Cover

title:
author:
publisher:
isbn10 | asin:
print isbn13:
ebook isbn13:
language:
subject

The Barbarian Plain : Saint

Sergius between Rome and
Iran Transformation of the
Classical Heritage ; 28
Fowden, Elizabeth Key.
University of California Press
0520216857
9780520216853
9780585383873
English
Sargis,--Saint,--4th cent.-Cult, Rusafa (Extinct city)-Church history, Syria--

2/754

publication date:
lcc:
ddc:

subject:

Church history. , Iraq-Church history. , Christian

saints.
1999
BR1720.S23F68 1999eb
270.2/092
Sargis,--Saint,--4th cent.-Cult, Rusafa (Extinct city)-Church history, Syria-Church history. , Iraq-Church history. , Christian
saints.

Page a
This page intentionally left blank.

Page b
THE TRANSFORMATION OF THE CLASSICAL
HERITAGE
Peter Brown, General Editor
I Art and Ceremony in Late Antiquity,
by Sabine G. MacCormack
II Synesius of Cyrene: Philosopher-Bishop, by Jay Alan Bregman
III Theodosian Empresses: Women and
Imperial Dominion in Late Antiquity,
by Kenneth G. Holum
IV John Chrysostom and the Jews: Rhetoric and Reality in the Late Fourth
Century, by Robert L. Wilken
V Biography in Late Antiquity: The Quest
for the Holy Man, by Patricia Cox
VI Pachomius: The Making of a Community in Fourth-Century Egypt, by
Philip Rousseau

5/754

VII Change in Byzantine Culture in the

Eleventh and Twelfth Centuries, by A.
P. Kazhdan and Ann Wharton Epstein
VIII Leadership and Community in Late
Antique Gaul, by Raymond Van Dam
IX Homer the Theologian: Neoplatonist
Allegorical Reading and the Growth of
the Epic Tradition, by Robert
Lamberton
X Procopius and the Sixth Century, by
Averil Cameron
XI Guardians of Language: The Grammarian and Society in Late Antiquity,
by Robert A. Kaster
XII Civic Coins and Civic Politics in the Roman East, A.D. 180275, by Kenneth
Harl
XIII Holy Women of the Syrian Orient, introduced and translated by Sebastian
P. Brock and Susan Ashbrook Harvey

6/754

XIV Gregory the Great: Perfection in Imperfection, by Carole Straw

XV "Apex Omnium": Religion in the "Res
gestae" of Ammianus, by R. L. Rike
XVI Dioscorus of Aphrodito: His Work and
His World, by Leslie S. B. MacCoull
XVII On Roman Time: The Codex-Calendar
of 354 and the Rhythms of Urban Life
in Late Antiquity, by Michele Renee
Salzman
XVIII Asceticism and Society in Crisis: John
of Ephesus and "The Lives of the
Eastern Saints," by Susan Ashbrook
Harvey
XIX Barbarians and Politics at the Court of
Arcadius, by Alan Cameron and Jacqueline Long, with a contribution by
Lee Sherry
XX Basil of Caesarea, by Philip Rousseau

7/754

XXI In Praise of Later Roman Emperors:

The "Panegyrici Latini," introduction,
translation, and historical commentary by C. E. V. Nixon and Barbara
Saylor Rodgers

Page c
XXII Ambrose of Milan: Church and Court
in a Christian Capital, by Neil B.
McLynn
XXIII Public Disputation, Power, and Social
Order in Late Antiquity, by Richard
Lim
XXIV The Making of a Heretic: Gender,
Authority, and the Priscillianist Controversy, by Virginia Burrus
XXV Symeon the Holy Fool: Leontius's
Life and the Late Antique City, by
Derek Krueger
XXVI The Shadows of Poetry: Vergil in the
Mind of Augustine, by Sabine
MacCormack
XXVII Paulinus of Nola: Life, Letters, and
Poems, by Dennis E. Trout
XXVIII The Barbarian Plain: Saint Sergius
between Rome and Iran, by Elizabeth
Key Fowden

9/754

XXIX The Private Orations of Themistius,

translated, annotated, and introduced by Robert J. Penella
XXX The Memory of the Eyes: Pilgrims to
Living Saints in Christian Late
Antiquity, by Georgia Frank

Page d
This page intentionally left blank.

Page i

The Barbarian Plain

Page ii
This page intentionally left blank.

Page iii
The Barbarian Plain
Saint Sergius between Rome and Iran

Elizabeth Key Fowden

Page iv
Title page image: Rusafa. Aerial view from north.
University of California Press
Berkeley and Los Angeles, California
University of California Press, Ltd.
London, England
1999 by the Regents of the University of California
Library of Congress Cataloging-in-Publication Data
Fowden, Elizabeth Key, 1964
The barbarian plain : Saint Sergius between Rome and
Iran / Elizabeth Key Fowden.
p. cm. (The transformation of the classical
heritage ; 28)
Includes bibliographical references and index.
ISBN 0-520-21685-7 (alk. paper)
1. Sargis, Saint, 4th centCult. 2. Rusfa (Extinct
city)Church history. 3. SyriaChurch history.
4. IraqChurch history. I. Title. II. Series.
BR1720.S23F68 1999
270.2'092dc21

15/754

[B]

99-22672
CIP

Manufactured in the United States of America

08 07 06 05 04 03 02 01 00 99
10 9 8 7 6 5 4 3 2 1
The paper used in this publication meets the minimum
requirements of ANSI/NISO z39.48-1992 (R 1997)
(Permanence of Paper).

Page v
In memory of my father
R. E. Key
grandfathers
R. W. Key
K. W. Crusius
and godfather
P. O. A. Sherrard

Page vi
City built on a high mountain, which cannot be taken!
Fortress of our country, by which the whole region is
protected!
Beautiful name, Sergius, which is beloved to all ears,
and whose story is the favorite and the pride of the
populace.
JACOB OF SARUG (D. 521), MIMRA ON THE
VICTORIOUS SERGIUS
AND BACCHUS 117118, TR. ANDREW PALMER

Page vii

CONTENTS
LIST OF ILLUSTRATIONS

ix

PREFACE

xi

ACKNOWLEDGMENTS

xiii

NOTE ON TRANSLITERATION

xv

ABBREVIATIONS

xvii

MAPS

xx

Introduction

1. Portraits of a Martyr

The Passio of SS. Sergius and Bacchus

History in the Passio: The Emperor

11

The Hagiographic Milieu

17

Alexander of Hierapolis and the Early Cult26

of S. Sergius

19/754

Images of S. Sergius

29

2. Martyr Cult on the Frontier: The Case of45

Mayperqat
Relics and Defense

45

At the World's Center

48

The Ecumenical City of Martyrs

52

The Iranian Viewpoint

56

3. Rusafa

60

The Frontier Zone

60

A Place of Convergence

67

Page viii
The Fortress-Shrine

77

Pilgrimage to Rusafa

92

4. The Spread of the Sergius Cult in Syria and101

Mesopotamia
The Hawran

105

West of the Steppe

112

Northern Syria and Mesopotamia

117

The Iranian Empire

120

5. Frontier Shrine and Frontier Saint

130

Justinian, Theodora, and S. Sergius

130

Iranian Interest in S. Sergius

133

The Ghassanid Confederation

141

al-Mundhir

149

6. The Cult of S. Sergius after the Islamic174

Conquest

21/754

Umayyad Rusafa

174

Rusafa after Hisham

183

Between East and West

185

Between Christianity and Islam

189

BIBLIOGRAPHY

193

INDEX

219

Page ix

ILLUSTRATIONS
Rusafa. Aerial view from north.

title
page

1. SS. Sergius and Bacchus. Silver flask from30

the Kaper Koraon treasure, Syria.
2. SS. Sergius and Bacchus. Encaustic icon31
from the Monastery of S. Catherine, Sinai.
3. S. Sergius (?). Silver bowl from Cyprus.

33

4. S. Sergius. Mosaic, basilica of S. Demetri-34

us, Thessalonica.
5. The imperial couple Honorius and Maria,36
later inscribed as SS. Sergius and Bacchus.
Rothschild cameo.
6. a: S. Sergius. Bronze attachment. b: S.37
Sergius. Bronze bracelet. c: S. Sergius (?).
Lead seal. d: "Holy rider." Silver armband.

23/754

7. Rider god. Stone relief found south of41

Damascus.
8. Rusafa: Plan of settlement.

79

9. Rusafa: Plan of basilica A.

81

10. Rusafa: Plan of basilica B.

88

11. Rusafa: al-Mundhir building, interior150

from west.
12. Rusafa: Plan of al-Mundhir building.

152

13. Rusafa: al-Mundhir building, apse.

161

14. The Armenian monastery of the Apostle170

Thaddeus in northwest Iran surrounded by
pilgrims' tents for the patronal festival.
15. Rusafa: Plan of basilica A and mosque. 176
16. Silver chalice found at Rusafa.

186

Page x
This page intentionally left blank.

