses!) had been a critical issue for Eusebian apologet-
ics, but nothing of the sort is at stake in the present reference. Flourishing –
flourished, that is, by Jerome –  years before the founding of the city,
this textual figure of Homer instead presages, and in a sense already inau-
gurates, the line of nonmilitary, Latin literary heroes (or antiheroes?) for
whom places were to be found in Jerome’s annals. The Latin Homer of
the preface, it turns out, was more than a declamatory diversion; it was the
index to a peculiar program of cultural translatio to be carried out by Jerome
under cover of Eusebius’s already powerfully tendentious and teleological
historical comparatism. As the great kingdoms of the world would one day
be subsumed in the empire of the Romans, an event signaled in the scheme
of the Chronicle by the amalgamation of their previously parallel histories
into a single strand or filum after the sack of Jerusalem by the emperor
Titus, so their most precious intellectual goods – essentially, the complex
inheritances of Greek and Hebrew literature – would now be inventoried
in a Latin history, drawn up in Constantinople at the beginning of the
reign of Theodosius. If an analog to this project of cultural thesaurization
is sought within the compass of the Chronicle itself, it will be found most
obviously in the Greek translation of the Hebrew scriptures (the Septu-
agint) made at the behest of King Ptolemy Philadelphus, who is said to
have added the work to the library that he had assembled at Alexandria
“out of every kind of learned writing,” ex omni genere litteraturae.
Latin literary history begins with Jerome in Constantinople. Since, with
a few important exceptions, Suetonius’s lives of the Roman poets are now
beyond recovery, Jerome’s notices and datings in the Chronicle supply the
indispensable basis for any modern chronology of those writers. The same
goes for his mentions of historians, philosophers, and orators of the classi-
cal period, for which a Suetonian source can usually also be assumed. Only
in the cases of grammarians and rhetoricians, for which Suetonius’s work
is still largely extant, is Jerome technically redundant. Such accountings
as these, however, do less than justice to the chronicler. It is not just as
the excerptor of a lost Suetonius that Jerome becomes, as if by accident,
the protohistorian of Latin letters. Even had Suetonius’s work survived in

 Chron. c: In Latina historia haec ad verbum scripta repperimus: Agrippa apud Latinos regnante
Homerus poeta in Graecia claruit. For the interpretation of this passage, see Burgess :  n. 
and refs. given there.
 Chron. f., echoing Tertullian, Apol. .: Ptolemaeus . . . eruditissimus rex et omnis litteraturae
sagacissimus.
 Helm ; Suerbaum : ; Vessey .
 Reinventing history 

its entirety, Jerome would still have deserved to be seen as an innovator in

literary historiography. Suetonius arranged his biographies of eminent men
of letters chronologically, to convey a sense of the historical development
of their disciplines or professions at Rome. He treated each discipline in
a separate book (De poetis, De philosophis, etc.) and assembled the books
into a distinct, higher-order historiographical enterprise, De viris [in literis]
inlustribus. Jerome, obedient to his Eusebian model, overrode both Sueto-
nian principles of distinction: in the Chronicle, practitioners of different
literary disciplines follow each other in a single file, intermingled with their
peers from other, nonliterary professions. And even as he undoes Suetonius,
Jerome subverts his own Greek source. These Latin literary performers are
found acting and suffering in his Roman history in ways incompatible with
the brevity and decorum of Eusebius’s notices on their Greek counterparts,
which normally specified only the genre of their performance. Thus we
learn from Jerome that Ennius lived frugally on the Aventine with just one
maidservant (a), that Plautus was once reduced to working in a mill
(h), and that Lucretius went mad and killed himself after swallowing
a love-philter (g). There is a good deal more of this kind of gossip
about what Latin writers did during their lives, along with fairly frequent
information on the circumstances of their deaths and inhumations.
When Ennius expired from joint disease aged over seventy, he was buried
in the tomb of the Scipios on the Via Appia, or so it was generally thought
(Chron.a). Modern studies of the politics of Latin literature in the
classical period have paused at the monuments of Ennius and other early
poets without always noticing that Jerome was there before them. Only
the burial arrangements of emperors receive fuller coverage in his Chronicle
than the obsequies of the poets. Tempting as it is to imagine Jerome as a
student in Rome, visiting the tombs of legendary figures from the literary
past like Ennius and Caecilius Statius (b), his references to them are
at least as likely to be derived from prior texts as from personal memory,
whatever he may, in fact, have seen and recalled. (His transcription of the
epitaph of Virgil at Naples [h] is presumably straight from Suetonius.)
A similar point could be made for his numerous references to famous
buildings in the city. Renaissance antiquarians might use the Chronicle as
a guidebook, but it could have had no such utility in the fourth century.
There is a monumental landscape implicit in Jerome’s Chronicle, and its

 Bloomer : –; see also Habinek .

 Burgess (: ) assumes his use of a distinct source providing “detailed information on imperial
building projects.”
 mark vessey

most heavily built-up area is that of Rome itself, but it is a landscape already
fully textualized and portable.
In Jerome’s account, the foundation of Constantinople was achieved by
the stripping of other cities (Chron. f.). His Latin redaction of Eusebius’s
universal Greek-Roman history likewise founds a new Rome, but a textual
or virtual one, and in doing so strips the real Rome of its material presence.
Within the confines of the emergent other city of the Latin Chronicle, the
funerary monuments of ancient Latin writers occupy a certain amount of
ground. The rest of the space that Jerome claims for them is filled by the
recital of their res gestae, trivial and devoid of genuinely literary-historical
interest though they mainly are. By simultaneously exploiting the most
ephemeral and the most monumental registers of his Suetonian and other
classical sources, Jerome may have hoped to attract a readership that was
already captive, or so Ammianus complained (..), to the specious
urbanities of Juvenal and Marius Maximus. Despite all the differences, the
mock-antiquarian scenography of the Latin Chronicle can also be seen to
anticipate the Historia Augusta.
If there is one place in the Chronicle where Jerome might be supposed to
bear personal topographic witness, it is in the entry for the year , when
he writes of the rhetorician Victorinus and the grammarian Donatus, both
famous at Rome, the latter his own teacher there. Victorinus, he adds, had
been honored with a statue in the Forum of Trajan, the literary-cultural
hub of the orbis Romanus for as long as Rome itself could be imagined
the center of anything. The context in the Chronicle is as symbolically
loaded as the physical location. The entry itself falls in the middle of a
distinct cluster of literary-historical records for the years –, which in
turn marks the midrange of the second major, semi-continuous series of
such notices from Jerome’s hand, following that generated by him from
Suetonius for the period from Ennius to Quintilian. This later series is
nearly coterminous with the post-Eusebian section of the Chronicle and
hence with the post-Constantinian pax Christiana, interrupted only by
the reign of Julian: in short, this is the period within which Jerome’s own
floruit must fall.
The series opens with an entry for Lactantius, the description of whose
lifestyle – he was “so poor, there were times he lacked even the necessities”
(d) – seems calculated to recall the impoverished Roman poets of yore.
There follows a string of notices on rhetoricians, poets, and others, almost
 Chron. e: Victorinus rhetor et Donatus grammaticus praeceptor meus Romae insignes habentur.
E quibus Victorinus etiam statuam in foro Traiani meruit. The same monument (or Jerome’s evocation
of it) would also strike Augustine’s eye: Conf. ..; Vessey : –.
 Reinventing history 

all of them Latin, for the years – (c, g, d, e, h, k,
l, m). After an interval of nearly twenty years, dotted with a single
literary-historical entry (d), the line of Latin orators and rhetoricians
picks up again with three figures celebrated at Rome, including Victorinus
(a, b, d), and two active in Gaul (h). Although the nature
of Jerome’s sources for this miniature set of viri inlustres – almost all of
them renowned for their eloquence, identified as Gallic and/or as having
professed at Rome – cannot be known, the resemblances between his list
and those presented by the index to the Panegyrici Latini and Ausonius’s
Commemoratio professorum Burdigalensium are strong enough to suggest
that he was working with something like a canon. These men were the
Latin literary stars of the generation or so before Jerome’s own. Whatever
fame he might aspire to, theirs would be the example he had to reckon
with. As we have seen, he positively invites the comparison. The entry for
 puts him at Rome as squarely as the statue of Victorinus. We may even
suspect that the statue is there for that purpose: less to reestablish Trajan’s
Forum as the physical center of the literary universe than to place Jerome
himself at the ideological center of his own times.
The sequel confirms as much. The years – mark the onset of a new
phase in the Arian controversy, associated with the exile of intransigent
Latin churchmen by Constantius II (f, i), and the accession of Julian
as Caesar. The scene has now been set for the formative action of Jerome’s
lifetime. The next notice (b) is his cue to enter upon it:  saw the death
of the hermit Antony, who had often spoken well of his ascetic predecessor
Paul of Thebes, who in turn . . . was already the subject of a work by Jerome!
The Vita Pauli is referred to in the Chronicle as a brevis libellus de exitu Pauli,
“A Short Death of Paul.” The description fits. Most of the text is taken
up with a narrative of the hermit’s demise, followed by the inhumation of
his corpse by obliging lions. At the end, Jerome took his stand on Paul’s
rough burial mound, fiercely contrasting it with the overburdened tombs
(operosa sepulcra) of the Roman aristocracy of his day. Some years later, in
a story told in a letter to a young noblewoman at Rome, he would present
his own Christian-ascetic literary vocation as occurring at a moment when
his friends were ready to bury him. The dynamics of that episode, with
its violent confrontation of secular and divine scriptures, are explicable
only in terms of the new option for white martyrdom suggested by the
emperor Julian’s short-lived attempt to prohibit the double profession

 Jerome, VPauli  (PL  [], cols. –).

 Jerome, Ep. ..: Interim parabantur exsequiae (CSEL : ).
 mark vessey

of Christianity and Hellenism – an option already clearly touted in the

Chronicle’s notice on the sophist Prohaeresius. As the prospects of dying
heroically for Christ receded into a pre-Constantinian past, the imagination
of near-death experiences seems to have grown more pressing in Christian
writing. Jerome was not alone in his mortifications. That said, few literary
beginnings of the time can have been as lavishly sepulchral as his in his
own telling, or as gravid with futurity.

As Ammianus is the last great Roman historian, the beginning of the end
of Roman historiography falls circa . At that point, we may speculate, a
late antique reader or copyist joined a preexisting compendium of events
of the previous millennium to the more ample narrative supplied for the
following twenty-five years by the last eighteen books of Ammianus’s Res
gestae. The splice, if it was made, would have been assisted by the capsule
history of the Roman race inserted by Ammianus, at ..–, to justify his
generally negative treatment of the affairs of the city of Rome itself. Roma,
as a personification of all that was manly and just in the conduct of the
Roman people, may have been fated to endure and triumph for all time, as
her most famous Latin poet had claimed (..), but the manners of those
who now inhabited the place of that name were so vicious as to prevent
anything worthy of record ever being enacted there (..). In a much-
quoted passage, Ammianus pours equal scorn on nobles who maintained
troupes of dancing girls while keeping their private libraries shut up like
tombs (.., bibliothecis sepulcrorum ritu in perpetuum clausis) and on
the common folk who had no time for anything but chariot races. Roman
history, we are given to understand, was now carried on elsewhere. And
so, at the end of this digression, the historian returns to the real matters
in hand, east and west: Ergo redeundum ad textum (..). Taken at his
word, Ammianus – even as a Roman historian after Gibbon’s heart, writing
at Rome – was already writing after Rome. The city itself was no longer
a theater, let alone the theater, of his action; at best it was an occasional
sideshow. Granted, Ammianus wrote as an outsider, albeit one with the
advantage of good connections; this chapter of his history is thick with
the experiences of peregrini or would-be-resident aliens. But was not an
outsider more likely to see the truth in such a case?
 Chron. f.: Prohaeresius sofista Atheniensis lege data, ne Christiani liberalium artium doctores essent,
cum sibi specialiter Iulianus concederet, ut Christianus doceret, scholam sponte deseruit. The sophist’s
refusal of the emperor’s waiver is the prototype of Jerome’s spontaneous literary-professional mar-
tyrdom, which laid down a law for his own and future generations; Vessey : –.
 For a caution against too quickly conflating Ammianus’s own experience with that of the outsiders
mentioned in his text, see now G. Kelly : –.
 Reinventing history 

Jerome’s emotional attachment to Rome is well documented. When the

Goths sacked the city in , no one lamented more loudly than he. For
all that, he made a career, indeed a profession, of being somewhere else.
Not until after his Roman banishment and departure for Palestine in 
does he appear to have won any kind of general reputation, and from early
on that reputation was reflected in language that explicitly made Jerome,
rather than Rome, the center of a (Christian) Latin literary universe.
Our analysis of the Chronicle has shown how adroitly he himself proposed
that substitution, at a juncture exactly coinciding with Ammianus’s “first”
Roman digression, in a text that played on the places of the city’s literary
past while implicitly describing a space of literary production that would be
independent of any geographical coordinates. In a book written in Istanbul
at the beginning of the s, reflecting the crisis of a Western cultural,
spiritual, and literary tradition whose origins could be plausibly traced to
Latin late antiquity, Erich Auerbach presents Gregory of Tours – the first
post-Roman writer in a chronological array extending from Homer and
the Hebrew scriptures to Virginia Woolf – as doomed to have no unifying
historical point of view, because “no longer situated in a place where all the
news from the orbis terrarum [was] received, sorted, and arranged according
to its significance for the state.” It is likely that Auerbach’s preoccupation
with Christian typological exegesis, and specifically with its powerlessness
to impart meaning to “vast regions of [historical] event” once accounted
for by classical historiography, prevented him from seeing how heavily
Gregory was indebted to the Latin chronicle tradition for the organization
of data for his history of the Franks, including those that he derived from
classicizing Roman historians who had written after Ammianus. For all
that he used the Historiae of Sulpicius Alexander and Profuturus Renatus
Frigeridus in book  of his history, Gregory took his initial historiographical
coordinates from Eusebius-Jerome. Auerbach imagined that the fifth-
century fall of Rome would decenter all Western literary representations of
reality. He failed to register the impact of a post-Roman reorientation of the
Latin literary universe engineered in the first place by another expatriate at
Constantinople, circa .
Jerome’s Eusebian replotting and extrapolation of Suetonius’s Latin-
Roman literary history – entailing, as we have seen, a Hieronymian recast-
ing of Eusebius’s Christian-Roman universal history – did not terminate
with the Chronicle. It would achieve larger expression in an even more