Page xi

PREFACE
My interest in the cultural landscape of Syria grew out
of a visit to southeastern Turkey and northern Syria in
1989, after a year spent in Greece. Looking down on
the Euphrates from Samosata, not yet inundated, I
was struck by the diversity of the landscapes and
communities that the river would pass before dissipating itself in the marshes of southern Iraq. The idea of
using the cult of Saint Sergius as a way into these
overlapping communities arose, gradually, after my
study of Syrian epigraphy with Glen Bowersock in
1991.
While writing this book, I have been kept in good company by three groups of people, variously enmeshed
with Syria. Each, I believe, has left its impression.
They resist chronological classification, so I shall begin with the representatives of the French tradition of
scholarship in Mandate SyriaRen Dussaud, Jean
Sauvaget, Daniel Schlumberger, Henri Seyrigwho
set the standard for fusing literary, archeological, architectural, and topographical evidence in historical
enquiry, broadly writ. The second presence here is

26/754

Princeton University, both as the sponsor of the Princeton Archaeological Expedition to Syria at the beginning of the twentieth century, and as the academic institution in whose Prentice, Firestone, and JonesPalmer libraries much of this book's framework was
laid. At Princeton, conversations with Glen Bowersock, Ted Champlin, Danny
, Oleg Grabar, and
especially my supervisor, Peter Brown, were always
stimulating, even if the sediment stirred settled into
new shapes only once I was back at my home in
northern Euboea.
Unlike the first two, the third presence reaches back
much further than a century, back, at least, to King
Abgar of Edessa. This is the Syrian Orthodox community of Aleppo, guided by its excellent metropolitan
Mor Yohanna Grigorios Ibrahim. Five months with the
Aleppo community taught me

Page xii
much about historical memory and community identity, as I waded through the torrential winter floods to
sing Syriac morning hymns with the children at S.
George's under the zealous leadership of Malfono
Abrohom Nuro; read S. Ephrem with the metropolitan
before his daily stream of petitioners began to flow;
and traveled in convoy to the jubilant consecration in
Qamishli of the new archbishop and patriarchal vicar
for the eastern United States. For their generosity and
interest, I warmly thank Metropolitan Yohanna and the
Syrian Orthodox community of Aleppo.
One of the great pleasures of this project has been
the glimpse it has allowed me into the world of Syriac
scholarship. I was fortunate to begin Syriac with John
Marks at Princeton in his last year before retirement.
None of the Syriac literature concerned with S. Sergius has been previously studied, and I owe much gratitude to Sebastian Brock and Andrew Palmer, to
whom I several times turned with questions relating to
these texts. For matters Arabic (as well as countless
others), I have benefited from Garth Fowden's enthusiasm for both that language and S. Sergius. Where
not otherwise attributed, translations from Arabic are

28/754

his. al-Isfahani's Kitab al-aghani served as the leitmotif of our stay in Aleppo, and some of the labor exerted
there has borne fruit in this book.
The German archeologists and architectural historians
involved with Rusafa have been exceptionally generous in their response to my work. I would like to thank
their leader, Thilo Ulbert, as well as Gunnar Brands
and Stephan Westphalen. I am also indebted to
Jonathan Rae and George 'Arab of the International
Center for Agricultural Research in Dry Areas, with
whom we spent invaluable hours in the north Syrian
steppe. I am grateful to Glen Bowersock, Peter
Brown, Ted Champlin, Garth Fowden, and James
Howard-Johnston for reading and commenting on the
entire typescript at various stages; as well as to Tom
Sinclair and Joel Walker, who commented on an earlier draft of chapter 2, and to Gunnar Brands and Thilo
Ulbert, who did the same for earlier drafts of chapter 3
and the end of chapter 5. None of these should be
held responsible for authorial error or gamble.
Limni, Euboea, Greece
21 December 1998

Page xiii

ACKNOWLEDGMENTS
I have been able to travel and conduct research in
Syria, Turkey, Jordan, and Palestine thanks to the
Robert F. Goheen Prize in Classical Studies of the
Woodrow Wilson National Fellowships Foundation, as
well as travel grants from the Association of Princeton
Graduate Alumni, the Department of Classics, the
Group for the Study of Late Antiquity, and the Seeger
Fellowship Fund of the Program of Hellenic Studies,
all at Princeton University. A Mellon Fellowship in the
Humanities and a Dissertation Grant from the National
Endowment for the Humanities allowed me two free
years for the completion of my dissertation. Subsequently, I was able to spend five months in Syria
thanks to the Olivia James Travelling Fellowship of
the American Institute of Archaeology, and enjoyed a
full year as the Gennadeion Fellow at the American
School of Classical Studies at Athens, working out the
implications of my fieldwork and writing the final
chapter on S. Sergius in Islamic Syria. Finally, thanks
to a research fellowship at the Centre for Greek and
Roman Antiquity at the National Research Foundation
in Athens, I have had the opportunity to be part of an

30/754

academic community engaged in exemplary interpretation of material evidence in its physical and literary
context. I would like to express my gratitude in particular to Miltiades Hatzopoulos, Louisa Loukopoulou,
and Athanasios Rizakis
For permission to illustrate objects or reproduce photographs I am obliged to the Walters Art Gallery, Baltimore, Maryland (fig. 1); the Trustees of the British
Museum (fig. 3); the Ninth Ephoria of Byzantine
Antiquities, Thessaloniki (fig. 4); the Museum of Art
and Archaeology, University of MissouriColumbia
(fig. 6a); Janet Zacos (fig. 6c); the Royal Ontario Museum (fig. 6d); Verlag Philipp von Zabern (figs. 8, 9,
12, and 15); Richard

Page xiv
Anderson, courtesy of Dumbarton Oaks (fig. 13);
Domos Publications, Athens (fig. 14); Thilo Ulbert,
courtesy of Deutsches Archologisches Institut
Damaskus (fig. 16).
Sources for the photographs are R. Mouterde and A.
Poidebard, Limes de Chalcis (Paris, 1945), pl. LXXIV
(title page image); K. Weitzmann, The Monastery of
Saint Catherine at Mount Sinai: The Icons, I: From the
Sixth to the Tenth Century (Princeton, N.J., 1976), pl.
XII (fig. 2); J. Meischner, AA (1993), pl. 1 (fig. 5); C.
Mondsert, Syria 37 (1960), fig. 4 (fig. 6b); G. Zacos
and A. Veglery, Byzantine Lead Seals (Basel, 1972),
no. 2975 (fig. 6c); G. Vikan, DOP 38 (1984), fig. 10
(fig. 6d); P. Ronzevalle, CRAI 1904, fig. 1 (fig. 7); J.
Kollwitz, AA (1957): 70, fig. 2 (fig. 10); author (fig. 11).

Page xv

NOTE ON TRANSLITERATION
In my transliterations of Syriac and Arabic I have kept
diacritics to a minimum. Syriac and Arabic titles are
transliterated where that form is well known; otherwise, they are cited in the established Latin or English
translation. Where available, translations of the oriental sources are included in the bibliography. Greek authors and sources are given their conventional Latin
spelling and titles.
In cases where both ancient and modern names are
known for a site, both appear, in that order, on first occurrence in the text, but only the ancient name appears subsequently. Map 3 uses modern names,
cross-referenced in the index where possible, since
the majority of the Hawranian sites I discuss are
known only by their Arabic names.

Page xvi
This page intentionally left blank.

Page xvii

ABBREVIATIONS
Abbreviations follow the conventions of Prosopography of the Later Roman Empire, ed. A. H. M.
Jones, J. R. Martindale, J. Morris et al., and F. R.
Adrados, Diccionario Griego-Espaol; for periodicals,
see L'Anne philologique. Note also the following:
AAS

Annales archologiques syriennes

(Damascus)

ACOec.

Acta conciliorum oecumenicorum,

ed. E. Schwartz (Strasbourg, 1914)

Acta
Sanct.

Acta Sanctorum (Antwerp 1643)

AMS

Acta martyrum et sanctorum, ed.

P. Bedjan (Paris, 189097)

BAFIC

I. Shahd, Byzantium and the

Arabs in the Fifth Century (Washington, D.C., 1989)

35/754

BAFOC

I. Shahd, Byzantium and the

Arabs in the Fourth Century
(Washington, D.C., 1984)

BA-IAS

Bulletin of the Anglo-Israeli Archaeological Society (London)

BASIC

I. Shahd, Byzantium and the

Arabs in the Sixth Century (Washington, D.C., 1995)

BHG

Bibliotheca hagiographica graeca,

ed. F. Halkin (Brussels, 1957)

BHO

Bibliotheca hagiographica orientalis, ed. P. Peeters (Brussels,

1910)

Bib.Sanct.Bibliotheca
196169)

Sanctorum

(Rome,

BIFAO

Bulletin de l'Institut franais

d'archologie orientale (Cairo)

BMGS

Byzantine and Modern

Studies (Birmingham)

Greek

36/754

BRL

Bulletin of the John

Library (Manchester)

Rylands

BSOAS

Bulletin of the School of Oriental

and African Studies (London)

Bull.

Bulletin pigraphique (Paris)

CArch

Cahiers archologiques (Paris)

Page xviii
CHIran

The Cambridge History of Iran,

ed. W. B. Fisher et al.
(Cambridge, 196891)

Chron 1234 Chronicum ad annum Christi

1234 pertinens, ed. J.-B. Chabot
(Louvain, 1920), and tr. (Louvain,
1937)
CIG

Corpus inscriptionum graecarum (Berlin, 182877)

CIL

Corpus inscriptionum latinarum

(Berlin 1863)

CIMRM

Corpus inscriptionum et monumentorum religionis mithriacae,

ed. M. J. Vermaseren (The
Hague, 195660)

CSEL

Corpus scriptorum
icorum latinorum

DACL

Dictionnaire d'archologie chrtienne et de liturgie (Paris,

190753)

ecclesiast-

38/754

DaM

Damaszener
(Mainz)

Mitteilungen

DCAE
(Athens)
DHGE

Dictionnaire de l'histoire et de la
gographie ecclsiastiques (Paris
1912)

Dict.
deDictionnaire de thologie caththologie olique (Paris, 190950)
EI1

The Encyclopaedia of Islam1, ed.

M. Th. Houtsma et al. (Leiden,
191338)

EI2

The Encyclopaedia of Islam2, ed.

H. A. R. Gibb et al. (Leiden,
1960)

Gatier,
P.-L. Gatier, Inscriptions de la
InscriptionsJordanie, vol. 2: Rgion centrale
(Amman,
Hesban,
Madaba,
Main, Dhiban) (Paris, 1986)

39/754

Hist.Ahud. History of Ahudemmeh, ed. F.

Nau, PO 3.1.751
IEJ

Israel
Exploration
(Jerusalem)

Journal

IGLS

Inscriptions grecques et latines

de la Syrie, ed. L. Jalabert, R.
Mouterde et al. (Paris, 1929)

JA

Journal asiatique (Paris)

JECS

Journal of Early Christian Studies (Baltimore)

JRAS

Journal of the Royal Asiatic Society (London)

MDOG

Mitteilungen
der
Orient-Gesellschaft
(Berlin)

Deutschen
zu Berlin

MGH, SRM Monumenta germaniae historica,

Scriptores
rerum
Merovingicarum

40/754

MHR

Mediterranean
Review (Tel Aviv)

Historical

MUSJ

Mlanges de l'Universit SaintJoseph (Beirut)

NJKA

Neue Jahrbcher fr das klassische Altertum

Nouvelles
archives

Nouvelles archives des missions

scientifiques et littraires (Paris)

OC

Oriens christianus (Wiesbaden)

OCP

Orientalia christiana periodica

(Rome)

PAES

Princeton University Archaeological Expedition to Syria in

19041905 and 1909 (Leiden,
190722)

Page xix
PEQ

Palestine Exploration Quarterly

(London)

PG

Patrologia graeca, ed. J. P.

Migne (Paris, 185766)

PL

Patrologia latina, ed. J. P.

Migne (Paris, 184464)

PLRE

The Prosopography of the Later

Roman Empire, ed. A. H. M.
Jones (vol. 1), J. R. Martindale
(vols. 13) and J. Morris (vol. 1)
(Cambridge, 197192)

PO

Patrologia orientalis, ed. R.

Graffin and F. Nau et al. (Paris,
1907)

P-OChr

Proche-orient
(Jerusalem)

chrtienne

Prentice,
W. K. Prentice, Greek and Latin
Greek andInscriptions, Publications of an
Latin
American
Archaeological
Inscriptions

42/754

Expedition
to
Syria
in
18991900, part 3 (New York,
1908)
PSAS

Proceedings of the Seminar for

Arabian Studies (Cambridge)

RAC

Reallexikon fr Antike und

Christentum (Stuttgart, 1950)

RE

Real-Encyclopdie der klassischen Altertumswissenschaft, ed.

A. Pauly, G. Wissowa, and W.
Kroll (Stuttgart, 18931980)

REArm

Revue des tudes armniennes

(Paris)

RSO

Rivista degli
(Rome)

SC

Sources chrtiennes (Paris)

SEG

Supplementum
epigraphicum
graecum (Leiden, 1923)

studi

orientali

43/754

Waddington,W. H. Waddington, Inscriptions

Inscriptions grecques et latines de la Syrie
(Paris, 1870)
ZDMG

Zeitschrift der Deutschen Morgenlndischen

Gesellschaft
(Leipzig)

ZDP-V

Zeitschrift
des
Deutschen
Palstina-Vereins (Weisbaden)

Page xx

Map 1. The eastern Mediterranean world in late

antiquity.

Page xxi

Map 2. Sergius sites in Syria and Mesopotamia.

Page xxii

Map 3. Sergius sites in the Hawran.

Page 1

Introduction.
In the sixth century A.D., the Greek-speaking world
knew the Syrian steppe that spread out around
Rusafa as the "Barbarian Plain."1 Rusafa was situated in the frontier zone between the great late antique empires of Rome and Sasanian Iran, a geographical position that gave the fortified settlement
political and cultural importance for the Romans, for
the Iranians, and for their Arab allies who inhabited
the region. All these parties sought in varying ways,
from military invasion to more subtle cultural penetration, to stake claims to this crucial region. It was at
Rusafa in the Barbarian Plain that the cult of S. Sergius was sown and gradually took root in the region's diverse communities. The widespread dissemination of
reverence for the martyr offers a rare opportunity to
view the overlapping of the cultural traditions that
flourished in late antique Syria and Mesopotamia.
Viewed on the ground, Rusafa occupies the near-center of Syria and Mesopotamia combined. But on the
mental maps of the region's diverse inhabitants and
visitors, Rusafa's position varied. To a wandering

48/754

monk from Constantinople, Rusafa belonged to the

"Persian desert"; to a Western pilgrim, it lay in "Saracen country" just beyond the Holy Land's eastern limit;
the orbit of influence claimed by a Nestorian monastery east of the Euphrates extended as far west as
Rusafa; while for the Arabs of the Banu Ghassan, the
fortress-shrine marked the northeasternmost reach of
their territory.
The Barbarian Plain is a space within the much greater plain that is Syria-Mesopotamia, an enormous,
variegated area that owes what unity it possesses to
the mountains around itfrom the hills of Palestine to
the
1.
. Proc., BP 2.5.29; Theoph. Sim., Hist.
5.13.3; Evagr., HE 6.21; Vita Gulanducht 2426. For
further discussion, see pp. 6566.

Page 2
Amanus on the west, the Taurus on the north, and the
Zagros on the east. To the south the plain blurs into
the Syro-Arabian desert. The mountains, which form
an arch around the plain, did not act as impenetrable
barriers, since trade and transhumance linked Syria
and Mesopotamia with lands beyond.2 The mountains' abruptness could impose an obstacle, but a
more significant determinant of a mountain's permeability was the cultural character of those who
dwelt in or around it. The great Zagros range
stretches from the Persian Gulf to the Caspian Sea
and divides the Syro-Mesopotamian plain from the
harsher Iranian plateau, the stronghold of a more homogeneous Iranian culture than ever flourished in the
plain below. Just below the Caspian Sea, the Zagros
links up with the Taurus range, which reaches as far
as southwestern Anatolia. The Taurus, abrupt but not
impervious, rises high above the Syro-Mesopotamian
plain. At the plain's northwesterly limit, the Orontes
river flows out to the Mediterranean between Amanus
and Mount Cassius. Caught between the sea and the
mountains, the westward-looking coastal plain was
married to the Mediterranean world and to some

50/754

extent alienated from the experience of Syria and

Mesopotamia east of the mountains.
Bounded on the west by this mountain belt, the high
Syro-Mesopotamian plain is always much more elevated, at ca. 500 m, than the Mediterranean coastal
plain. Its altitude, its isolation from the sea's temperate
influence and its exposure to the Syro-Arabian desert
to the south combine to create an extreme continental
climate, in which temperatures may vary dramatically
according to season, but also within twenty-four
hours. Most of the Syro-Mesopotamian plain is best
described as steppe, often treeless. Its seasonal
grasses and bushes can be surprisingly abundant
after rains, which, in turn, determine the movements
of pastoralists with their flocks. Steppeland differs
from desert in potential. For practical purposes, the
steppe is defined here as potentially cultivable arid
land. The desert too sprouts grasses, but thanks to
occasional wells, springs, and seasonal rains, the
steppe can, in places, support cultivation. The SyroMesopotamian steppe is crossed by a network of
ridgesQalamun, Abu Rujmayn, Bishriextending
northeast from the Anti-Lebanon mountains to the Euphrates. The north Syrian steppe undulates eastward

51/754

toward Mesopotamia with little interruption, save salt

marshes, basalt outcrops, and scattered tells. Low
ridges and wadis punctuate the gravel- and stonecovered south Syrian steppe, which merges with the
great Syro-Arabian desert.
Because of its size and diversity, the Syro-Mesopotamian plain could not
2. For particularly memorable evocations of the cultural unity of Syria-Mesopotamia in late antiquity, see
Garsoan, CHIran, 3: 56875, and Brown, Religion
and Society, 1015. See also Howard-Johnston in
Byzantine and Early Islamic Near East, ed. Cameron,
3: 157226, for a study of the Sasanian empire that
emphasizes the role of geography in its political
institutions.