 Jerome, Ep. ; Rebenich .  Vessey .  Auerbach : .
 Ibid.: .  Gregory of Tours, Decem libri historiarum  pref. , ;  pref.
 mark vessey

radically innovative work of historiography, fit to rank with such

other Rome-translating prodigies of the s as Claudian’s panegyrics,
Ammianus’s Res gestae, and the Historia Augusta. Jerome’s chronological
directory of Christian writers and their works, the De viris inlustribus or
(as he himself preferred to call it) De scriptoribus ecclesiasticis, would stand
alongside his Chronicle as one of the standard historiographic resources of
the Latin Middle Ages. Compiled in Bethlehem “in the fourteenth year
of the emperor Theodosius,” with material from Eusebius’s Ecclesiastical
History and a preliminary nod to classical models offered by Suetonius
and Cicero’s Brutus, the catalog of Church Writers allowed full develop-
ment to principles of Christian literary production already discernible in
the Chronicle, both scriptural (Origenian-Eusebian) and pseudo-martyrial
(neo-Julianic). It was the closest Jerome would ever come to writing the
church history mooted in the preface to his Vita Malchi, and it was also a
fulfillment of the prefatory pledge and disclaimer of a latior historia given
earlier at Constantinople.
Michel Foucault once suggested that it would be worth discovering “at
what point we [in the West] began to recount the lives of authors rather
than of heroes,” and he already fingered Jerome as partly responsible for
the shift. Jerome’s mobilization in Latin discourse of a new cadre of actor
and agent in Roman history, the scriptor ecclesiasticus, came about at a
moment when – as Ammianus’s experience shows – the stylistic challenge
of sustaining a classical historical narrative “connecting the eternity of
Rome with human will and effort” was becoming almost insuperable.
Could Roman history still be written after the death of Julian or in the
aftermath of Hadrianople? In the short term, it was the panegyrists who
had to answer that question. It was their responsibility to speak in the
presence (or absence) of the emperor and to make “vast regions of event”
revolve once more around his sacred person – if not at Rome, then wherever
“Rome,” as embodied in the emperor, now notionally stood. Such risks as
Jerome himself might once have been willing to take in this regard had
been dispelled at Trier on the northwestern frontier, around the year 

 McKitterick : –, –. Gregory of Tours already signed off in the new Hieronymian
idiom: Decem libri historiarum . (“I, Gregory, have written the ten books of this history”),
echoing Jerome, Vir. inl.  (“I, Jerome, have written these [works].”).
 Jerome, Vir. inl. pref.  Foucault : , –.
 Matthews : ; see also footnote  in this chapter. For a reading of the Res gestae in which
books – reveal the mounting difficulty experienced both by emperors in imposing unity on
the empire and by the Roman historian in imposing unity on his narrative, see C. Kelly , esp.
–, citing Matthews : , –. On this view, the Romans risked “being stranded on the
margins of their own history” () as early as the s.
 Reinventing history 

when, by his own testimony, he abandoned his quest of a civil service

career. (What did befall him on the “semi-barbarian banks of the Rhine,”
as both he and Symmachus called them, and in what company? Could we
construe it, his decision not to include either Symmachus or Ausonius in his
prosopographies might prove as revealing as Ammianus’s silence in respect
of one or Augustine’s studious mention of the other.) Jerome – unlike such
Latin Christian contemporaries of his as Augustine and Paulinus of Nola –
would never extend the line of later panegyristae Latini, never place himself
directly in a tradition of Roman state eloquence derivable through the
younger Pliny from Cicero, Brutus, and more remote republican forebears.
Whereas their profession, as old-style orators and historians of the moment,
ultimately presupposed the presence of the emperor at Rome, his, as a
new-style scriptor ecclesiasticus intent on the sacred letters of a divine law
code, was predicated on absence of and from both the emperor and (his)
city. Jerome’s closest counterparts, in this last respect, were the drafters
of the Gesta senatus by which the Theodosian Code was received (from
Constantinople) and promulgated at Rome in .
More than four decades earlier, as we learn from pseudo-Aurelius, the
mortal remains of Theodosius I had been brought back from the west
to Constantinople, as if to guarantee that classical Roman and Byzantine
history would one day be seen to make a single, marmoreal whole. As
John Matthews has taught us, however, the later Roman historian needs
to have a nose for the rotting corpse as well as an eye for brilliant surfaces.
Whatever accidents of personal history and prior historiography may have
dictated the endpoint of the beginning of the Latin chronicle tradition,
the unrecovered body of Valens on the field of Hadrianople still speaks –
in Jerome’s report (Chron. c) – for volumes of post-Roman writing to
come.
 Jerome, Ep. ..; Symmachus, Or. .; Augustine, Conf. .., with Ebbeler and Sogno .
 Matthews : – (“Senatus Amplissimi Gesta”); Humphries : . See also Vessey :
– (“Manu divina: transmitting divine authority”) and, for the (manu)scriptural turn in Jerome’s
oeuvre, Vessey .
 References

Aldrete, G. S. . Gestures and Acclamations in Ancient Rome. Baltimore, Md.

Amato, E., A. Roduit, and M. Steinrück, eds. . Approches de la Troisième
Sophistique. Hommages à Jacques Schamp. Brussels.
Ando, C. . Review of J. F. Matthews, Laying Down the Law. Phoenix :
–.
Arjava, A. J. . Women and Law in Late Antiquity. Oxford.
. “Paternal Power in Late Antiquity.” JRS : –.
Ashbrook Harvey, S. . “The Sense of a Stylite: Perspectives on Simeon the
Elder.” VC : –.
Auerbach, E. . Mimesis: The Representation of Reality in Western Literature.
Trans. W. R. Trask. Princeton, N.J.
Auliard, C. . La diplomatie romaine: l’autre instrument de la conquête de la
fondation à la fin des guerres samnites (753–290 av. J.-C.). Rennes.
Ayres, L. . Nicaea and its Legacy: An Approach to Fourth-Century Trinitarian
Theology. Oxford.
Baldovin, J. F. . The Urban Character of Christian Worship: The Origins,
Development, and Meaning of Stational Liturgy. Rome.
Banaji, J. . Agrarian Change in Late Antiquity: Gold, Labor, and Aristocratic
Dominance. Oxford.
Barceló, P., and G. Gottlieb. . “Das Glaubensedikt vom . Februar :
Adressaten und Zielsetzung.” In Dietz and Hennig : –.
Bardill, J. . “The Golden Gate in Constantinople: A Triumphal Arch of
Theodosius I.” AJA : –.
Bardon, H. –. La littérature latine inconnue.  vols. Paris.
Barnes, T. D. . Constantine and Eusebius. Cambridge, Mass.
. The New Empire of Diocletian and Constantine. Cambridge, Mass.
. “Ammianus Marcellinus and His World.” CP : –.
. Review of S. Corcoran, The Empire of the Tetrarchs: Imperial Pronounce-
ments and Government, A.D. 284–324. AJPh : –.
Barnish, S. J. B. . The Variae of Magnus Aurelius Cassiodorus Senator. Translated
Texts for Historians. Liverpool.
Barsanti, C. . “Il foro di Teodosio I a Constantinopoli.” In Arte profana e arte
sacra a Bisanzio, ed. A. Iacobini and E. Zanini. Milion . Rome: –.


 References 

Bassett, S. . The Urban Image of Late Antique Constantinople. Cambridge.

Bauer, F. A. . Stadt, Platz und Denkmal in der Spätantike. Untersuchungen
zur Ausstattung des öffentlichen Raums in den spätantiken Städten Rom, Kon-
stantinopel und Ephesos. Mainz.
. “Urban Space and Ritual: Constantinople in Late Antiquity.” AAAH :
–.
Bayer, K., ed. . Vergil-Viten. Munich.
. Suetons Vergilvita: Versuch einer Rekonstruktion. Tübingen.
Benesevic, V. N., ed. . Syntagma XIV titulorum sine scholiis secundum versionem
Palaeo-Slovenicam, adjecto textu Graeco e vetustissimis codicibus manuscriptis
exarato. St. Petersburg. Repr. Leipzig ().
Berger, A. . Untersuchungen zu den Patria Konstantinupoleos. Poikila Byzantina
. Bonn.
. “Regionen und Strassen im frühen Konstantinopel.” Istanbuler Mitteilun-
gen : –.
. “Streets and Public Spaces in Constantinople.” DOP : –.
. “Imperial and Ecclesiastical Processions in Constantinople.” In Byzantine
Constantinople: Monuments, Topography and Everyday Life, ed. N. Necipoglu.
Leiden: –.
Bernardi, J. . La prédication des pères cappadociens: Le prédicateur et son audi-
toire. Paris.
. “Un réquisitoire: les Invectives contre Julien de Grégoire de Nazianze.” In
Braun and Richer : –.
ed. and trans. . Grégoire de Nazianze. Discours 4–5 Contre Julien. Sources
chrétiennes . Paris.
Bertrand, J.-M. . “A propos du mot provincia.” JS: –.
Bird, H. W., trans. and annot. . Aurelius Victor: De Caesaribus. Liverpool.
Birley, A. R. . “The Historia Augusta and Pagan Historiography.” In Marasco
: –.
Bispham, E. . From Asculum to Actium: the Municipalization of Italy from the
Social War to Augustus. Oxford.
Bloomer, W. M. . Latinity and Literary Society at Rome. Philadelphia, Pa.
Boeft, J. den, J. W. Drijvers, D. den Hengst, and H. C. Teitler, eds. .
Ammianus after Julian: The Reign of Valentinian and Valens in Books 26–31 of
the “Res Gestae.” Leiden.
Boissevain, J. . “Patronage in Sicily.” Man : –.
. Friends of Friends. Oxford.
Bonini, R. . Ricerche sulla legislazione giustinianea dell’anno 535: Nov. Justini-
ani 8: venalità delle cariche e riforme dell’ammistrazione periferica. rd edn.
Bologna.
Bonner, S. F. . Education in Ancient Rome from the Elder Cato to the Younger
Pliny. Berkeley, Calif.
Booth, A. D. . “Elementary and Secondary Education in the Roman Empire.”
Florilegium : –.
. “The Academic Career of Ausonius.” Phoenix : –.
 References

Boulluec, A. . “L’école d’Alexandrie. De quelques aventures d’un con-

cept historiographique.” In Alexandrina: Hellénisme, judaı̈sme et chris-
tianisme à Alexandrie, Mélanges offerts à Claude Mondésert. Paris:
–.
Bowersock, G. . Julian the Apostate. Cambridge, Mass. Repr. London ().
a. Hellenism in Late Antiquity. Ann Arbor, Mich.
b. Review of J. F. Matthews, The Roman Empire of Ammianus. JRS :
–.
. Fiction as History: Nero to Julian. Berkeley, Calif.
Bowersock, G. W., P. Brown, and O. Grabar, eds. . Late Antiquity: A Guide
to the Postclassical World. Cambridge, Mass.
eds. . Interpreting Late Antiquity: Essays on the Postclassical World.
Cambridge, Mass.
Bowes, K., and M. Kulikowski, eds. . Hispania in Late Antiquity: Current
Approaches. Leiden.
Boys-Stone, G. R. . Post-Hellenistic Philosophy: A Study of Its Development
from the Stoics to Origen. Oxford.
Brakke, D. . Athanasius and the Politics of Asceticism. nd edn. Baltimore, Md.
Brandes, W., ed. . Poetae Christianae Minores, I. Corpus Scriptorum Ecclesi-
asticorum Latinorum , . Vienna.
Braun, R., and J. Richer, eds. . L’empereur Julien I: De l’histoire à la légende
(331–1715). Paris.
. L’empereur Julien II: De la légende au mythe (de Voltaire à nos jours). Paris.
Brennecke, H.-C. . Studien zur Geschichte der Homöer: Der Osten bis zum Ende
der homöischen Reichskirche. Tübingen.
Broek, R. van den. . “The Christian ‘School’ of Alexandria in the Second
and Third Centuries.” In Centres of Learning: Learning and Location in Pre-
modern Europe and the Near East, ed. J. W. Drijvers and A. M. MacDonald.
Leiden: –.
Brown, P. . Augustine of Hippo. Berkeley, Calif.
. The World of Late Antiquity, AD 150–750. New York.
. “Eastern and Western Christendom in Late Antiquity: A Parting of the
Ways.” In The Orthodox Church and the West. Studies in Church History :
– (reprinted in P. Brown, Society and the Holy in Late Antiquity (London,
): –).
. The Cult of the Saints. Chicago, IL.
. Power and Persuasion in Late Antiquity: Towards a Christian Empire.
Madison, Wis.
. Authority and the Sacred. Cambridge.
. Poverty and Leadership in the Later Roman Empire. Hanover, N.H., and
London.
Browning, R. . Review of J. F. Matthews, Western Aristocracies and the Imperial
Court A.D. 364–425. CR, N.S. .: –.
Bruggisser, P. . Symmaque ou le rituel épistolaire de l’amitié littéraire: Recherches
sur le premier livre de la correspondence. Paradosis . Fribourg, Switz.
 References 

Brugnoli, G., ed. . Foca: Vita di Virgilio. Pisa.

ed. . “Vita Focae.” In Vitae Vergilianae Antiquae, ed. G. Brugnoli and F.
Stok. Rome: –.
Brunt, P. A. . “Charges of Provincial Maladministration under the Early
Principate.” Historia : –.
. “Lex de imperio Vespasiani.” JRS : –.
. The Fall of the Roman Republic. Oxford.
Bruun, C., ed. . The Roman Middle Republic: Politics, Religion and Historiog-
raphy c. 400–133 B.C. Acta Instituti Romani Finlandiae . Rome.
Burgess, R. W. . “Jerome’s Chronici canones, Quellenforschung and Fourth-
Century Historiography.” In Historiae Augustae Colloquium Argentoratense,
ed. G. Bonamente, F. Heim., and J.-P. Callu. Bari: –.
. “Jerome Explained: An Introduction to His Chronicle and a Guide to Its
Use.” AHB : –.
. “The Passio S. Artemii, Philostorgius and the Dates of the Invention and
Translation of the Relics of Andrew and Luke.” Analecta Bollandiana :
–.
. Review of Marasco . BMCR ...
. “A Common Source for Jerome, Eutropius, Festus, Ammianus and the
Epitome de Caesaribus between  and , along with Further Thoughts on
the Date and Nature of the Kaisergeschichte.” CP : –.
Burns, T. S. . Barbarians within the Gates of Rome: A Study of Roman Military
Policy and the Barbarians, Ca. 375–425 A.D. Bloomington, Ind.
Cairns, J. W., and P. de Plessis, eds. . Beyond Dogmatics: Law and Society in
the Roman World. Edinburgh.
Callu, J. P. . Symmaque: lettres. Vol. : Livres VI–VIII. Paris.
. “À la frange de l’histoire: cinq textes ou paratextes de l’antiquité tardive.”
Numismatica e antichità classiche : –.
. Symmaque: lettres. Vol. : Livres IX–X, Paris.
Cameron, Alan. . “Wandering Poets: A Movement in Byzantine Egypt.”
Historia : –.
. “Gratian’s Repudiation of the Pontifical Robe.” JRS : –.
. Claudian: Poetry and Propaganda at the Court of Honorius. Oxford.
. “The Date and the Owners of the Esquiline Treasure.” AJA : –.
. “Poetry and Literary Culture in Late Antiquity.” In Approaching Late
Antiquity: The Transformation from Early to Late Empire, ed. S. Swain and
M. Edwards. Oxford: –.
. “The Imperial Pontifex.” HSCP : –.
Cameron, Alan, J. Long, and L. Sherry. . Barbarians and Politics at the Court
of Arcadius. Berkeley, Calif.
Cameron, Averil. . “New Themes and Styles in Greek Literature, a Title
Revisited.” In Greek Literature in Late Antiquity: Dynamism, Didacticism,
Classicism, ed. S. F. Johnson. Aldershot: –.
Cameron, Averil, and S. Hall, eds. . Eusebius, Life of Constantine. Oxford.
Campbell, J. B. . The Emperor and the Roman Army, 31 BC–AD 235. Oxford.
 References

Canfora, L. . The True History of the So-called Artemidorus Papyrus. Bari.
Carrié, J.-M. . “Patronage et propriété militaire au IVe siècle: Objet rhétorique
et objet réel du discours ‘Sur les Patronages’ de Libanius.” BCH : –.
Cecconi, G. A. . Commento storico al libro secondo dell’epistolario di Q. Aurelio
Simmaco. Pisa.
Champlin, E. . “The Testament of the Piglet.” Phoenix : –.
. Final Judgments: Duty and Emotion in Roman Wills. Berkeley, Calif.
Chastagnol, A. . L’album municipal de Timgad. Bonn.
. L’évolution politique, sociale et économique du monde romain de Dioclétien
à Julien: la mise en place du Bas-Empire (284–363). Paris.
Christol, M. . Essai sur l’évolution des carrières sénatoriales dans la 2e moitié du
IIIe s. ap. J.-C. Paris.
Clausen, W., ed. . Virgil: Eclogues. Oxford.
Comparetti, D. [] . Virgil in the Middle Ages. Trans. E. F. M. Benecke.
Reprint, Princeton.
Connolly, S. . “½mnuw aÉt¼n t¼n Sebast¼n: The Greek Oath in the Roman
World.” In Horkos: The Oath in Greek Society, ed. A. Sommerstein and J.
Fletcher. Bristol: –.
Conte, G. B. . Latin Literature: A History. Trans. J. B. Solodow. Baltimore
and London.
Cooper, K., and J. Hillner, eds. . Religion, Dynasty, and Patronage in Early
Christian Rome, 300–900. Cambridge.
Corcoran, S. . “Hidden from History: The Legislation of Licinius.” In Harries
and Wood : –.
. The Empire of the Tetrarchs: Imperial Pronouncements and Government,
AD 284–324. Oxford.
a. The Empire of the Tetrarchs: Imperial Pronouncements and Government,
AD 284–324. nd edn. Oxford.
b. “The lex Ovinia and the Emancipation of the Senate.” In Bruun :
–.
. “Emperor and Citizen in the Era of Constantine.” In Constantine the
Great: York’s Roman Emperor, ed. E. Hartley et al. Aldershot: –.
Cornell, T. . Beginnings of Rome: Italy and Rome from the Bronze Age to the
Punic Wars (c. 1000–264 BC). London.
. “The lex Ovinia and the Emancipation of the Senate.” In Bruun :
–.
Corti, G. . Lucifero di Cagliari: Una voce nel conflitto tra chiesa e impero alla
metà del IV secolo. Milan.
Coşkun, A. a. “Chronology in the Eucharisticos of Paulinus Pellaeus: A
Reassessment.” Mnemnosyne : –.
b. Die gens Ausonia an der Macht. Oxford.
. “Notes on the Eucharisticos of Paulinus Pellaeus: Towards a New Edition
of the Autobiography.” Exemplaria Classica : –.
. “The Eucharisticos of Paulinus Pellaeus: Towards a Reappraisal of the
Worldly Convert’s Life and Autobiography.” VChrist : –.
 References 

. Review of Paulinus Pellaeus, Carmina, by C. M. Lucarini. BMCR

...
Cotton, H. . Documentary Letters of Recommendation in Latin from the Roman
Empire. Königstein.
. “The Guardianship of Jesus Son of Babatha: Roman and Local Law in the
Province of Arabia.” JRS : –.
Courcelle, P. . Les lettres grecques en Occident: De Macrobe à Cassiodore. Paris.
. Les Confessions de saint Augustin dans la tradition littéraire. Paris.
. Histoire littéraire des grandes invasions germaniques. rd edn. Paris.
Cox, P. . Biography in Late Antiquity: A Quest for the Holy Man. Berkeley, Calif.
Cox Miller, P. . “Strategies of Representation in Collective Biogra-
phy: Constructing the Subject as Holy.” In Rousseau and Hägg :
–.
Crawford, M. H. . Roman Republican Coinage. Cambridge.
. Roman Statutes. Bulletin of the Institute of Classical Studies . London.
Cribiore, R. . Writing, Teachers, and Students in Graeco-Roman Egypt. Atlanta,
Ga.
. Gymnastics of the Mind: Greek Education in Hellenistic and Roman Egypt.
Princeton, N.J.
. The School of Libanius in Late Antique Antioch. Princeton, N.J.
Croke, B. . “AD : The Manufacture of a Turning Point.” Chiron : –.
. The Chronicle of Marcellinus: Translation and Commentary. Sydney.
. Count Marcellinus and His Chronicle. Oxford.
Crook, J. A. . Law and Life of Rome. London.
. Legal Advocacy in the Roman World. London.
Curran, E. . Pagan City and Christian Capital: Rome in the Fourth Century.
Oxford.
Dagron, G. . “Les moines et la ville: Le monachisme à Constantinople jusqu’au
Concile de Chalcédoine.” T&MByz : –.
. Naissance d’une capitale: Constantinople et ses institutions de 330 à 451. Paris.
. “Le christianisme dans la ville Byzantine.” DOP : –.
. Constantinople imaginaire. Paris.
. Emperor and Priest. Cambridge.
Demarolle, J. . “L’image de l’empereur Julien dans la deuxième moitié du
XVIième s.” In Aiôn. Les temps chez les Romains, ed. R. Chevalier. Paris:
–.
Deniaux, E. . Clientèles et pouvoir à l’époque de Cicéron. Paris.
De Ste. Croix, G. E. M. . “Suffragium: From Vote to Patronage.” British
Journal of Sociology : –.
Diefenbach, S. . “Zwischen Liturgie und civilitas. Konstantinopel im
 Jahrhundert und die Etablierung eines städtischen Kaisertums.” In
Bildlichkeit und Bildort von Liturgie: Schauplätze in Spätantike, Byzanz und
Mittelalter, ed. R. Warland. Wiesbaden: –.
Dietz, K., and D. Hennig, eds. . Klassisches Altertum, Spätantike und frühes
Christentum: Festschrift für Adolf Lippold. Würzburg.
 References

Dillon, J. . The Heirs of Plato: A Study of the Old Academy. Oxford.
Dionisotti, A. C. . “From Ausonius’ schooldays? A schoolbook and Its Rela-
tives.” JRS : –.
Dmitriev, S. . City Government in Hellenistic and Roman Asia Minor. Oxford.
Drinkwater, J. F. . Review of Matthews . CR : –.
Drinkwater, J. F., and H. Elton, eds. . Fifth-Century Gaul: A Crisis of Identity?
Cambridge.
Duckworth, G. E. . “The Cradle of Flowers (Ecl. .).” TAPA : –.
Ebbeler, J. V., and C. Sogno. . “Religious Identity and the Politics of Patron-
age: Symmachus and Augustine.” Historia : –.
Edwards, M. . “Birth, Death, and Divinity in Porphyry’s Plotinus.” In
Rousseau and Hägg : –.
Edwards, M., and S. Swain, eds. . Approaching Late Antiquity: The Transfor-
mation from Early to Late Empire. Oxford.
Effenberger, A. . “Überlegungen zur Aufstellung des Theodosios-Obelisken
im Hippodrom von Konstantinopel.” In Innovation in der Spätantike: Kollo-
quium Basel 1994, ed. B. Brenk. Wiesbaden: –.
Eigler, U. . Lectiones vetustatis: Römische Literatur und Geschichte in der lateinis-
chen Literatur der Spätantike. Monographien zur klassischen Altertumswis-
senschaft . Munich.
Eilers, C. . Roman Patrons of Greek Cities. Oxford.
Ellis, L., and F. Kidner, eds. . Travel, Communication and Geography in Late
Antiquity. Aldershot and Burlington, Vt.
Ellis, S. . “The ‘Palace of the Dux’ at Apollonia, and Related Houses.” In
Cyrenaica in Antiquity, ed. G. Barker, J. Lloyd, and J. Reynolds. BAR Inter-
national Series . Oxford: –.
. “Power, Architecture and Decor: How the Late Roman Aristocrat Appeared
to His Guests.” In Roman Art in the Private Sphere: New Perspectives on the
Architecture and Décor of the Domus, Villa and Insula, ed. E. K. Gazda. Ann
Arbor, Mich.: –.
. “Classical Reception Rooms in Romano-British Houses.” Britannia :
–.
Elm, S. . “Hellenism and Historiography: Gregory of Nazianzus and Julian
in Dialogue.” Journal of Early Medieval Europe : –.
Elm, S., E. Rebillard, and A. Romano, eds. . Orthodoxie, christianisme, histoire.
Rome.
Ernesti, J. . Princeps christianus und Kaiser aller Römer: Theodosius der Grosse
im Lichte zeitgenössischer Quellen. Paderborn/Munich/Vienna.
Errington, R. M. a. “The Accession of Theodosius I.” Klio : –.
b. “Theodosius and the Goths.” Chiron : –.
a. “Church and State in the First Years of Theodosius I.” Chiron : –.
b. “Christian Accounts of the Religious Legislation of Theodosius I.” Klio
: –.
. “Themistius and His Emperors.” Chiron : –.
. Roman Imperial Policy from Julian to Theodosius. Chapel Hill, N.C.
 References 