Page 3
be dominated by one group of its inhabitants. Instead,
pastoralists, farmers, craftsmen, merchants, monks,
and soldiers were joined in a network of symbioses.
The eastern, northern, and western fringes were
densely populated in many places, and their inhabitants, especially in the eastern half, were very diverse.
They included Zoroastrians, Jews, polytheists, and
Christians of Nestorian, Chalcedonian, and nonChalcedonian ("monophysite") allegiance. Many of
both the permanent and of the less permanent inhabitants of Syria and Mesopotamia were mobile by vocationpastoralists,
semi-pastoralists,
merchants,
teachers, students, and pilgrims.
Seen from the outside, life in the more arid regions of
the steppe, and even more so in the desert, is restless. Settled populations often view the steppe as an
obstacle, a sterile, alien territory. Much of this impression derives from ignorance of the landscape. Knowledge of the steppe's variety and potential was the basic tool of pastoralists, and this knowledge had to be
both detailed and local, but at the same time command the grander patterns of the steppe. Because
movement in the steppe, whether for seasonal

53/754

migration, commercial interests, warfare, or pilgrimage, frequently took one through lonely expanses,
great importance was assumed by points of convergence, especially round water sources. In the great
Syro-Mesopotamian spaces, these fixed places were
flash points where all forms of human interactionsocial, economic, political, and religioustended to
concentrate.
Within this landscape, the political frontier between
the Roman and Iranian empires acted as an artificial
divide. In order to maintain their territorial claims, both
empires turned to military alliances with the Arab inhabitants of the region and to the construction of fortifications. It is in part the intention of this study to
demonstrate how from the point of view of Romans,
Iranians, and Arabs alike, divine defense went hand in
hand with arms and walls, a fact of late antique history
often overlooked or ignored. If one of the historian's
goals ought to be to "attempt to ascertain contemporary thinking and values and then look at how these
were realized in practice," we cannot afford to project
onto our evidence a separation of religious belief and
political or military action.3 Unless we take seriously
the widely held belief in the power of saints and their

54/754

relics to influence the political and military history of

the frontier zone, any attempt to explain developments in the region and their broader impact will be
one-dimensional.
Procopius names towers and fortresses throughout
the Roman empire that bore the names of saints.4 In
Roman Syria, it was not uncommon for
3. The quotation is from Isaac in Byzantine and Early
Islamic Near East, ed. Cameron, 3: 150.
4. See Proc., Aed. 4.4, for fortresses dedicated to S.
Sabianus and S. Stephen in Epirus Nova, one to S.
Sabianus and two to S. Donatus in Epirus Vetus; ibid.
4.11, to S. Trajan, S. Julian, and S. Theodore in
Thrace; ibid. 5.7, to S. Cyril in Scythia.

Page 4
military architecture to be placed under the protection
of Christian saints. For instance, between Chalcis /
Qinnasrin and Emesa / Homs, three fortified posts
dating from the late fifth and sixth centuries
were associated with holy patrons: the archangel Michael and the centurion Longinus at Burj ("tower" in
Arabic); S. Longinus along with S. Theodore and S.
George at Ghur; and, finally, the martyr Sergius, dedicatee of a metaton at Raphanaea / Rafniya, a vital
station in a pass through the Anti-Lebanon.5 At another Burj, near Dumayr, northeast of Damascus, an inscription relates that the Ghassanid phylarch alMundhir dedicated a tower to S. Julian.6 Saints
Longinus, Theodore, George, Sergius, and Julian all
distinguished themselves as soldier-martyrs, and all
earned respectable followings in the Roman East.7
What was unusual about the soldier-martyr Sergius
was the breadth of his appeal: by the sixth century,
his reputation had spread beyond Roman territory into
Iran, as well as among the Arab tribes between the
two empires. S. Sergius seems to have been politically colorblind. S. George also attracted devotees in
the Sasanian empire. But although the soldier saints
George and Sergius were sometimes confused, it was

56/754

S. Sergius and not S. George who was chosen by

political leaders in the region as a means of affirming
their authority. Vital to this development was the fact
that in hagiographical tradition, S. Sergius had been
executed during the Great Persecution at the fort of
Rusafa near the middle Euphrates.
This study of the cult of S. Sergius in Syria and Mesopotamia focuses on the nature of the cult in the frontier zone in late antiquity. It by no means aspires to be
a comprehensive account of the Sergius cult
wherever and whenever it took root, which would take
us as far east as Mongolia in geographical space,8
and in time right up to the present day, inasmuch as
together with George and Thecla, Sergius remains
one of the most revered saints in the Middle East
among both Christian and Muslims. The Barbarian
Plain consists of six chapters. In the first, an examination of the Passio and artistic images of S. Sergius attempts to reconstruct the rise of the martyr's cult at
Rusafa in the fifth century. With S. Sergius established at Rusafa, chapter 2 takes a broader view of
the increasing interest in martyrs in connection with
frontier defensethe particular example chosen to illustrate this development is Bishop Marutha's cult of

57/754

Iranian martyrs at Mayperqat. This provides the necessary context for the buildup of S. Sergius's cult at
Rusafa. The third chapter discusses the historical and
archeological evidence for the Sergius cult at Rusafa:
the strategic importance of the site at
5. For references and discussion, see pp. 11314
below.
6. Waddington, Inscriptions, no. 2562c. On alMundhir's tower, see Shahd, BASIC, 1: 49550,
52426.
7. On the cult of military saints, see Delehaye, Lgendes grecques, and Orselli, Santit militare.
8. For the easternmost appearance of S. Sergius, see
Mingana, BRL 9 (1925): 297371.

Page 5
Rome's eastern limit; Rusafa's place in trade and
transhumance patterns; and its significance as a pilgrimage center. With the growth of pilgrimage, the
Sergius cult spread outward from the site of his martyrdom, and chapter 4 traces the material and literary
evidence for the saint's popularity in Syria and Mesopotamia. In chapter 5, the leaders of Rome, Iran, and
the Arab tribes that dwelt between them take center
stage, and the discussion focuses on their attempts to
tap into the martyr's popularity and influence in Syria
and Mesopotamia for their own political ends. The final chapter is concerned with the Sergius cult after
the Islamic conquests, joining together disparate evidence to reconstruct a picture, not only of the continuity
of Christian cult, but also of the symbiosis achieved by
Christian and Muslim followers of S. Sergius.
"Barbarian Plain" is, of course, an outsider's label. It is
useful, though, since in the name itself the steppe is
fused with its inhabitants. While much of this book focuses on the empires of Rome and Iran, in whose
frontier zone the cult of S. Sergius flourished, one of
the most important factors that emerges is the role of
architecture and fixed points, such as Rusafa, in the

59/754

lives of the pastoralists and semi-pastoralists who inhabited the frontier zone. This study is offered as one
step toward the history of what Fergus Millar has
called "the marginal zone between the steppe and the
cultivable land."9
9. Millar, Roman Near East, 333.

Page 6
This page intentionally left blank.

Page 7

ONE
Portraits of a Martyr
Not long before 431, Bishop Alexander of Hierapolis /
Manbij in northern Syria invested three hundred
pounds of gold to build a church in the middle of the
Syrian steppe, a region he himself described as a wilderness. The church rose up within the walls of the
fortress called Rusafa and was dedicated to the martyr Sergius. This event is the first evidence we can
comfortably associate with S. Sergius and his
cultbefore the church's construction, there is silence
in the sources. Afterward, the cult gained impetus and
the Passio of SS. Sergius and Bacchus was written
down. Gradually, the cult of S. Sergius established itself as a prominent feature in the cultural landscape of
late antique Syria and Mesopotamia. Rusafa, located
about 30 kilometers south of a bend in the middle Euphrates, lay at the eastern limit of the Roman empire,
at the western limit of the Iranian empire, and near the
center of the Syro-Mesopotamian world. The shrine
attracted a wide range of patrons, including the

62/754

Roman emperor Justinian, the Iranian king of kings

Khusrau II, and the Ghassanid phylarch al-Mundhir.
Already in the fifth century, the martyr's influence had
begun to radiate outward from Rusafa across Syria
and Mesopotamia. It was to extend as far afield as the
foothills of the Zagros Mountains, Armenia, Egypt,
and Gaul. But S. Sergius's prominence in fifththrough seventh-century Syria and Mesopotamia remained closely linked to his principal shrine's location
in a region where diverse cultures mingled and two
great powers, Rome and Iran, confronted each other.
The Passio of SS. Sergius and Bacchus relates the
martyrdom of both saints and the miracles worked by
S. Sergius after his death. Hagiography is a notorious
minefield for the historian, and we would be wise to
approach the Passio from the perspective of the fifthcentury world it reflectsthat is, of the time of its composition. Anachronism and confusion in the Passio

Page 8
reflect the quality of the author's information and
should be seen within the framework of his purpose,
which was to describe the crowning of a martyr. In
such a work, accounts of events might be historically
inaccurate, but it did not matter as long as they fulfilled their purpose of setting the symbolic scene in
which the current of God's redemptive grace could
flow through the miracle-working saint to his followers.
This Passio seems to have done just that. The course
of history is influenced as much, if not more, by what
is perceived to have happened in the past as by what
can perhaps be documented from written sources.
What is of concern to the historian of the cult of S.
Sergius is the martyr's posthumous reputation.
Whether or not there was a soldier named Sergius
who was martyred outside Rusafa, and whether this
was at the command of the emperor Galerius, Maximinus Daia, or even Julian, was more or less irrelevant to the cult's later development. From at least the
early fifth century onward, Sergius was believed in the
hearts of the faithful to have been martyred at Rusafa
sometime in the distant past, in the Age of Persecution when martyrs were made. Reading the Passio
from a fifth-century perspective, the story ceases to

64/754

be a frustrating puzzle that no philological or archeological sleuth will ever be able to put right and becomes
instead a precious portrait of a martyr cult in its early
stages.