Escribano Paño, M. V. . “Graciano, Teodosio y el Ilı́rico: la constitutio Nullus

(locus) haereticis.” RIDA : –.
. “Heresy and Orthodoxy in Fourth-Century Hispania: Arianism and
Priscillianism.” In Bowes and Kulikowski : –.
Evans Grubbs, J. . Law and Family in Late Antiquity: The Emperor Constantine’s
Marriage Legislation. Oxford.
. “Marrying and Its Documentation in Later Roman Law.” In Reynolds
and Witte : –.
Evelyn White, H. G., ed. –. Ausonius.  vols. London/Cambridge, Mass.
Faedo, L. . “Il complesso monumentale del Foro di Teodosio a Constantinop-
oli.” Corsi di cultura sull’arte ravennate e bizantina : –.
. “Teodosio, Temistio e l’ideologia erculea nella Nea Rome: a proposito
dell’arco del Forum Tauri.” MDAI(R) : –.
Fairweather, J. A. . “Fiction in the Biographies of Ancient Writers.” Anc. Soc.:
–.
Festy, M., ed. and trans. . Pseudo-Aurélius Victor: Abrégé des Césars. Paris.
Finn, R. . Almsgiving in the Later Roman Empire: Christian Promotion and
Practice, 313–450. Oxford.
Fontaine, J. . Review of Matthews . REL : –.
Foucault, M. . “What Is an Author?” In The Foucault Reader, ed. P. Rabinow.
New York: –.
Fowden, G. . “The Pagan Holy Man in Late Antique Society.” JHS : –.
French, D. . “Maintaining Boundaries: The Status of Actresses in Early Chris-
tian Society.” VCh : –.
Gagos, T., and P. Heilporn. . “A New agnitio bonorum possessionis.” In Essays
and Texts in Honor of J. D. Thomas (American Studies in Papyrology ), ed.
T. Gagos and R. Bagnall. Exeter: –.
Gaiser, K. . Philodemus Academica: Die Berichte über Platon und die Alte
Akademie in zwei herkulanensischen Papyri. Stuttgart.
Gallazi, C. and B. Kramer. . “Artemidorus im Zeichensaal. Eine Papyrus-
rolle mit Text, Landkarte und Skizzenbüchern aus späthellenistischer Zeit.”
Archive für Papryusforschung : –.
Galvão-Sobrinho, C. R. . “The Rise of the Christian Bishop: Doctrine and
Power in the Later Roman Empire, AD –.” PhD diss. Yale University.
Garnsey, P., and C. Humfress. . Evolution of the Late Antique World.
Cambridge.
Garnsey, P., and G. Woolf. . “Patronage of the Rural Poor in the Roman
World.” In Patronage in Ancient Society, ed. A. Wallace-Hadrill. London and
New York: –.
Gautier, F. . La retraite et le sacerdoce chez Grégoire de Nazianze. Turnhout.
. “À propos du témoignage de Grégoire de Nazianze sur le concile de
Constantinople (mai–juillet ) aux vers – du De uita sua.” REAug
: –.
Gellner, E., and J. Waterbury, eds. . Patrons and Clients in Mediterranean
Societies. London.
 References

Geyssen, J. . “Presentations of the Victory on the Theodosian Obelisk.”

Byzantion : –.
Giardina, A. –. Dialoghi di Archeologia –: –.
. “Esplosione di tardoantico.” Studi storici : –.
Gibbon, E. . The History of the Decline and Fall of the Roman Empire. Ed. D.
Womersley. Vols. –. London.
Gillett, A., ed. . On Barbarian Identity: Critical Approaches to Ethnicity in the
Early Middle Ages. Turnhout.
Gleason, M. . “Festive Satire: Julian’s Misopogon and the New Year at Antioch.”
JRS : –.
Goetz, G. . Corpus glossariorum Latinorum, III. Hermeneumata Pseudo-
dositheana. Leipzig.
Goetz, H.-W. . “Der Umgang mit der Geschichte in der lateinischen
Weltchronistik des hohen Mittelalters.” In Julius Africanus und die christliche
Weltchronik, ed. M. Wallraff. Berlin and New York: –.
Gómez Villegas, N. . Gregorio de Nazianzo en Constantinopla. Madrid.
Goody, J. . Representations and Contradictions: Ambivalence towards Images,
Theatre, Fiction, Relics and Sexuality. Oxford.
Grafton, A., and M. Williams. . Christianity and the Transformation of the
Book: Origen, Eusebius, and the Library of Caesarea. Cambridge, Mass., and
London.
Greatrex, G. . “The Background and Aftermath of the Partition of Armenia
in A.D. .” AHB : –.
Green, R. P. H. . “Still Waters Run Deep: A New Study of the Professores of
Bordeaux.” CQ : –.
. “Greek in Late Roman Gaul: The Evidence of Ausonius.” In “Owls to
Athens”: Essays on Classical Subjects Presented to Sir Kenneth Dover, ed. E. M.
Craik. Oxford: –.
Grey, C. . “Letters of Recommendation and the Circulation of Rural Laborers
in the Late Roman West.” In Ellis and Kidner : –.
Grierson, P. . “The Tombs and Obits of the Byzantine Emperors (–).”
DOP : –.
Grumel, R. P. V. . “L’Illyricum de la mort de Valentinien Ier () à la mort
de Stilicon ().” REByz : –.
Gryson, R., ed. . Scolies ariennes sur le concile d’Aquilée. Sources chrétiennes
. Paris.
Habinek, T. . The Politics of Latin Literature: Writing, Identity, and Empire in
Ancient Rome. Princeton, N.J.
Hadas, M., and M. Smith. . Heroes and Gods: Spiritual Biographies in Antiquity.
New York.
Halsall, G. . Barbarian Migrations and the Roman West 376–568. Cambridge.
Hanson, R. P. C. . The Search for the Christian Doctrine of God: The Arian
Controversy, 318–381. Edinburgh.
Hardie, C., ed. . Vitae Vergilianae Antiquae. Oxford.
 References 

Harries, J. . “How to Make a Law Code.” In Modus Operandi: Essays in Honor
of Geoffrey Rickman, ed. M. Austin, J. Harries, and C. Smith. London: –.
. Law and Empire in Late Antiquity. Cambridge.
Harries, J., and I. Wood, eds. . The Theodosian Code: Studies in the Imperial
Law of Late Antiquity. London.
Harris, W., ed. . The Transformations of Vrbs Roma in Late Antiquity.
Portsmouth, R.I.
Hauken, T. . Petition and Response: An Epigraphic Study of Petitions to Roman
Emperors, 181–249. Bergen.
Heather, P. J. . “The Anti-Scythian Tirade of Synesius’ De Regno.” Phoenix
: –.
. Goths and Romans 332–489. Oxford.
a. “Literacy and Power in the Migration Period.” In Literacy and Power in
the Ancient World, ed. A. Bowman and G. Woolf. Cambridge: –.
b. “The Historical Culture of Ostrogothic Italy.” In Teoderico il Grande e i
Goti d’Italia: Atti del XIII Congresso internazionale di studi sull’Alto Medioevo
(Spoleto): –.
. “New Men for New Constantines? Creating an Imperial Elite in the East-
ern Mediterranean.” In New Constantines: The Rhythm of Imperial Renewal
in Byzantium, 4th–13th Centuries, ed. P. Magdalino. Aldershot: –.
. “Senates and Senators.” In Cambridge Ancient History. Vol. : The Late
Empire, A.D. 337–425, ed. A. Cameron and P. Garnsey. Cambridge: –.
. “The Late Roman Art of Client Management and the Grand Strategy
Debate.” In The Transformation of Frontiers from late Antiquity to the Car-
olingians, ed. W. Pohl and I. N. Wood. Leiden: –.
. “State, Lordship, and Community in the Early Medieval West, –
 AD.” In Cambridge Ancient History. Vol. , ed. A. M. Cameron et al.
Cambridge: –.
. The Fall of the Roman Empire: A New History. London.
. “Ethnicity, Group Identity, and Social Status in the Migration Period.” In
Franks, Northmen, and Slavs: Identities and State Formation in Early Medieval
Europe, ed. I. Garipzaonov et al. Leiden: –.
Heather, P. J., and J. F. Matthews. . The Goths in the Fourth Century. Translated
Texts for Historians. Liverpool.
Heather, P., and D. Moncur. . Politics, Philosophy, and Empire in the Fourth
Century: Themistius’ Select Orations. Liverpool.
Helm, R. . Hieronymus’ Zusätze in Eusebius’ Chronik und ihr Wert für die
Literaturgeschichte. Philologus, Supplementband , Heft . Leipzig.
Hoek, A. van den. . “The ‘Catechetical’ School of Early Christian Alexandria
and its Philonic Heritage.” HThR : –.
Hoffer, S. E. . The Anxieties of Pliny the Younger. Atlanta, a.
Hölkeskamp, K.-J. . “Fides–deditio in fidem–dextra data et accepta: Recht,
Religion und Ritual in Rom.” In Bruun : –.
Holum, K. . Theodosian Empresses. Berkeley, Calif.
 References

Honoré, A. M. . “The Making of the Theodosian Code.” ZSS RA : –.
. Law in the Crisis of Empire, 379–455: The Theodosian Dynasty and Its
Quaestors. Oxford.
Hopkins, K. . “Social Mobility in the Later Roman Empire: The Evidence of
Ausonius.” CQ : –.
Horsfall, N. . “Virgil: His Life and Times.” In A Companion to the Study of
Virgil, ed. N. Horsfall. Leiden: –.
Humfress, C. . “Roman Law, Forensic Argument and the Formation of Chris-
tian Orthodoxy (III–VI Centuries).” In Elm, Rebillard, and Romano :
–.
. “Civil Law and Social Life.” In The Cambridge Companion to the Age of
Constantine, ed. N. Lenski. Cambridge: –.
. Orthodoxy and the Courts in Late Antiquity. Oxford.
Humphries, M. . “Roman Senators and Absent Emperors in Late Antiquity.”
AAAH : –.
. Review of Matthews . CR : –.
. “From Emperor to Pope? Ceremonial, Space, and Authority at Rome from
Constantine to Gregory the Great.” In Cooper and Hillner : –.
Hunter, D. . “Marrying and the tabulae nuptiales in Roman North Africa.”
In Reynolds and Witte : –.
Inglebert, H. . Les Romains chrétiens face à l’histoire de Rome: Histoire, chris-
tianisme et romanités en Occident dans l’Antiquité tardive (IIIe au Ve siècles).
Paris.
Iskandar, A. Z., ed. and trans. . Galen: On How to Recognize the Best Physicians.
CMG Supplementum Orientale . Berlin.
Isnardi Parente, M. . “Per la biografia di Senocrate.” RFIC : –.
Janin, R. . Constantinople byzantine. Paris.
. Les églises et les monastères. Vol.  of La géographie ecclésiastique de l’Empire
byzantin, part . Le siège de Constantinople et le patriarcat oecuménique. nd
edn. Paris.
. Les églises et les monastères des grands centres byzantins. Paris.
Jeanjean, B. . “Saint Jérôme, patron des chroniqueurs en langue latine.” In
Saint Jérôme, Chronique: Continuation de la Chronique d’Eusèbe, années 326–
378, suivie de quatre études sur les chroniques et chronographies dans l’Antiquité
tardive (IVe–VIe siècles), ed. B. Jeanjean and B. Lançon. Rennes: –.
Johnson, M. a. “On the Burial Places of the Theodosian Dynasty.” Byzantion
: –.
b. “On the Burial Places of the Valentinian Dynasty.” Historia : –.
Jones, A. H. M. . “The Roman Civil Service (Clerical and Sub-clerical
Grades).” JRS : –.
a. The Later Roman Empire: A Social Economic and Administrative Survey.
 vols. Oxford.
b. “Collegiate Prefectures.” JRS : –.
. The Decline of the Ancient World. London.
Jungck, C. . Gregor von Nazianz: De vita sua. Heidelberg.
 References 

Kaster, R. A. . “Notes on ‘Primary’ and ‘Secondary’ Schools in Late Antiquity.”

TAPA : –.
. Guardians of Language: The Grammarian and Society in Late Antiquity.
Berkeley, Calif.
Kelly, C. . Ruling the Later Roman Empire. Cambridge, Mass., and London.
. “Crossing the Frontiers: Imperial Power in the Last Book of Ammianus.”
In den Boeft et al. : –.
Kelly, G. . “The New Rome and the Old: Ammianus Marcellinus’ Silences
on Constantinople.” CQ : –.
. “Ammianus and the Great Tsunami.” JRS : –.
. Ammianus Marcellinus: The Allusive Historian. Cambridge.
Kelly, J. N. D. . Jerome. London.
Kennedy, G. . Greek Rhetoric under Christian Emperors. Princeton, N.J.
Kent, J. . “‘Concordia’ Solidi of Theodosius I: A Reappraisal.” Numismatic
Chronicle : –.
Kiilerich, B. . The Obelisk Base in Constantinople: Court Art and Imperial
Ideology. Rome.
Klein, H. A. . “Sacred Relics and Imperial Ceremonies at the Great Palace
of Constantinople.” In Visualisierungen von Herrschaft (Byzas ), ed. F. A.
Bauer. Istanbul: –.
Kolias, G. T. . Ämter- und Würdenkauf im früh- und mittelbyzantinischen Reich.
Athens.
Kopocek, T. A. . “Curial Displacements and Flight in Later Fourth Century
Cappadocia.” Historia : –.
Koster, S. . Die Invektive in der griechischen und römischen Literatur. Meisen-
heim am Glan.
Krause, J.-U. . Spätantike Patronatsformen im Westen des Römischen Reiches.
Munich.
Kulikowski, M. () “Nation versus Army: A Necessary Contrast?” In Gillett
: –.
Kurmann, A. . Gregor von Nazianz: Oratio 4 gegen Julian. Ein Kommentar.
Basle.
La Bléterie, Abbé de. . Vie de l’empereur Julien. nd edn. Paris.
Lane Fox, R. . Pagans and Christians. New York.
Lavan, M. . “Rome and the Provinces: Slavery, Patronage and Other
Paradigms for Empire.” Unpublished PhD diss., Cambridge University.
Leader-Newby, R. . Silver and Society in Late Antiquity: Functions and Mean-
ings of Silver Plate in the Fourth to Seventh Centuries. Burlington, Vt.
Lefkowitz, M. R. . The Lives of the Greek Poets. Baltimore, Md.
Lenski, N. . “Initium mali Romano imperio: Contemporary Reactions to the
Battle of Adrianople.” TAPhA : –.
. Failure of Empire: Valens and the Roman State in the Fourth Century A.D.
Berkeley/London.
a. Failure of Empire: Valens and the Roman State in the 4th Century AD.
Berkeley, Calif.
 References

b. Review of Matthews . CJ /: –.