THE PASSIO OF SS. SERGIUS AND BACCHUS

With this said by way of introduction, the Passio of
SS. Sergius and Bacchus now deserves to be retold.1
Sergius was primicerius scholae gentilium, that is, the
highest-ranking junior officer in the horse guard, a regiment originally
1. The account is preserved in the original Greek, ed.
van den Gheyn, AB 14 (1895): 37395 (cited below
as Pass. gr.); a Syriac translation of the original
Greek, ed. Bedjan, AMS 3: 283322 (cited below as
Pass. syr.); a Latin translation of the original Greek,
ed. Byeus, Acta sanct. Oct. 3: 86370; and the Metaphrastic version, a recasting of the Greek original in a
more literary style, ed. Byeus, Acta sanct. Oct. 3:
87182. On this later redaction, see Hgel, Metaphrasis, 1821. The Syriac translation retains many
Latin military terms from Pass. gr. and occasionally
substitutes Syriac toponyms. The Chronicle of Seert

65/754

12.25354 (Scher) preserves in Arabic a much-distilled version of the Passio, as does the Coptic redaction of the Arab Jacobite Synaxarion for 1 Oct. (Bacchus) and 7 Oct. (Sergius). The Coptic redaction cannot be dated precisely, but its debt to the Passio is
clear in the narrative outline, despite certain innovations that are otherwise unattested. For the Arabic redactions, see Graf, Geschichte der christlichen arabischen Literatur, 1: 152. The Armenian Synaxarion of
Ter Israel for 7 Oct. also contains a much-abbreviated
version of the Passio. This thirteenth-century synaxarion is based on a tradition of translations of saints'
lives extending back in some cases as far as the fifth
century: see the introductory comments in PO
5.3.35053. Boswell, Same-Sex Unions, 37590,
provides a somewhat inaccurate English translation of
Pass. gr., in which, for example, schola gentilium is
rendered as "school of the Gentiles" and Sergius (always referred to as Serge) and Bacchus are called
"directors of our school" (and, in conjunction with this
mistranslation, see ibid., 377 n. 12, on discrimination
against homosexual teachers in schools, evoking the
literary parallel of Lillian Hellman's play The Children's
Hour [1934]).

Page 9
composed of non-Romans. He was highly favored at
the court of an emperor whom the texts call Maximianus, renowned for his persecution of Christians.2
Sergius was a friend of the emperor and, according to
the Syriac account, his kinsman. Together with Bacchus, by analogy the secundocerius, he enjoyed great
freedom of speech before the throne.3 Sergius and
Bacchus were also Christians, whose faith had not interfered with their worldly prosperity until envious colleagues used it against them. Once, when the two officers were absent from Maximianus's side, they were
denounced for their impiety toward the gods and proselytism among the ranks. To find out the truth, Maximianus ordered the pair to accompany him and his
assembled soldiers to a temple of Zeus for a sacrifice.
As he was partaking of the sacrificial meat there, the
emperor looked around but did not see Sergius and
Bacchus, who were discovered outside the temple
singing hymns against idolatry. Maximianus had the
two soldiers brought inside and ordered them to sacrifice and partake in the ritual meal. Reciting Psalm 135
against senseless images, Sergius and Bacchus refused. At once they were stripped of their belts,
cloaks, and golden torques (maniakia), their symbols

67/754

of power, and dressed instead in women's garments.

Heavy chains were clamped around their necks, and
in this fashion they were paraded through the city to
the palace, where Maximianus pressed them about
their betrayal of him and the gods of Rome. Again the
officers responded with a recitation of Christian doctrine. Furious, Maximianus ordered their entire
2. For the identity of the emperor, see below.
3. Pass. syr., p. 285. After Bacchus's execution, Sergius mourns the loss of his "brother," and Bacchus,
who appears to Sergius that night, likewise addresses
Sergius as "brother" (Pass. gr. 9, 20,and Pass. syr., p.
307). Both the Greek and Syriac emphasize that Sergius and Bacchus were joined not by a familial but a
spiritual relationship (Pass. gr. 2; Pass. syr., p. 285).
Yet the Piacenza Pilgrim (ca. 570) understood Bacchus to be Sergius's actual brother (Itinerarium, 191).
It was perhaps from the Pilgrim's account that
Devreesse acquired the erroneous view that they
were literally brothers (Patriarcat, 284). Boswell,
Same-Sex Unions, 1725, discusses a wide range of
ancient and modern meanings given to the term
"brother" and concludes that Sergius and Bacchus did

68/754

not call each other brother because they were biologically related, adding "so the appellation 'brother'
must be understood as reflective of ancient usage in
erotic subcultures or as reflecting biblical usage (particularly in Greek versions). Either way, it would have
distinctly erotic connotations" (151). Boswell does not
seem to notice that the Passio's author also attaches
the appellation to the monks who retrieve Bacchus's
body only a few lines before the new martyr addresses Sergius as his brother (Pass. gr. 19; Pass.
syr., p. 306). Severus of Antioch, who in his homily on
S. Sergius draws heavily on the Passio, also emphasizes the similarities between the two young men and
would surely have alluded to a blood relation had tradition recalled one (Sev. Ant., Hom. cath. 57, pp.
8586; see discussion below). Although Boswell acknowledges that the term "brother" could be employed
simply to describe fellow Christians (23), his commentary ignores this usage. In general, Boswell is not
much concerned with the historical background of the
cult of Sergius and Bacchus (see, e.g., his rendering
of the name of the well-known castrum Sura
as
Souros and Syrum, both inaccurate), but rather with
the saintly pair's appearance from the tenth century
onward.

Page 10
bodies to be bound in chains and had them sent to
the empire's edge, to Antiochus, dux of the province
of Augusta Euphratensis, who owed his position to
Sergius's recommendation. Maximinus sent with them
a letter in which he instructed Antiochus that if Sergius
and Bacchus repented, they were to be restored to
their former status and would enjoy even greater privileges. For refusal, they would pay with their lives.
The first night they stayed at an inn twelve miles from
the unnamed location of their fall from grace. They
traveled from city to city until they arrived "in the frontier zone near where the race of Saracens dwell."4 In
the praetorium at the fortress of Barbalissus / Beth
Blash, they were brought before the dux Antiochus;
but he had no success in undermining their resolve,
which had in the meantime been strengthened by angelic visitations. Antiochus had Sergius thrown into
prison. Bacchus was left to be tortured. After exhausting several crews of floggers, Bacchus rebutted for
one last time Antiochus's arguments against Christianity. He then hearkened to a heavenly voice that
called the "athlete and soldier" to receive his eternal
reward. The martyr's soul fled from Bacchus's tattered

70/754

body, which was left to the dogs outside the fortress

walls. Antiochus rose from the judgment seat and
went to dinner. Outside the walls, wild animals circled
around Bacchus's corpse, but contrary to nature,
guarded it until nightfall, when monks emerged from
nearby caves to retrieve the body and provide the
martyr with a fitting burial. That night Bacchus, now
received into the heavenly ranks and dressed again
as a soldier, appeared to Sergius in a dream with consolation and encouragement.
The next day, Antiochus offered Sergius another opportunity to sacrifice, which Sergius again loquaciously refused. Antiochus then had him taken to the
fortress of Sura / Shura, where, at another praetorium,
the dux conducted yet another futile inquisition, after
which Sergius was made to put on shoes spiked inside with long nails and forced to run nine miles in
front of the ducal chariot to the next fortress, Tetrapyrgium / Magdla d-mayya. Upon arrival, the dux went in
for a meal and sent Sergius to the military prison. During the night an angel healed Sergius's feet. The next
morning Sergius appeared before the tribunal unaided
by even a crutch, provoking Antiochus to accuse him
of sorcery. Still mindful of his former debt to Sergius,

71/754

Antiochus again in vain offered him a chance to repent. Then he remounted his chariot and drove toward the fortress of Rusafa with Sergius leading the
way, singing hymns. At Rusafa, Sergius made his final refusal, Antiochus thereupon condemned him to
death by the sword, and the public executioner led
him away from the tribunal. A great crowd of men, women, and children gathered to witness the execution
and were moved by
4. Pass. gr. 13:
; and cf.
Pass. syr., pp. 29899: "limiton, a desert place."