Leo, F. . Die griesch-römische Biographie nach ihrer litterarischen Form. Leipzig.
Lepelley, C. . “Le patronat épiscopale aux IVe et Ve siècles: Continuités et
ruptures avec le patronat classique.” In L’évêque dans la cité du IVe au Ve
siècle: Image et autorité, ed. E. Rebillard and C. Sotinel. Rome: –.
Leppin, H. . Von Constantin dem Grossen zu Theodosius II: Das christliche
Kaisertum bei den Kirchenhistoriken Socrates, Sozomenus und Theodoret.
Göttingen.
. Theodosius der Grosse. Darmstadt.
Leroy-Mohlinghen, A. . “A propos de la Vie de Syméon Stylites.” Byzantion :
–.
Leroy-Mohlinghen, A. and P. Canivet, eds. –. Histoire des moines de Syrie:
Histoire Philothée de Théodoret de Cyr.  vols. Paris.
Lewis, N. . “NOHMATA LEGONTOS.” BASP : –.
Leyerle, B. . Theatrical Shows and Ascetic Lives: John Chrysostom’s Attack on
Spiritual Marriage. Berkeley, Calif.
Liebeschuetz, J. H. W. G. . Antioch: City and Imperial Administration in the
Later Roman Empire. Oxford.
. “Alaric’s Goths: Nation or Army?” In Drinkwater and Elton : –.
a. “Late Antiquity and the Concept of Decline.” Nottingham Medieval
Studies : –.
b. The Decline and Fall of the Roman City. Oxford.
. “Pagan Historiography and the Decline of the Empire.” In Marasco :
–.
Liebs, D. . “Rechtsschulen unter Rechtsunterricht im Prinzipat.” ANRW II.:
–.
. “Ämterkauf und Ämterpatronage in der Spätantike: Propaganda und
Sachzwangt bei Julian dem Abtrünnigen.” ZRG : –.
Lim, R. . “Converting the Un-Christianizable: The Baptism of Stage Per-
formers in Late Antiquity.” In Conversion in Late Antiquity and the Early
Middle Ages: Seeing and Believing, ed. K. Mills and A. Grafton. Rochester,
N.Y.: –.
Lindsay, J. . Song of a Falling World: Culture during the Break-up of the Roman
Empire (A.D. 350–600). London.
Lippold, A. . Theodosius der Grosse und seine Zeit. Munich.
Lizzi Testa, R. . Senatori, popolo e papi: Il governo di Roma al tempo dei
Valentiniani. Bari.
Lobel, E., E. G. Turner, and R. P. Winnington-Ingram. . The Oxyrhynchus
Papyri: Part XXV [nos 2426–2437]. Graeco-Roman Memoirs . London.
Lucarini, C. M., ed. . Paulinus Pellaeus, Carmina. Munich and Leipzig.
Lugaresi, L., ed. and trans. . Gregorio di Nazianzo: Contro Giuliano l’Apostata:
Orazione 4. Florence.
. “Ambivalenze della rappresentazione: Riflessioni patristiche su riti e spet-
tacoli.” ZAC/JAC : –.
Lyman, R. . “Hellenism and Heresy.” JECS : –.
 References 

MacCormack, S. . “Latin Prose Panegyrics.” In Empire and Aftermath: Silver

Latin II, ed. T. A. Dorey. London and Boston: –.
. “Latin Prose Panegyrics: Tradition and Discontinuity in the Later Roman
Empire.” REAug : –.
. Art and Ceremony in Late Antiquity. Berkeley, Calif.
MacMullen, R. . Corruption and the Decline of Rome. New Haven, Conn.
Magie, D., trans. . The Scriptores Historiae Augustae. Cambridge, Mass.
Mango, C. . Le développement urbain de Constantinople (IVe–VIIe siècles). Paris.
. “The Triumphal Way of Constantinople and the Golden Gate.” DOP
: –.
. “The shoreline of Constantinople in the fourth century.” In Byzantine
Constantinople: Monuments, Topography and Everyday Life, ed. N. Necipoglu.
Leiden: –.
Mansfeld, J. . Prolegomena: Questions to be Settled before the Study of an Author
or Text. Leiden.
 Prolegomena Mathematica: From Apollonius of Perge to Late Neoplatonism.
Leiden.
Marasco, G., ed. . Greek and Roman Historiography in Late Antiquity: Fourth
to Sixth Century A.D. Leiden and Boston, Mass.
. Filostorgio: Cultura, fede e politica in uno storico ecclesiastico del V secolo.
Rome.
Maraval, P. . Lieux saints et pèlerinages d’Orient. Paris.
Marcone, A. . Commento storico al libro sesto dell’epistolario di Q. Aurelio
Simmaco. Pisa.
. “La corruzione politica in Roma antica. Un incontro fra storici e giuristi
sulla corruzione nella tarda antichità.” Riv.Stor. Ant. : –.
. “La riforma giustinianea dell’amministrazione periferica: La Novella .”
In Poteri centrali e poteri periferici nella tarda antichità: Confronti conflitti, ed.
L. di Paola and D. Minutoli. Florence: –.
. “A Long Late Antiquity? Considerations on a Controversial Periodiza-
tion,” tr. C. Sogno. JLA .: –.
Marrou, H. I. . Saint Augustin et la fin de la culture antique. Paris.
. A History of Education in Antiquity. New York.
Mateos, J. –. Le Typicon de la Grande Église.  vols. Rome.
Mathisen, R. W. . Roman Aristocrats in Barbarian Gaul: Strategies for Survival
in an Age of Transition. Austin, Tex.
Matthews, J. F. . “Continuity in a Roman Family: The Rufii Festi of Volsinii.”
Historia : –.
. “Olympiodorus of Thebes and the history of the west (A.D. –).”
JRS : –.
. “Gallic Supporters of Theodosius.” Latomus : –. (Repr. in J. F.
Matthews, Political Life and Culture in Late Roman Society. London, :
no. IX).
. “The Letters of Symmachus.” In Latin Literature of the Fourth Century,
ed. J. W. Binns. London: –.
 References

. Western Aristocracies and Imperial Court, A.D. 364–425. Oxford.

. “Anicius Manlius Severinus Boethius.” In Boethius, His Life, Thought and
Influence, ed. M. Gibson. Oxford: –.
. “Ammianus Marcellinus.” In Ancient Writers: Greece and Rome, ed. T. J.
Luce. New York: vol. : –.
. The Roman Empire of Ammianus. Baltimore.
. Western Aristocracies and Imperial Court, A.D. 364–425. Paperback edn.
Oxford.
. “The Making of the Text.” In Harries and Wood : –.
. “The Origin of Ammianus.” CQ n.s. : –.
. Laying Down the Law: A Study of the Theodosian Code. New Haven, Conn.
. The Journey of Theophanes: Travel, Business, and Daily Life in the Roman
East. New Haven, Conn.
. The Roman Empire of Ammianus. nd edn. Ann Arbor, Mich.
Matthews, J. and T. Cornell. . The Atlas of the Roman World. New York.
Mayer, M. . “De nuevo sobre la tradición biográfica virgiliana.” AFB : –.
Mayer, W. . “Cathedral Church or Cathedral Churches? The Situation at
Constantinople (c. – AD).” OCP : –.
Mazhuga, V. . “À quelle époque vivait le grammairien Phocas?” RPh :
–.
Mazzarino, A. –. “Appunti sul metodo II: Intorno all’età e all’opera di Foca.”
Helikon –: –.
McCormick, M. . Eternal Victory: Triumphal Rulership in Late Antiquity,
Byzantium, and the Early Medieval West. Cambridge.
McGill, S. . Virgil Recomposed: The Mythological and Secular Centos in Antiq-
uity. Oxford.
McGuckin, J. A. a. Saint Gregory of Nazianzus: An Intellectual Biography.
Crestwood, N.Y.
b. “Autobiography as Apologia in St. Gregory of Nazianzus.” Studia Patris-
tica . Leuven: –.
McKitterick, R. . History and Memory in the Carolingian World. Cambridge.
McLynn, N. B. a. Ambrose of Milan: Church and Court in a Christian Capital.
Berkeley, Calif.
b. “Paulinus the Impenitent: A Study of the Eucharisticos.” JECS : –.
. “The Transformation of Imperial Churchgoing in the Fourth Century.”
In Approaching Late Antiquity: The Transformation from Early to Late Empire,
ed. S. Swain and M. Edwards. Oxford: –.
. “‘Genere Hispanus’: Theodosius, Spain and Nicene Orthodoxy.” In Bowes
and Kulikowski : –.
Meehan, D. M. . Saint Gregory of Nazianzus: Three Poems. Washington, D.C.
Meischner, J. . “Das Missorium des Theodosius in Madrid.” JDAI : –
.
Mergiali-Sahas, S. . “Byzantine Emperors and Holy Relics.” JöByz : –.
Mervaud, C. . “Julien l’Apostat dans la correspondance de Voltaire et
Frédéric II.” Revue d’histoire littéraire de la France : –.
 References 

Mesnard, P., ed. . Œuvres philosophiques de Jean Bodin. Paris.

Méthy, N. . Les lettres de Pline le Jeune: une représentation de l’homme. Paris.
Meyer, E. . Legitimacy and Law in the Roman World: Tabulae in Roman Belief
and Practice. Cambridge.
. “Diplomatics, Law and Romanisation in Documents from the Judaean
Desert.” In Cairns and de Plessis : –.
Millar, F. . Review of Les carrières procuratoriennes équestres sous le Haut-Empire
romain, by H.-G. Pflaum. JRS : –.
. The Emperor in the Roman World. London.
. The Emperor in the Roman World. nd edn. Ithaca, N.Y.
. The Crowd in Rome in the Late Republic. Ann Arbor, Mich.
. A Greek Roman Empire: Power and Belief under Theodosius II, 408–450.
Berkeley, Calif.
Mitchell, S. . “Maximinus and the Christians in AD ; A New Latin Inscrip-
tion.” JRS : –.
. A History of the Later Roman Empire, AD 284–641: The Transformation of
the Ancient World. Malden, Mass.
Moffatt, A. . “Celebrating a Byzantine Imperial Birthday.” In Philhellen.
Studies in Honour of Robert Browning, ed. C. N. Constantinides, N. M.
Panagiotakes, E. M. Jeffreys, and A. D. Angelou. Istituto Ellenico di Studi
Bizantini e Postbizantini di Venezia, Bibliotheke, . Venice: –.
Mommsen, T. . Römisches Staatesrecht. rd edn. Berlin.
Moreschini, C., ed. . Grégoire de Nazianze: Discours 32–37. Sources chrétiennes
. Paris.
Morgan, T. . Literate Education in the Hellenistic and Roman Worlds.
Cambridge.
Morstein-Marx, R. . Mass Oratory and Political Power in the Late Roman
Republic. Cambridge.
Mouritsen, H. . Plebs and Politics in the Late Roman Republic. Cambridge.
Moussy, C., ed. . Paulin de Pella, Poème d’action de grâces et prière. Paris.
Namier, L. . The Structure of Politics at the Accession of George III. nd edn.
London.
Naumann, H. . “Suetons Vergilvita.” RhM : –.
. “Suetonius’ Life of Virgil: The Present State of the Question.” HSCP :
–.
Naumann, R. . “Neue Beobachtungen am Theodosiusbogen und Forum Tauri
in Istanbul.” Istanbuler Mitteilungen : –.
Nesselrath, H.-G. . “Zur Wiederentdeckung von Julian Apostata in der
Renaissance: Lorenzo de’ Medici und Ammianus Marcellinus.” Antike und
Abendland : –.
Nixon, C. E. V., and B. S. Rodgers, trans. and annot. . In Praise of Later
Roman Emperors: The Panegyrici Latini. Berkeley and Los Angeles, Calif.
Norman, A. F. . “Notes on Some Consulares of Syria.” ByzZ : –.
ed. and trans. . Libanius: Autobiography and Selected Letters. Vol. .
Cambridge, Mass.
 References

North, J. . Review of Brunt, . JRS : –.