Page 11
the youth of the condemned officer. Animals abandoned their folds5 and harmlessly added their inarticulate cries to the general mourning. After a final prayer, Sergius crossed himself and knelt. His head was
cut off and a heavenly voice welcomed him to join the
army of angels, ranks of patriarchs, and choirs of
apostles and prophets.
To protect the holy site from contamination by unbelievers, a chasm appeared where the martyr's blood
had been spilled and remained there as a sign of divine intervention. Some witnesses buried Sergius's
body at the site of his martyrdom. A long time afterward, zealous Christians from Sura attempted to steal
the relics for themselves, but the martyr would not
permit himself to be removed from his victory seat and
sent a brilliant flare from his grave to alert those within
Rusafa's walls to the Surans' clandestine activity. The
soldiers garrisoned in Rusafa took this as a sign of
brewing hostilities and set out from the walled town
fully armed to drive away the offenders. The would-be
thieves then asked permission to build a funerary
monument at the potent site, and they left a few days
afterward.6 Later, when Christianity had spread

73/754

across the land, abolishing polytheism, a martyrium

was built within the walls and consecrated by fifteen
bishops. On the anniversary of Sergius's martyrdom,
7 October, the martyr's relics were transferred to the
new site. The Passio records that many miracles and
cures were worked by the saint at Rusafa, especially
at the old monument outside the walls. The martyr
displayed the power to cure the sick and possessed
and to tame beasts, especially on the anniversary of
his death, when wild animals would flock to the site
from the desert's furthest reaches, mixing peaceably
with the other pilgrims to honor the martyr established
there by God's command.

HISTORY IN THE PASSIO: THE EMPEROR

The story moves from an impressionistic, often stereotyped setting at the Zeus temple of an unnamed
city to an increasingly tangible world of familiar placenames in the Syrian steppe. By the end of the story,
fifteen bish5. Pass. syr., p. 316: "holes."
6. In the Greek it is unclear whether the Surans ask
the Rusafans for permission to build the monument or

74/754

whether the Rusafans ask the Surans to build one.

The latter interpretation is preferred by Boswell,
Same-Sex Unions 389, with n. 77, and Woods, JECS
3 (1997): 338 and 365, where, contrary to the Greek
that is his source, Woods identifies the Surans as soldiers and describes the building project as a joint effort of Rusafan and Suran soldiers. In this case reference to the Syriac is helpful. At Pass. syr., p. 321, the
Syriac translator clearly understood the former interpretation, which I have also adopted on grounds of logic as well as the Syriac evidence: "[T]hose people
[i.e., the Surans] who had attempted to steal the holy
martyr used persuasion and requested that the Rusafans allow them to be there for a few days in order to
build a sepulcher for the veneration of the bones of
the blessed one and then to go away."

Page 12
ops consecrate a specific building at Rusafawhich
has recently been identified, with the help of both the
literary texts and archeological investigation. Finally,
the author testifies that even in his day, more miracles
were worked at the original site of the martyr's bones
than at the new martyr shrine built within the walls.
We move from the mists of oral tradition to the clarifying moment of composition. The number of possible
anachronisms suggests that what we have is an accumulation of ideas and historical details, and even of literary motifs, that were felt to be compatible with the
basic story of the soldier's martyrdom. Whether this
assimilation occurred gradually over a long period of
time, or in the mind of a single author is not certain,
but the tendency of other more or less contemporary
saints' stories to take shape over time makes the
former more likely also in the case of our Passio.7
Because the identity of Sergius's persecutor has been
for some time a matter of scholarly comment, a brief
look at the historical difficulties posed by the Passio
cannot be avoided. That way we can hope either finally to lay to rest the temptation to learn anything
about the historical Sergius before the fifteen bishops

76/754

appear at Rusafa or, if it proves impossible to deny

the story any historical plausibility, at least arrive at
the most probable historical setting for the events described. First of all there is the question, who was the
emperor called Maximianus? The Passio's setting is
one of general repression of Christianity, so our emperor must be one known to have been involved in
persecutions. Was he, then, Galerius Maximianus
(caesar 293305; augustus 30511) or Maximinus
(Maximin) Daia (caesar 30510; augustus 31013)?
Or was he, as has recently been suggested, the
apostate Julian? All three were famous persecutors,
and Galerius and Maximinus Daia, at least, were easily and commonly confused. One possible clue might
be the title
used to describe Maximianus in the
Greek text.
would most naturally denote an augustus, making it unlikely though not impossible that a
caesar (such as Galerius on campaign in the East
between 293 and 299, or Maximinus Daia at Antioch
in 3056) should be considered.8 On the other hand,
it would be uncharacteristic of this particular text to
provide a precise portrait of the imperial college in the
late third and early fourth centuries.9

77/754

7. See, e.g., Walter, REB 53 (1995): 295326, on the

cult of S. George.
8. It was not unheard of for a caesar to appear as sebastos or augustus already on third-century coins: see
Ziegler, Tyche 9 (1994): 189. On nonofficial use of
basileus through the third century, see Wifstrand, in
Dragma, 52939; on the official use of imperator for
both augustus and caesar, see Kolb, Diocletian, 22 n.
54 and 46 n. 122; also Millar, Emperor, 61315, for
the use of basileus in both official documents and
literature.
9. This means that little weight can be placed on the
term basileus, either to rule out entirely the possibility
that Maximianus was in fact a caesar or to suggest,
as does Woods, JECS 5 (1997): 354, that the persecuting emperor was a sole ruler, since no partner is
mentioned.

Page 13
The choice is not made easier by the texts' silence
concerning the site of the accusation against Sergius
and Bacchus. The Passio's omission of the city's
name is suggestive in itselfprobably it was too obvious to merit mention, or considered irrelevant. All we
are told is that it had a palatium, a marketplace, and a
temple of Zeus, and that the officers traveled thence
city by city, without any indication of time, until they
reached the province of Augusta Euphratensis and
the fortress of Barbalissus on the Euphrates.10 The
most natural choice is Antioch, whence one traveled
some 170 kilometers along the northern Syro-Mesopotamian route via Beroea and Hierapolis to Barbalissus on the Euphrates.11
Since the city mentioned in the Passio had a palatium,
Nicomedia should be briefly noted as an alternative,
although the journey would obviously have been
much longer. If we consider Nicomedia as the first
scene of action, then the augustus Galerius becomes
a candidate for "Maximianus." According to the existing record of his movements, the caesar Galerius left
Syria in 299, following his eastern campaigns, and is
not known to have returned.12 After 299, Galerius

79/754

fought and ruled primarily in the Balkans, but he could

be found occasionally at Nicomedia, where on 24
February 303, Diocletian's edict against Christians
was promulgated. The Passio's action takes place in
an atmosphere of persecution, with anti-Christian proclamations posted in marketplaces, resulting in the
torture and execution of non-conformists.13 However,
Galerius's absence from that capital for most of 3035
leaves only a few brief occasions when the action of
the Passio could have taken place during his presence at Nicomedia, if we are to identify the text's Maximianus with Galerius as augustus.14
The argument in favor of Julian is inspired by discomfort with certain details in the Passio that have been
thought irreconcilable with a date under the tetrarchy,
and by the existence of literary parallels to the Passio
in historical and hagiographical sources that can be
matched with events during
10. Pass. gr. 13; Pass. syr., p. 298. To my knowledge,
there is no evidence to support the assertion that Sergius was martyred "on the Euphrates front where he
was stationed against the Persians" (Shahd, BASIC,
1: 953).

80/754

11. Mouterde and Poidebard, Limes, 27.

12. Barnes, New Empire, 6164.
13. Pass. gr. 1, 3; Pass. syr., pp. 284, 286; cf. Eusebius, HE 9.37.
14. A. H. M. Jones, who accepts the Passio as "rhetorical" but "accurate in factual detail," takes "Maximianus" to mean "Galerius as caesar before his proclamation as augustus in 305; for the incident took
place in Syria" (Later Roman Empire, 54); cf. the date
of 3035, published a decade earlier by Jones as the
date of their martyrdom and reprinted in Roman
Economy, 268 n. 32. However, as stated above, there
is no clear evidence that Galerius ever returned to
Syria after 299 (Barnes, New Empire, 6466). It
should be noted that the PLRE entry for Antiochus
dux of Augusta Euphratensis (PLRE, 1: 71, Antiochus
2) relies on Jones's dating of the martyrdom to 3035
(Roman Economy, 268 n. 32) for its dating of
Antiochus.

Page 14
Julian's brief rule.15 It is possible to reconstruct a context in which the Sergius and Bacchus story might
make sense during his reign. Zosimus, for example,
records that Julian once paraded soldiers dressed in
women's clothing as a punishment for desertion16
and, indeed, Sergius and Bacchus are accused of
desertion by the emperor, who inflicts on them the
same punishment. Julian is also known to have sacrificed regularly at the temple of Zeus in Antioch.17 But
none of these parallels is sufficiently distinctive to
render an identification with Julian unavoidable.
Besides being a rather obvious form of humiliation,
dressing delinquent soldiers in women's clothing is a
frequently noted punishment for deserters in later
Byzantium, as well as in the Sasanian and
empires.18 It also turns up long before Julian, in the fifth
century B.C., in the law code of the Thurii.19 In the
Republican period, one characteristic element in a
soldier's attire was the military belt, the cingulum militiae. One form of degradation was to remove the cingulum and compel the delinquent soldier to stand with
his tunic ungirt in public.20

82/754

Certainly the identification of the text's Maximianus

with Julian would relieve the Passio of some possible
anachronisms regarding the frontier situation, which is
discussed below. And it is not utterly impossible (although quite unprovable) that the author had read
some lost accounts of Christian soldiers persecuted
by Julian, as has been suggested by David Woods.21
But even if the author had read and borrowed from
such accounts, that does not imply that the martyrdom
of Sergius that he was narrating also took place under
Julian.22 It remains difficult to explain why it is that
Sergius's publicist
15. I would like to thank David Woods for sending me
his article "The Emperor Julian and the Passion of
Sergius and Bacchus," JECS 5 (1997): 33567.
Woods develops in great detail, treated here only
briefly, a suggestion made by Franchi de' Cavalieri,
Note agiografiche 9: 169200, esp. 19499.
16. Zos., Hist. nov. 3.3.45.
17. Jul., Misop. 346 BC.
18. For Byzantine incidents, see Cinnamus, Epitome
1.5.1113; Pachymeres, De Mich. et Andron.

83/754

Paleolog. 3.25; Nicetas Choniates, Hist. 6, p. 196;

Scylitzes, Synop. hist. p. 429.4. Theoph., Chron., A.
M. 6080, p. 263, records that, adding insult to injury,
Hormisdas had women's garments delivered to his
general Bahram Chobin after his defeat by the Romans in 587/8 (589 in other sources). See Scylitzes,
Synop. hist., pp. 44445.9 for the
practice.
19. For the Thurii, see Diodorus Siculus 12.16.1. For
more general discussion, see Polites in
4
(191314): 628, with n. 5; Koukoules,
, 3: 198,
with n. 9; Guilland in Byzantinoslavica 27 (1966):
3045; and Alan Cameron, Circus Factions, 172, with
n.4.
20. For numerous examples, see Mller, NJKA 17
(1906): 56068.
21. Woods, JECS 5 (1997): 344, 357, 361, 36364,
36667, posits that such accounts existed and were
subsequently lost.
22. Woods assumes that the Passio is a complete fiction. This is a position with a certain pedigree. Without
ever arguing his point directly, the influential Bollandist Hippolyte Delehaye several times dismissed the

84/754

Passio of Sergius and Bacchus as devoid of historical

value, a fantasy: see, e.g., AB 23 (1904): 478;
Mlanges d'hagiographie, 238; Origines, 210; also
see

Page 15
chose to call the tyrannical emperor Maximianus
rather than using thesurely sufficiently infamousname of Julian, if indeed the Passio depended
so heavily on stories of Julianic persecutions. Woods
does not raise this issue, although in a note he draws
attention to Christian frustration at Julian's policy of
persecution without execution, hence starving the
Christians of martyrs. Precisely because of this
dearth, and of Julian's lasting reputation as
Christianity's enemy par excellence, it would be peculiar for any Julianic heroes to have been transposed to
the time of a vague "Maximianus."
Maximinus Daia, to consider the third possible emperor, arrived at Antioch in the late spring of 305, after his
proclamation as caesar.23 In his capacity as caesar in
the East, but especially as augustus from 310 to 313,
Maximinus pursued a well-publicized anti-Christian
campaign with great energy. He directly involved governors and duces in the harassment of Christians, a
situation that again calls to mind the active role given
to the dux Antiochus in the Passio.24 A scurrilous
confession by two allegedly ex-Christian prostitutes at
Damascus was posted in marketplaces around the

86/754

eastern Empire, rescripts encouraged communities to

drive Christians from their midst, and the ancient
priesthoods and temples were revamped in the spirit
of the buoyant polytheism that Maximinus espoused.25 It would seem most natural to place the
martyrdom of Sergius and Bacchus in the later phase
of Maximinus's persecution, when he was augustus
between spring 310 and 313. During late 311 and
early 312, he was resident at Nicomedia, which cannot be absolutely excluded as the scene of Sergius's
and Bacchus's denouncement. But Maximinus's antiChristian campaign intensified in

Delehaye et al., eds., Martyrologium romanum, 439.

On Delehaye's influential contribution to the study of
hagiography, see Peeters, Oeuvre, 10349, and
Knowles, Historical Enterprises, esp. 2527. This view
has been embraced by Honigmann, EI1, 3: 1265, who
nonetheless confirms the accuracy of the Passio's topographical details, and by Barnes, New Empire, 186.
But Barnes has demoted Antiochus to fictional status
following Delehaye's unargued claim that the Passio
is fictitious. In an attempt to bolster Delehaye's summary judgment, Barnes suggests that "the name

87/754

'Antiochus' is an obvious allusion to the Seleucid king

of the second century B.C. against whom the Maccabees rebelled" (New Empire, 186). But PLRE, 1:
7173, has sixteen entries for Antiochus, not an unusual name in the late antique East. The excavators
of Rusafa have continued this uneasy relationship
with the Passio, which is both a hagiographical text
and a mine of historical and topographical information.
Kollwitz, the first director of the excavations, claimed
on the one hand that "im brigen sind diese Akten
spt und ohne Quellenwert" (Neue Deutsche Ausgrabungen, 46), while proceeding, on the other, to exploit the Passio's clues to Rusafa's architectural development. Ulbert, Kollwitz's successor, continued to
use the Passio as one of several guides to understanding the early phases of architectural development at Rusafa, e.g., in Actes, 445.
23. Barnes, New Empire, 6566; Downey, History of
Antioch 31823, 64445.
24. Eusebius, HE 8.14.910 and Mart. Pal. 9.2; cf.
Lact., De mort. persec. 36.4.
25. For prostitutes: Euseb., HE 9.5.12; rescripts:
Mitchell, JRS 78 (1988): 10524, esp. 11112;

88/754

polytheist reform: Nicholson, JEH 45 (1994): 310,

esp. 34.

Page 16
312. During that year the dioceses of Asiana, Pontica,
and Oriens are recorded to have sent embassies to
Maximinus concerning the Christian blight in their
communities, and the emperor responded with a rescript giving carte blanche to the polytheists on 6 April
312.26 Among the cities requesting the emperor's assistance against the Christians was Antioch, led by its
dynamic curator Theotecnus.27 It was a suitable climate for the accusation against Sergius and Bacchus.28 Maximinus was in Antioch in the summer of
312, before his Armenian campaign that autumn,
which would allow for the tradition that Sergius and
Bacchus were martyred on the frontier in early October, Sergius on the seventh.29 But we cannot assume
that the festal date necessarily represents the precise
date of Sergius's martyrdom.
Further support for postulating the date of 312 under
Maximinus Daia is found in the martyrdom's setting.
The frontier fortresses that provide the backdrop for
the officers' execution are said in the Passio to belong
to the province of Augusta Euphratensis.30 This
province is attested in the Laterculus Veronensis,
whose record of the eastern provinces can be fixed

90/754

between 314/15 and 324.31 This date, supported by

epigraphical and papyrological evidence, dissolves
Honigmann's skepticism about the Passio's chronological trustworthiness, which was founded on his belief
that Augusta Euphratensis was established in 341.32
The 314/1524 date for the Laterculus Veronensis
would easily accommodate a martyrdom in Augusta
Euphratensis in 312.33
26. Euseb., HE 9.7.314, translates the original Latin
rescript that was posted at Tyre. A fragmentary copy
was found at Arycanda in Lycia, which includes also
the petition to the emperor by the province of Lycia
and Pamphylia (CIL 3.12132). Another copy was recently discovered at Colbasa, a Pisidian community
also located in the province of Lycia and Pamphylia;
see Mitchell, JRS 78 (1988): 10524, for publication
and commentary. Eusebius, HE 9.2, and Lactantius,
De mort. pers. 36.3, accuse Maximinus of inspiring
the petitions himself.
27. Euseb., HE 9.23; PLRE, 1: 908, Theotecnus 2;
see also Mitchell, JRS 78 (1988): 114.
28. The claim made by Woods, JECS 5 (1997):
35254, that between 299312, two Christians could

91/754

not possibly have risen to the highest echelons of the

imperial bodyguard without detection, paints an
unjustifiably black-and-white picture of historical realities, which were influenced by the unpredictability and
flexibility of both the Christians' and the authorities' attitudes toward cohabitation.
29. The Chronicle of Seert 25, p. 292, records that
"Maximin" killed Sergius and Bacchus, although earlier in the Chronicle the passio is recounted with "Maximianus" as the antagonist. There is a long tradition in
favor of Maximinus Daia, which includes the early
commentators on and editors of the texts: Tillemont,
Mmoires, 5: 492; Byeus, Acta sanct. Oct. 3: 838,
844; and van den Gheyn, AB 14 (1895): 375 n. 1; see
also Chabot in his edition of Mich. Syr., Chron., 1: 188
n. 3; Honigmann in RE, 4A: 1706; Poidebard and
Mouterde, AB 67 (1949): 110; Halkin, BHG, 2: 238;
Hoffmann, Sptrmische Bewegungsheer, 282 n. 788
(in Antioch).
30. Pass. gr. 2, 13.
31. Barnes, New Empire, 205.
32. Honigmann in RE, 4A: 1698, 1706.