Noy, D. . “Matchmakers and marriage markets in Antiquity.” EMC : –
.
Oakley, S. . A Commentary on Livy Books 6–10. Vol. . Oxford.
. A Commentary on Livy Books vi–x. Vol. . Oxford.
O’Donnell, J. . “Paganus.” Classical Folia : –.
ed. . Augustine: Confessions. Vols. –. Oxford.
Oliver, J. H. . “Minutes of a Trial Conducted by Caracalla at Antioch in A.D.
.” In Mélanges helléniques offerts à Georges Daux. Paris: –.
Orlin, E. M. . Temples, Religion and Politics in the Roman Republic. Leiden.
Oroz, J. . “De las vitae vergilianae a la vida de Virgilio.” Perfecit : –.
Osgood, J. . “Nuptiae Iure Civili Congruae: Apuleius’s Story of Cupid and
Psyche and the Roman Law of Marriage.” TAPhA : –.
Pack, R. C. . “Curiales in the Correspondence of Libanius.” TAPA :
–.
Parsons, P. . “Philippus Arabus and Egypt.” JRS : –.
Paschoud, F. . “Ammien ,,: Une recusatio?” REL : –.
Peschlow, U. . “Betrachtungen zur Gotensäule in Istanbul.” In Tesserae:
Festschrift für Josef Engemann. Jahrbuch für Antike und Christentum.
Ergänzungsband . Münster.
Peter, H. . Die geschichtliche Litteratur über die römische Kaiserzeit bis Theodo-
sius I und ihre Quellen.  vols. Leipzig.
ed. –. Historicorum Romanorum Reliquiae.  vols. Leipzig.
Petit, P. . “Les senateurs de Constantinople dans l’oeuvre de Libanius.”
L’Antiquité Classique : –.
Petzl, G. and E. Schwertheim. . Hadrian und die dionysischen Künstler: Drei in
Alexandria Troas neugefundene Breife des Kaisers an die Künstler Vereinigung.
Bonn.
Pflaum, H.-G. . Le Marbre de Thorigny. Paris.
Pharr, C. . “The Text and Interpretation of the Theodosian Code , , .”
AJPh .: –.
Piganiol, A. . L’empire chrétien (325–395). Paris.
Potter, D.S. . “Palmyra and Rome: The Issue of Odaenathus’s Titulature and
the Use of the imperium maius.” ZPE : –.
. The Roman Empire at Bay: AD 180–395. London and New York.
. “The Transformation of the Empire: – CE.” In A Companion to
the Roman Empire, ed. D. S. Potter. Oxford: –.
. Ancient Rome: A New History. London.
Price, R. . A History of the Monks of Syria. Kalamazoo, Mich.
Priessnig, A. . “Die literarische Form der spätantiken Philosophenromane.”
ByzZ : –.
. “Die biographische Form der Plotinvita des Porphyrios und das Anto-
niosleben des Athanasios.” ByzZ : –.
Purcell, N. . “The Apparitores: A Study in Social Mobility.” PBSR : –.
Quasten, J. . Patrology: Vol. . Utrecht.
 References 

Quiroga, A. . “From Sophistopolis to Episcopolis. The Case for a Third

Sophistic.” Journal for Late Antique Religion and Culture : –.
Rapp, C. . Holy Bishops in Late Antiquity: The Nature of Christian Leadership
in an Age of Transition. Berkeley, Los Angeles, London.
Rauschen, G. . Jahrbücher der christlichen Kirche unter dem Kaiser Theodosius
dem Grossen. Freiburg.
Rea, J. . “The Date of the Prefecture of Statilius Aemilianus.” Chronique
d’Égypte : –.
Rebenich, S. . “Gratian, a Son of Theodosius, and the Birth of Galla Placidia.”
Historia : –.
a. Hieronymus und sein Kreis: Prosopographische und sozialgeschichtliche
Untersuchungen. Historia Einzelschriften . Stuttgart.
b. “Zum Theodosiusobolisken in Konstantinopel.” Istanbuler Mitteilungen
: –.
. “Der heilige Hieronymus und die Geschichte: Zur Funktion der exempla
in seinen Briefen.” RQA : –.
. Jerome. London and New York.
. “Christian Asceticism and Barbarian Incursion: The Making of a Chris-
tian Catastrophe.” Journal of Late Antiquity .: –.
Rees, R. D. . “Diocletian and the Efficacy of Public Law.” In Cairns and de
Plessis : –.
Reifferscheid, A., ed. . C. Suetoni Tranquilli Praeter Caesarum Libros Reliquiae.
Leipzig.
Reynolds, L. D., and N. G. Wilson. . Scribes and Scholars: A Guide to the
Transmission of Greek and Latin Literature. rd edn. Oxford.
Reynolds, P. L., and J. Witte, Jr., eds. . To Have and To Hold: Marrying and
Its Documentation in Western Christendom, 400–1600. Cambridge.
Richardson, E. C., ed. . Texte und Untersuchungen. Vol. . Leipzig.
Richardson, J. S. . “Imperium Romanum: Empire and the Language of Power.”
JRS : –.
. The Language of Empire: Rome and the Idea of Empire from the Third
Century BC to the Second Century AD. Cambridge.
Riginos, A. S. . Platonica: The Anecdotes concerning the Life and Writings of
Plato. Leiden.
Ritter, A. M. . Das Konzil von Konstantinopel und sein Symbol: Studien zur
Geschichte und Theologie des II. Ökumenischen Konzils. Göttingen.
Ritzerfeld, U. . “‘Omnia Theodosio cedunt subolique perenni’: Überlegungen
zu Bildprogramm und Bedeutung des Theodosiusobelisken in Kon-
stantinopel.” JAC : –.
Roberts, M. . “The Treatment of Narrative in Late Antique Literature.” Philo-
logus : –.
Roda, S. . Commento storico al libro nono dell’epistolario di Q. Aurelio Simmaco.
Pisa.
. “La polifunzionalità della lettera commendaticia: teoria e prassi
nell’epistolario simmachiano.” In Colloque genevois sur Symmaque à l’occasion
 References

du mille six centième anniversaire du conflit de l’autel de la Victoire, Genève,

4–7 Juin 1984, ed. F. Paschoud. Paris: –.
Rose, M., and S. Aydingun. . “Under Istanbul (Ancient Byzantine port at
Constantinople’s Ancient Harbor).” Archaeology .: –.
Rosen, K. . Review of Matthews . Gnomon : –.
. Julian: Kaiser, Gott und Christenhasser. Stuttgart.
Rostagni, A., ed. . Suetonio De poetis e biografi minori. Turin.
Roueché, C. . “Acclamations in the Later Roman Empire: New Evidence from
Aphrodisias.” JRS : –.
. Aphrodisias in Late Antiquity. JRS Monographs . London.
Rousseau, P. . “Antony as a Teacher in the Greek Life.” In Rousseau and
Hägg : –.
Rousseau, P. and T. Hägg, eds. . Greek Biography and Panegyric in Late
Antiquity. Berkeley, Calif.
Roussel, P., and F. de Visscher. –. “Les inscriptions du temple de Dmeir.”
Syria : –.
Rubenson, S. . “Antony and Pythagoras: A Reappraisal of the Appropriation
of Classical Biography in Athanasius’ Vita Antonii.” In Early Christianity in
the Context of Antiquity, ed. D. Brakke, A. Jacobsen, and J. Ulrich. Berlin:
–.
Russell, D. A., and N. G. Wilson, eds. . Menander Rhetor. Oxford.
Sabbah, G. . “Ammianus Marcellinus.” In Marasco : –.
Safran, L. . “Points of View: the Theodosian Obelisk Base in Context.” GRBS
: –.
Saller, R. P. . Personal Patronage under the Early Empire. Cambridge.
. “Familia, domus, and the Roman Concept of the Family.” Phoenix .:
–.
. “Men’s Age at Marriage and Its Consequences in the Roman Family.”
CPh : –.
. “Patronage and Friendship in Early Imperial Rome.” In Patronage in
Ancient Society, ed. A. Wallace-Hadrill. London and New York: –.
. Patriarchy, Property and Death in the Roman Family. Cambridge.
Salzman, M. R. . “Aristocratic Women: Conductors of Christianity in the
Fourth Century.” Helios : –.
. The Making of a Christian Aristocracy. Cambridge, Mass.
Sande, S. . “Some New Fragments from the Column of Theodosius.” AAAH
: –.
Saradi, H. . “Constantinople and its Saints.” Studi Medievali : –.
Sarris, P. . “Managing the Great Estate: Aristocratic Property and Economic
Growth in the Late Antique East.” In Recent Research in the Late Antique
Countryside, ed. W. Bowden, L. Lavan, and C. Machado. Leiden: –.
Schäfer, T. . Imperii insignia, sella curulis und fasces: Zur Repräsentation
römischer Magistrate. Mainz.
Scheid, J. . “L’investiture impériale d’après les commentaries des arvales.”
Cahiers du Centre G. Glotz : –.
 References 

. Commentarii Fratrum Arvalium qui supersunt; les copies épigraphiques des
protocols annuels de la conferrie Arvale (21 av.–304 ap. J-C.). Roma Antica 4.
Rome.
Scheidel, W. . “Emperors, Aristocrats and the Grim Reaper: Towards a Demo-
graphic Profile of the Roman Elite.” CQ : –.
Scholten, C. . “Die Alexandrinische Katechetenschule.” JbAC : –.
Schuller, W. . “Ämterkauf im römischen Reich.” Der Staat : –.
. “Zwischen Klientel und Korruption: zum römischen Beamtenwesen.” In
Festschrift Robert Werner, ed. W. Dahlheim et al. Konstanz: –.
Scourfield, J. H. D. . “Textual Inheritances and Textual Relations in Late
Antiquity.” In Texts and Culture in Late Antiquity: Inheritance, Authority,
and Change, ed. J. H. D. Scourfield. Swansea: –.
Screech, M. A. (ed.) . Michel de Montaigne. The Complete Essays, with tr. and
comm. London.
Secondat, Ch. de, Baron de Montesquieu. . Reflexions on the Causes of The
Rise and Fall of the Roman Empire Translated from the French. rd edition.
London.
Seeck, O. –. Geschichte des Untergangs der antiken Welt. Berlin.
. Regesten der Kaiser und Päpste für die Jahre 311 bis 476 n. Chr. Stuttgart.
Seyfahrt, W. . “Philologische Probleme um Ammianus Marcellinus: Gedanken
über die handschriftliche Überlieferung und eine moderne Textgestaltung.”
Klio : –.
Shanzer, D. R. . “The Date and Literary Context of Ausonius’ Mosella:
Ausonius, Symmachus, and the Mosella.” In Style and Tradition: Studies in
Honor of Wendell Clausen, ed. P. E. Knox and C. Fuss. Stuttgart and Leipzig:
–.
. “Avulsa a latere meo: Augustine’s Spare Rib: Confessions ...” JRS :
–.
Shaw, B. D. a. “The Family in Late Antiquity: The Experience of Augustine.”
P&P : –.
b. “The Age of Roman Girls at the Time of Marriage: Some Reconsidera-
tions.” JRS : –.
. “Raising and Killing Children: Two Roman Myths.” Mnemosyne : –.
Sherwin-White, A. N. . The Letters of Pliny: A Historical and Social Commen-
tary. Oxford.
Siniossoglou, N. . Plato and Theodoret: The Christian Appropriation of Platonic
Philosophy and the Hellenic Intellectual Resistance. Cambridge.
Sirks, B. . “The Sources of the Code.” In Harries and Wood : –.
Smith, C. S. . The Roman Clan: The gens from Ancient Ideology to Modern
Anthropology. Cambridge.
Smith, R. . “The Imperial Court of the Late Roman Empire.” In The Court
and Court Societies in Ancient Monarchies, ed. A. J. S. Spawforth. Cambridge:
–.
Sogno, C. . Q. Aurelius Symmachus: A Political Biography. Ann Arbor,
Mich.
 References

Soraci, C. . “Approvvigionamento e distribuzioni alimentari: considerazioni

sul ruolo dei vescovi nel tardo impero.” Quaderni catanesi di studi antichi e
medievali : –.
Speck, P. . “Beobachtungen zur Unterbasis des Theodosios-Obelisken im
Hippodrom von Konstantinopel.” Boreas : –.
Spira, A. . Gregorii Nysseni Opera 9.1. Leiden.
Stäcker, J. . Princeps und miles: Studien zum Bindungs- und Nahverhältnis von
Kaiser und Soldat im 1. und 2. Jahrhundert n. Chr. Hildesheim.
Stallknecht, B. . Untersuchungen zur römischen Aussenpolitik in der Spatantike.
Bonn.
Stein, E. . Histoire du Bas-Empire. Paris.
Stok, F., ed. . “Vita Donatiana.” In Vitae Vergilianae Antiquae, ed. G. Brugnoli
and F. Stok. Rome: –.
a. “Sulla tradizione della Vita Vergilii di Foca (F.S.).” In Studi sulle Vitae
Vergilianae, ed. G. Brugnoli and F. Stok. Pisa: –.
b. “Stemma Vitarum Vergilianarum.” In Studi sulle Vitae Vergilianae, ed.
G. Brugnoli and F. Stok. Pisa: –.
Strzelecki, L. . “De Phocae vita Vergiliana.” In Munera philologica Ludovico
Cwiklinski oblata. Poznan: –.
Stuart, D. R. . Epochs of Greek and Roman Biography. Berkeley, Calif.
Suerbaum, W., ed. . Die archaische Literatur: Von den Anfängen bis Sullas Tod.
Vol.  of Handbuch der lateinischen Literatur der Antike. Munich.
Syme, R. . Tacitus. Oxford.
. “Pliny’s Less Successful Friends.” Historia : –. Repr. in Roman
Papers, vol. , ed. E. Badian (Oxford, ), –.
. Roman Papers VII. Ed. A. R. Birley. Oxford.
Talbert, R. J. A. . Review of Matthews . BMCR . ..
Tiede, D. . The Charismatic Figure as Miracle Worker. Missoula, Mont.
Treggiari, S. . Roman Freedmen during the Late Republic. Oxford.
. “Digna condicio: Betrothals in the Roman Upper Class.” EMC : –.
a. “Ideals and Practicalities in Matchmaking in Ancient Rome.” In The
Family in Italy from Antiquity to the Present, ed. D. I. Kertzer and R. P. Saller.
New Haven, Conn., and London: –.
b. Roman Marriage: Iusti Coniuges from the Time of Cicero to the Time of
Ulpian. Oxford.
Tuilier, A., G. Bady, and J. Bernardi, eds. . Saint Grégoire de Nazianze:
Oeuvres poétiques. Vol. , pt. : Poèmes personnels II, 1, 1–11. Paris.
Turpin, W. . “The Purpose of the Roman Law Codes.” ZSS RA : –
.
Urbainczyk, T. . Socrates of Constantinople: Historian of Church and State. Ann
Arbor, Mich.
. Theodoret of Cyrrhus: the Bishop and the Holy Man. Ann Arbor, Mich.
Uytfanghe, M. van. . “L’Hagiographie: Un ‘genre’ chrétien ou antique tardif?”
AB : –.
Vaggione, R. P. . Eunomius of Cyzicus and the Nicene Revolution. Oxford.
 References 

Van Dam, R. . Leadership and Community in Late Antique Gaul. Berkeley,
Calif.
. Kingdom of Snow: Roman Rule and Greek Culture in Cappadocia. Philadel-
phia, Pa.
Van Nuffelen, P. . Un héritage de paix et de piété: Étude sur les histoires
ecclésiastiques de Socrate et de Sozomène. Orientalia Lovaniensia Analecta .
Leuven.
Vanderspoel, J. . Themistius and the Imperial Court: Oratory, Civic Duty, and
Paideia from Constantius to Theodosius. Ann Arbor, Mich.
Várady, L. . Das letzte Jahrhundert Pannoniens 376–476. Budapest.
Verboven, K. . The Economy of Friends: Economic Aspects of “Amicitia” and
Patronage in the Late Republic. Brussels.
Vessey, M. . “From cursus to ductus: Figures of Writing in Western Late
Antiquity (Augustine, Jerome, Cassiodorus, Bede).” In European Literary
Careers: The Author from Antiquity to the Renaissance, ed. P. Cheney and F. A.
de Armas. Toronto: –.
. “Sacred Letters of the Law: The Emperor’s Hand in Late Roman (Literary)
History.” Antiquité Tardive : –.
. “Latin Literary History after Saint Jerome: The Scriptorum illustrium
latinae linguae libri of Sicco Polenton.” Neulateinisches Jahrbuch/Journal of
Neo-Latin Language and Literature : –.
. “History, Fiction, and Figuralism in Book  of Augustine’s Confessions.”
In The Cultural Turn in Late Ancient Studies: Gender, Asceticism, and His-
tiography, ed. D. B. Martin and P. C. Miller. Durham, N. C., and London:
–.
. “Quid facit cum Horatio Hieronymus? Christian Latin Poetry and
Scriptural Poetics.” In Poetry and Exegesis in Premodern Latin Chris-
tianity: The Encounter between Classical and Christian Strategies of Inter-
pretation, ed. W. Otten and K. Pollmann. Leiden and Boston, Mass.:
–.
. “Jerome and the ‘Jeromanesque.’” In Jerome of Stridon: His Life, Writings
and Legacy, ed. A. Cain and J. Loessl. Ashgate: –.
Vidal, J. L. . “La biografı́a de Virgilio escrita por Focas.” Boletin de la Sociedad
Catellonense de Cultura : –.
. “La Vita Vergiliana de Focas: Biografı́a y poesia de escuela.” ExcPhilol. :
–.
. “Intención de autor y estilo de la época en la composición de la vita
Vergilii de Focas.” In Scripturus Vitam: Lateinische Biographie von der Antike
bis in die Gegenwart: Festgabe für Walter Berschin zum 65. Geburtstag, ed. D.
Walz. Heidelberg: –.
Voltaire, J. . Œuvres complètes. Nouvelle édition avec notices, préfaces, variantes,
table analytique. Vol. . Paris.
Walbank, F. W. . Review of V. La Bua, Filino-Polibio Sileno-Diodoro: il prob-
lema delle fonti dalla morte di Agatocle alla Guerra mercenaria in Africa. CR
: –.
 References

Wallace-Hadrill, A. . “Patronage in Roman Society: From Republic to

Empire.” In Patronage in Ancient Society, ed. A. Wallace-Hadrill. London
and New York: –.
Walsh, P. G. . “Introductory” [Later Principate]. In The Cambridge History of
Classical Literature. Vol. , ed. E. J. Kenney. Cambridge: –.
Ward Perkins, B. . The Fall of Rome and the End of Civilization. Oxford/New
York.
Waszink, J. H. . Biene und Honig als Symbol des Dichters und der Dichtung in
der griechisch-römischen Antike. Opladen.
Watson, A. . International Law in Archaic Rome: War and Religion. Baltimore,
Md.
. “Law and Society.” In Cairns and de Plessis : –.
Watts, E. . “Orality and Communal Identity in Eunapius’ Lives of the Sophists
and Philosophers.” Byzantion : –.
. City and School in Late Antique Athens and Alexandria. Berkeley, Calif.
. “Creating the Academy: Historical Discourse and the Shape of Commu-
nity in the Old Community.” JHS : –.
Wehrli, F. ed. . Die Schule des Aristoteles. Vol. . nd edn. Basle.
Wesch-Klein, G. . “Recruits and Veterans.” In A Companion to the Roman
Army, ed. P. Erdkamp. Oxford: –.
Westerink, L. G. . Anonymous Prolegomena to Platonic Philosophy. Amsterdam.
Whitby, M., ed. . The Propaganda of Power: The Role of Panegyric in Late
Antiquity. Leiden and Boston, Mass.
White, C. . Gregory of Nazianzus: Autobiographical Poems. Cambridge.
Wickham, C. . Framing the Early Middle Ages. Oxford.
Williams, G. W. . “Some Aspects of Roman Marriage Ceremonies and Ideals.”
JRS : –.
Williams, R. . Arius: Heresy and Tradition. London.
Williams, S., and G. Friell. . Theodosius: The Empire at Bay. New Haven,
Conn.
Woods, D. . “Valens, Valentinian I, and the Ioviani Cornuti.” Latomus :
–.
. “On the Alleged Reburial of Julian the Apostate at Constantinople.”
Byzantion : –.
Wormald, P. . “The Decline of the Western Empire and the Survival of Its
Aristocracy.” JRS : –.
Yakobson, A. . Elections and Electioneering in Rome: A Study of the Political
System of the Late Republic. Historia Einzelschriften . Wiesbaden.
Zecchini, G. . “Latin Historiography: Jerome, Orosius and the Western
Chronicles.” In Marasco : –.
Ziolkowski, J. M., and M. C. J. Putnam, eds. . The Virgilian Tradition: The
First Fifteen Hundred Years. New Haven, Conn.
 Index

Ablabius  Anglo-Saxons 

Abraham  Anicius Achilleus Glabrio Faustus 
Acacius  Annius Anullinus 
Academy ,  Antioch , , , , , , , , , ,
Accius  , , , , , 
Acholius  Antinoopolis 
Acta Alexandrinorum  Antiochus Chuzon , 
Actium  Antoninus Pius , 
adventus , ,  Antony , –, , 
Advocate  Aphrodisias n.
Aelius Donatus See Chapter ; , n., , Apollinarians 
, , –,  Apollo , 
Aelius Ulpianus n. Apollonius of Tyana 
Aeneas of Gaza  Appius Claudius 
Aetius n. Apuleius 
Africa , ,  Aquileia , , n.
Africanus  Aquitania , , , , 
agora –,  Arabia n.
Ahab  Aradius Rufinus 
Alamanni  Arbito 
Alans  Arbogast 
Alaric , , , –,  Arcadius , , , , , , , ,
Alexandria , , , , , n., , , , , , , , ; baths of 
, ; Catechetical School ,  Arcadius Chrisius 
Allia, battle of  Arcadius Rufinus 
Alps  Armenia 
Ambrose , , , , ,  Arianism , , , , , , , ,
amicitia , ,  , , , ; homoeans , 
Amiens  aristocracy: of Italy , , , ; Christian 
Ammianus Marcelinus , , , , , , , , Aristotle , 
, n., , , , , n., , , Aristoxenus 
, , , , , , , , , Armathaim 
; on Julian n., – Artemidorus 
Amos  Artemisia 
Amphilochus, Bishop of Iconium , n. Arthanaric 
Anaglypha Traiani n. Arveni 
Anastasia , ,  asceticism , , , , 
Anastasia, Chapel of  Asia , , 
Anatolia  Asia Minor 
Anaximander  assiduus 
Andrew  astrology 


 Index

Athanaric n.,  Brixia 

Athanasius , , , , , ; Life of Brown, Peter 
Antony , , –, ,  Brunt, Peter , –, , 
Athaulf  Brutus 
Athena ,  Burgundians 
Athens n., , , ; Athenians 
Attic Greek  Caecilius Statius 
Atticus  Caesarius 
Attila ,  Campania 
Attius Insteius Tertullus  capitatio 
auctoritas n.,  Cappadocia , , 
Augusteon  Cappadocian Fathers 
Augustine , , , –, , , , –, Caracalla –
, , ; Confessions , , –, , Carinus 
, , ,  Carosa 
Augustus , , , , , , n., , Carthage 
n., , ; law of adultery  Carthage 
Aulus Gellius  Carus 
Aulus Plautius  Cato Major , 
Aurelian , ,  Causidici 
Aurelius Hermogenes ,  Ceionius Rufus 
Aurelius Victor ,  Ceionius Volusianus 
Ausonius , , , , , , , , , Celsus 
, , , , , ; Ephemeris n., Celtes, Conrad , , 
, ; Professores ; Protrepticus n., cervisia 
 Chalcedon 
Autun  Chalke 
Auxentius –, n. Charito , , 
Aventine  Christ , , , 
Avidius Cassius  Christianity , , , , , , , , , ,
, , , , , , , , , ,
Babatha n. , , , , , , , , , ,
Balkans , , , , , ,  , ; authors ; Constantine –;
barbarians , , , , , , ; education , ; Imperial Court ;
invasions  Julian , , ; Roman Emperor ,
Basil of Caesarea , n.,  , ; in Syria 
Batavian Revolt n. Christmas 
beatitudo  Chronicon Paschale n.
Bellovacori  Chrysogonus 
beneficium  Cicero , , , , –, , , , , ,
Benenatus  ; Brutus , ; Hortensius 
Berytus  Cilicia 
Bethlehem  citizenship 
Bible , ; Vulgate  civilitas n., –
biography , , , , , , , , , classical world 
, , , ; classical , ; Greek Claudian , , , , , –, ,
; Hellenistic ; philosophical , , , 
, , , ; Virgilian , ,  Claudius II Gothicus 
bishops –, , , , , , , , , Claudius Mamertinus , 
, ; of Alexandria ; of Claudius Pompeianus , 
Constantinople ; Eastern ; of Rome Claudius Severus , , n.
,  client , , 
Bodin  Clio 
Bordeaux  Clusium 
Bosporus , , ,  Codex Gregorianus 
 Index 

Codex Justinianus  Diocletian , , , , , , , , n.,
collatio lustralis  , , , ; and the law 
Commentarii  Diodorus Siculus n.
Commodus  Dionysius of Halicarnassus –
concilium  Dios of Antioch 
Concordia  Dioscuri 
consistorium  Dmeir n., 
Constantia  doctiores 
Constantine See Chapters –; , , , , Domnica , , , 
, , , , , , , , , ; Drys 
Christianity , ; column of ; Duke of Newcastle 
Imperial Court ; Laws , , see Chapter
; mausoleum of ; military , , , , Easter 
–, , , see Chapter  economy 
Constantinople , , , , , , , Edict on Maximum Prices 
n., , , , , , , , , education , , , ; Roman ,
, , , , , , , –, , 
, , , , , , , , ; Egeria 
Arians of , , , ; church of , Egypt , , , n., , , , , ,
; Constantine and ; Council of –, ; Palmyrene invasion 
; Gregory of Nazianzus in , , , Elegiac 
, , , , , , ; Theodosius I Eleusis 
in , , , , , , , , , Eleutherios 
, , , , ; Valens ; Walls  Elijah 
Constantius I , , , , , , , , Eloquium , 
 eminentissimus n.
Constantius II , , , , , ,  Empedocles 
consul  empire 
Corbulo  Ennius , , 
court: Christianity and ; Byzantine ; Ephesus 
Imperial , , , , ,  Epiphany 
Ctesiphon  Epirus 
Culleus  Epitome de Caesaribus 
Curia  equites , 
curiales  Eriulph , 
cursus honorum  Ethiopians 
Cypria  Euergetes 
Cyrenaica  Eugenius , , , , , 
Cyril of Alexandria  Eunapius , , n., n., , ,
n., , 
Dacia ,  Eunomians 
Dalmatius  Eunomius n., , 
Damascius  Euphemia, church of 
Damasus of Rome  Euphrates 
Danube , , ,  Europe ; western 
Darius  Eusebia 
David  Eusebius , , , , , , , , ,
deditio in fidem –,  , , , ; Ecclesiastical History ;
Demophilus , , , , , , Life of Constantine , –, ; Life of
 Origen –, 
Demosthenes  Eutropius , , –, , , 
devotio  evocati 
dies imperii 
Digest of Justinian , ,  Fabius Pictor 
dignitas n.,  Faustus 
 Index

Festus , ,  Grammaticus , 

Fides , –, , ,  Gratian , , , , , , , , ,
filius familias ,  , , , , , , , , , 
fines  gravitas 
Flaccilla , , , , , ,  Great Church 
Flavianus  Gregoria 
Flavius Eusignius  Gregorius , , 
Flavius Neoterius  Gregory of Nazianzus See Chapters  and ; ,
Florus ,  , , , , , , ; De vita sua
foedus , n. –, –; on Julian –; orations
food  –
fortitudo  Gregory of Nyssa , , , 
Forum ; of Constantine , ; of Gregory of Tours , 
Strategius ; of Taursus ; of Greece ; Classical , 
Theodosius ; of Trajan ,  Greek language , n., , , , ,
Foucault, Michel  , , , 
Franks , ,  Greeks , , , 
Fravitta ,  Greuthungi , 
Frigidus , , 
Fritigern  Hadrian , , 
Fronto , ,  Hadrianople, Battle of , , , , ,
Fulvius  , , , , 
Hadrianus Sallustius 
Gabinius Barbarus Pompeianus n. Hagia Sophia 
Gainas  Halys 
Gaius (jurist)  Hannibal 
G. Annius Anullinus Geminus Percennianus  Hebdomon , , , 
Gaius Gracchus , , – Hebrew 
Galen  Helena , 
Galerius , ,  Helenianae 
Galilea  Hellenism 
Galla , , ,  Hellenistic World 
Galla Placidia  Heraclea n.
Gallienus ,  Heracleides of Pontus , 
Gaul , , , , , , , , n., , Hercules 
, , , , , , , , , , Herenius Modestinus 
; Roman , ,  Hermes 
Gavius Clarus  Hermogenian , , 
Geiseric  Hermopolis 
generals  Hermopolis Magna 
Germanicus , , , , n., n., Hesiod , 
– hexameter , 
Gesta Senatus ,  Hiero of Syracuse 
Gibbon, Edward , , , , , , Hilary of Poitiers n.
– Himerius 
gladiators  Hippodrome, Constantinople –, , ,
Gnaeus Calpurnius Piso  , 
Golden Gate  Historia Augusta , , , , , 
Golden Horn  historiography, Roman , , 
Gordian III  Holy Apostles, Church of , , , ,
Gothic column  , , 
Goths , , , , , –, , , , Homer , , n., , , , , ;
, , , , , , , , , ; in education n.; Iliad ; Odyssey 
Gothic War , –, ; treaty of , homoeans see Arianism
– honestas 
 Index 

honorati  Lactantius 

Honorius , , , , ,  late antiquity , , , , , , , , ,
Horrea Theodosiaca  , , , 
Hortensius  Latin , , , , , , , , ,
Huns  , ; historian ; poetry 
Latium 
Iamblichus , , , ; On the Law of the Praetorian Provinces 
Pythagorean Life  Letters , , , , , , –, , ;
Illyricum , ; Illyrians  epistula commendaticia , , ; legal ,
imperium –, –, ,  n.
impietas  lex Gabinia 
indulgentia , , ,  lex generalis , , 
India  lex Ovinia 
Insteius Tertullus  lex Papia 
International Theatrical Association of the lex Porcia 
Artisans of Dionysus  lex regia 
Isaac ,  Libanius , , –, , , , , , ,
Ismael  , n.
Istanbul  liberalitas 
Italy , , , , , , , , ; cities of Libertus 
 Licinian-Sextian Reforms 
Italian Allies  Licinianus n.
iustitia  Licinius , , , n., , , 
liturgy , , , 
Jerome , , , , , , , , , Livius Drusus , , 
, , –, , , ; Chronicle Livy –, n., , n., , , 
,  Lollianus Avitus 
Jerome of Stridon ,  Lucan 
Jerusalem , ,  Lucerius Verinus 
Johannes  Lucius Caesius 
John ,  Lucius Verus , 
John the Baptist ; Church of  Lucretius 
John Chrysostom , n.,  Lucullus , 
John of Lycopolis  Lugdunensis Prima 
John Lydus ,  Luke 
Jones, A. H. M. ,  Lycus River , 
Jordanes  Lydia , 
Josiah 
Jovian , , , , , ,  Macedonia , , , 
Judea  Macedonianists , , 
Julian , , , n., , , , , –, Macedonius 
, , , , , , , , , , Macrobius 
, , ; in Athens ; Christians , magic 
; death of ; Hellenism ; and Rome magister memoriae 
 magister officiorum 
Julianus Licinianus  maiestas n.
Julio-Claudians  Maiuma 
Julius Caesar –, ,  Magnus Maximus 
Julius Marcellinus  Mamertines –
Junius Flavianus  Manichees 
Junius Mauricus , n., ,  Manlius Rusticianus 
Jurists  Manuel 
Justin n. Marcian 
Justinian , , , , ,  Marcian of Gaza , n.
Juvenal , n.,  Marcus Agrippa 
 Index

Marcus Aurelius , , , ,  Neo-Platonism , 
M. Grunnius Corocotta  Neoterius , , 
Marcus Popillius Laenas  Nero 
Marinus n., n.,  Nexum 
Marius  Nicaea ; Council of , ; Nicene
Marius Maximus ,  Orthodoxy , , , , , , ,
Maro  , , ; Nicene Creed , 
marriage , , , , , ; women’s ages Nichomachus 
n. Nichomachus Flavianus , , , , , ,
Marseilles  ; father of 
Martial  Nicias (philosopher) –
Martyrion of Peter and Paul  Nicomedia 
Martyrius  Nisibis 
martyrs , ,  Nomothesia 
Matthews, John , ; Journey of Theophanes North Africa , , 
; Laying Down the Law –; Roman Empire Notitia Dignitatum 
of Ammianus , ; Western Aristocracies and Novatian 
Imperial Court –, ,  Numerian 
Maxentius ,  Nummius Tuscus 
Maximian , , 
Maximinus , , n. Obelisk of Constantinople 
Maximinus Daia , ,  Odaenathus 
Maximus , , , ,  Oenoanda 
Medes  Olybrius 
Mediterranean, East ,  Olympiodorus , , n., , ,
Meletius of Antioch , ,  
Memmius, Q. Aurelius  Olympiodorus of Thebes 
Menander  Orcistus 
Menander Rhetor ,  Origen n., –, , 
Merodobuus  Oriens 
Mesopotamia ,  Orosius , , , 
Messana  Otium 
Middle Ages ,  ousia 
Milan , ,  Ovid 
Milion , , 
military , , , , , , ,  Pacatus , , , , , , , ,
Millar, Fergus  
mime  pagan , , , , –, , , , ;
Milvian Bridge, battle of  biography , ; philosophers 
Minicius Acilianus , – paideia , 
Moesia  Palatine 
Monacensis II, Life of  Palatini , 
money  Palestine , 
Monica  Palmyra –
Montaigne ,  panegyric , , , , , , , ;
Montesquieu, Charles de , n.,  in verse 
Mores  panegyrici Latini , , –, , , ,
Moses , ,  , 
munificentia  Pannonia , 
Pantomime 
Namier  Paris 
Naples  parrhesia , , , 
Nazarius  paterfamilias , , 
Near East  patria potestas , , , 
Nectarius , , , , , ,  patronage , , , –, , , 
 Index 

Patruinus ,  possessors 

Paul, Bishop of Constantinople ,  potestas , 
Paul “the Chain”  praepositus 
Paul of Thebes  praeteritio 
Paulinus of Milan  Praetorian Edict 
Paulinus of Nola  priests , 
Paulinus of Pella ; see Chapter , , , , Principate , , , , –, 
, , , , , , , , , ; principia –, 
education of –, , ; Eucharisticos Priscus of Panium n., –
Deo sub ephemeridis meae textu , , , Priscus 
,  Probinus 
Paulus (author of Sententiae) , ,  Probus , 
Pax Christiana  Proclus n., , 
Pella ,  Procopius , , , , 
Peloponessus  Procopius of Gaza 
Peter of Alexandria  Proculus 
Petilius Cerialis  prodigy , , , , , 
Persia , , , , , , , , , Prohaeresius 
,  Prolegomena to Platonic Philosophy 
Persius  Proletarius 
Pertinax ,  proskynesis 
Peter, Hermann  protector domesticus 
philanthropia  Proxenus 
Philinus of Agrigentum , n. prudentia 
Philip (emperor) ,  Psalms 
Philippi, battle of ,  psogos 
philosophers , , , ; biography of Ptolemy Philadelphus 
 P. Alfenus Rufus , 
philosophy , , , ; Christian , , P. Septimius Demetrius 
; classical , ; in late antiquity  Pulcheria , 
Philostratus  Pythagoras , , , , 
Philostorgius 
Phocas , –, , –; Ars de nomine et Quintilian n., 
verbo ; debt to Donatus ; Life of Vergil Q. Cervidius Scaevola 
, ; students of , see Chapter  Quintus Fabius Maximus 
Phoenicia 
Phrygia  Radagaisus 
pietas  Ravenna 
Pindar  recusatio 
Plato , , , , , , ; Life of , relatio 
; Platonism , ,  Remi 
Platonists ,  Renaissance 
Plautus n.,  Renatus Profuturus Frigeridus , 
plebs  Restitutus 
Pliny (minor) , , , , –, –, , rhetor 
, ; panegyricus , ,  rhetoric , , , , ; schools of 
Polemo  Rhine , 
polis ,  Roman Comedy n.
politeia  Roman empire , , , , , , , , ,
Polybius  , , , , , , , , , ,
Pompey ,  , , , , , ; collapse of ;
Pomponius (jurist)  eastern , , , ; late , , , ,
Pontifex Maximus  , , , , , , , , , ;
Porphyry n., ,  oikoumene , ; provinces ; size of
Portico of Domninus  ; western , , , , 
 Index

Roman emperor , , , , , , , Sidonius Apollinaris n., 
; fourth century  Silenus n.
Roman law , , ,  Simeon Stylites , , , , 
Roman republic , , –, ; late , –, Singara 
; middle  Sirmium 
Romans , , ,  Skaptopareni 
Rome , , , , , , , , , , slavery , , , 
, , , , , , , , , , Socrates , , , , 
, , , , , , , , , Socrates (historian) , , , , , 
, , , , , ; education ; and Sophocles 
Julian ; Olympiodorus of Thebes in  Soter 
Romulus , , , , ,  Sozomen , , , , , , , 
Romulus Augustus  Spain , ; Spanish Orthodoxy 
Roscii of Amerinum  Sparta 
Rufinus , , , ,  spes 
Ruricius Pompeianus  Speusippus 
Ruteni  Stama 
Statilius Ammianus , n.
Sabel  statio 
Sabellianism  Statius 
sacer  Statius Rufinus 
Saint Eirene, Church of , ,  Stephen (martyr) 
Saint Mark, Church of  Stilicho , , , , 
Saint Menas, Church of  Succi pass , 
Salutaris  Suetonius , n., , , , , ,
salutatio , , , , , , n., ,  , –, , 
Salvian  Sueves 
Samia  suffragium , , 
Samuel  Sulla 
Sarus  Sulpicius Alexander , 
satire n. Sulpicius Cornelianus 
Saturninus , ,  Sulpicius Severus , 
scholastici  Syme, Ronald , , 
Scipios  Symmachus (major) 
Sea of Marmara  Symmachus, Q. Aurelius See Chapter ; , ,
secretarium ,  , , , , , , , , 
senate , ; at Rome , , , , n., Symmachus, Q. Aurelius Memmius , 
, , , , , ; at Constantinople Synesius , 
, , , , , , , , –, Syria , , , , ; army of 

Seneca  Tacitus , , , 
Seneca (minor)  Tacitus (emperor) 
Septimius  Tarsus 
Septimius Severus  taxes , , , , 
Septuagint ,  tax collectors 
Serdica  teachers , , , 
Serena  Temperantia 
Sergeric  Tenagio Probus
Servius  Terentius n., 
Servius Sulpicius Rufus  Tervingi , , 
Severan period , ,  Tetrarchy , , , , 
Severus  Thagaste 
Sextus Aurelius Victor ,  Thales 
Shakespeare  Thalassius , , 
Sibylline Oracles  theater –, 
 Index 

Themistius , , , , , –, , Valentinian I , , , n., n., , ,
, , , , , ; as orator –; , , , , , , 
as philosopher –; and Theodosius Valentinian II , , , 
–, , , ,  Valentinian III n., n., –,
Theodore  
Theodoret , –, , , , ; Religious Valerian 
History n. Vandals 
Theodorus  Varro 
Theodosian Code , –, –, , , , , Varronianus , 
, , , , ,  Vellius 
Theodosius I See Chapters –; , , , , Velvocorians , , , 
, , , n.; Constantinople –, Vespasian , , n.
, , , , , , , –; veterans See chapter ; , , , , , ,
ecclesiastical policies –, , , ; , –
and Gregory of Naziensus , , , , Via Appia 
; and military , , , , , , Via Salaria 
; compared to Valens –; Theodosius vicarious 
II , , , , , ,  Victor 
Theophanes , , , , n., n. Victoria 
Thessalonica , , , , , ,  Victorinus n., , ,
Theveste  
Thrace ,  Victorinus () , 
Tiberius  Vincentius 
Tigris  vires n.
Timothy  Virgil , , , , , , , , , ,
Titus  , , , , ; birth of , –;
Titus Tatius  Eclogues , –, –; father of , ,
Trajan , , n., , , , , ; , ; as an infant ; mother of , ,
Column of n.,  ; in Roman education n.
tribunal  Virgin, Church of the 
tribute  virginity ; virgin birth 
Trier , , ,  Voltaire n.
Trinity ,  Vulgate see Bible
triumvirate, second 
Troy  Wallace-Hadrill, Andrew , 
Twelve Tables , ,  Wills , ; of soldiers –
Tyrtaeus 
Xenocrates , 
Ulpian , , , , , , –, ,  Xenophon 
Ulysses 
utilitas  York 
Ursicinus , 
Zenobia 
Vacca  Zeus-Helios 
Valens , , , , , –, , , , Zono de’Magnalis, Life of 
, , , , ; and Arians , , Zosimus , , , , n., ,
, ; and Goths , ; death of , 
, , ,  Zoticus of Philadelphia , 