92/754

33. Of course, the date of the Laterculus Veronensis

gives merely a terminus ante quem for the establishment of the provinces it records. A. H. M. Jones, who
dates the Laterculus Veronen-

Page 17
As is often the case with toponyms, the Syriac Passio
does not simply translate the Greek, but incorporates
local usage. In the Syriac Passio, Antiochus is called
the "dux of the Euphrates" and the province is named
"Phoenicia, in the authority of the western side of the
river."34 These variations resist easy explanation. It is
quite likely that the variation derives from the name
known to the translator. But the Syriac could also reflect the third-century provincial divisions when the
province of Phoenicia stretched at least as far northeast as Palmyra. Temporary reversions to former provincial names, either as part of actual administrative
restructuring or as literary slips, are not unknown. For
example, even though the province of Augusta Euphratensis existed at least by the second decade of
the fourth century, the Nicene signatures of 325 recognize the old third-century divisions of Coele Syria,
Phoenicia, Arabia, and Palestine.35 However we interpret the Syriac text's divergence, the name Augusta
Euphratensis in the original Greek corresponds to the
provincial nomenclature in 312.

94/754

THE HAGIOGRAPHIC MILIEU

Simply to brush aside the historical setting presented
in the Passio risks forfeiting traces of the background
that might aid our understanding of the cult's development. Nonetheless, the fact remains that the stereotypes, vagueness and even confusion of the
narrative's early sections exist in large part thanks to
the fact that the Passio of SS. Sergius and Bacchus
belongs to the fermenting hagiographical environment
of the fifth and sixth centuries, when accounts of martyrs' trials mutually inspired and reinforced each other.36 The author of the martyrion of S. Athanasius of
Clysma, for example, used the story of SS. Sergius
and Bacchus as a model for his martyr's ordeal, and
even gave Sergius and Bacchus parts to play.37 In
the martyrion, Athanasius is a high-ranking officer under Maximianus suspected of Christian sympathies.
To test Athanasius, Maximianus sends him to
Egyptnot as a humiliated, exiled officer, but as a
procurator. Athanasius bids farewell to his

sis to 312, sees no difficulty in dating the martyrdom

95/754

in Augusta Euphratensis to 3035 (Roman Economy,

265, 268 n. 32).
34. Pass. syr., p. 285, and cf. p. 298. See too Sev.
Ant., Hom. cath. 57, which relies heavily on the Greek
Passio. There Severus describes how Sergius and
Bacchus were led "to that region in Mesopotamia
which the local people call Euphratesia" (p. 90). See
below for a discussion of this homily.
35. Barnes, New Empire, 224, with n. 58; Jones, Roman Economy, 267.
36. On hagiographical topoi in the Passio, see Byeus,
Acta sanct. Oct., 3: 838, and Delehaye, Lgendes hagiographiques, 2728.
37.
Kerameus in

ed. Papadopoulos5: 36067.

Page 18
friends Sergius and Bacchus, members of the schola
gentilium, and encourages them in their future
struggles, which he foresees will take place "in the
province of Augusta Euphratesia." The debt of the
martyrion of Athanasius is straightforward.
A somewhat less clear connection can be discerned
in the Vita of Victor and Corona, an account with a
very complicated history and multiple recensions in
Latin, Coptic, Arabic, and Ethiopic. The original Greek
Vita is not dated, but can be placed in the milieu of
fifth- to sixth-century hagiographical writing.38 Even in
the original, the story appears to be confused: Victor
is an Italian soldier who refuses to sacrifice to the
gods. A dux named Sebastianus is assigned to persuade him, Victor is adamant in his refusal and is
eventually decapitated. At the end, it is stated that the
martyrdom took place in Damascus, a city in Italy,
during the reign of the emperor Antoninus. In a Coptic
recension, the story becomes more interesting and includes a scene in which the Christian soldier Victor
throws his golden chain in the face of the persecuting
emperor Diocletian. This gesture also appears in the
Vita of Gordius, a soldier in Antioch under

97/754

Maximianus who, like Sergius and Bacchus, is

dressed in female clothing as punishment.39 A millstone is then tied round Gordius's neck and he is
thrown into the Orontes River. The exchange of a military torque for a martyr's crown is a potent image and
appears as early as the fourth century, in Prudentius's
Peristephanon.40 These literary echoes are vexing,
since it is not usually possible to date the recensions
with much precision, and thereby to retrace the direction of influence, if any, from one account to another.
The slow absorption of literary commonplaces must
be viewed side by side with more direct literary borrowing.41 The emperor "Maximianus," the penal role
of the dux, and the removal of military insignia have
all been absorbed into the repertoire of military martyr
accounts, but may well have sprung originally from
actual events before they sparked the hagiographers'
imaginations.
Together with the influence of fifth- and sixth-century
hagiographical trends on the Passio of SS. Sergius
and Bacchus, we must also consider the problem of
seeming anachronism within the Passio. One possible
anach-

98/754

38. Delehaye, AB 40 (1922): 11718, 151. For comparison of the Latin recension and a Coptic eulogy on
Victor, see Galtier, BIFAO 4 (1905): 105221. The
brief account of the Passio of SS. Sergius and Bacchus in the Coptic recension of the Arab Jacobite
Synaxarion, 1023, 11314, belongs to this atmosphere, in which many strands of hagiography
crossed.
39. Halkin, AB 79 (1961): 515.
40. Prud., Perist. 1.65.
41. On the literary genre of hagiography, the "Epic
Passions" and the use of commonplaces, see Delehaye, Passions des martyrs, 236315. See also Garsoan in East of Byzantium, ed. id., Mathews, and
Thomson, 15152, with n. 11, who draws attention to
parallels between the trials of SS. Sergius and Bacchus and of S. Gregory of Armenia, as depicted in
their passiones.

Page 19
ronism in the Passio more serious than the name
Augusta Euphratensis is the alleged existence in the
early fourth century of the schola gentilium. The
Passio's representation of Sergius as primicerius and
Bacchus, by analogy, as the secundocerius of the
schola gentilium would be the earliest attestation of
that body, the foreign regiment of the imperial horse
guard, otherwise only known to have existed from the
reign of Constantine onward.42 Another possible anachronism is the presence of monks living in caves
near the middle Euphrates in the early fourth century.43 Monasticism had begun to take root in upper
Mesopotamia, in the Tur
, in the fourth century,
and certainly by the early fifth century a monastic
community had gathered round Alexander Akoimetes
on the banks of the middle Euphrates.44 But in its
suggestion that monks had taken up residence in
caves near Barbalissus by the early fourth century,
the Passio stands alone.45 As in the case of the early
date suggested by the Passio for the schola gentilium,
we cannot absolutely rule out the Passio's contribution
to our knowledge of early fourth-

100/754

42. Pass. gr. 1, 2; Pass. syr. p. 285. See Speidel, Riding, 7576, on the Constantinian reorganization of the
horse guard (scholae), and Frank, Scholae, 5455, on
the definition of gentiles, although neither he nor
Speidel discusses Jones's argument in Later Roman
Empire, 613, where with some caution he cites the
Passio as evidence of the existence of the schola
gentilium already under Diocletian. See also Byeus,
Acta sanct. Oct., 3: 839. Sergius and Bacchus, the
earliest attested members by Jones's calculation, bore
Roman names either because they were Roman officers set over foreign troops or because, as in the
case of many non-Romans serving the East, they had
adopted Roman names. Already in Ammianus any
distinction had blurred, since most regiments, the
scholae in particular, were manned by Germans.
Jones, Later Roman Empire, 61314n. 12, cites examples of barbarian tribunes of the scholae. On the
frequency of German officers in the scholae, see Hoffmann, Sptrmische Bewegungsheer, 1: 28185,
299300; and see Liebeschuetz, Barbarians and
Bishops, 8, on German officers' preference for Latin
names, and 23, on the upward mobility of German
soldiers in the scholae. Since Sergius is a well-attested Roman name and goes back to the Republican

101/754

period, Peeters's suggestion, Trfonds, 68, that the

name Sergius is a Hellenized version of the Syriac
shraga, meaning lamp or torch, is unnecessary.
However, it does seem that to the Semitic ear, the resemblance between Sarjis and shraga or siraj (the respective Syriac and Arabic words for "lamp") lent itself
to puns: e.g., Jacob of Sarug, Mimra on the Victorious
Sergius and Bacchus, v. 116 (p. 658 [Bedjan]), and alBuhturi, Diwan, 108. Unfortunately, the Passio
provides no background information on the martyrs
that might suggest their familiarity with the region.
43. Pass. gr. 19; Pass. syr., pp. 306.
44. Vita et institutum piisimi patris nostri Alexandri
26.31, pp. 67782. On the historical reliability of the
Euphrates episode in the Vita Alex., see Gatier, in
Frontires terrestres, 43557; also Brinkmann, BJ 99
(1896): 25254; Peeters, Trfonds, 150.
45. Palmer, Monk and Mason, 2122, 24, 3031, discusses a possible mid fourth-century foundation date
for the monastery of Mor
, as well as fourth-century
activities of monks in the frontier zone between the
Roman and Iranian empires. Excavation of the monastic settlement at Tal , near modern Raqqa, has

102/754

advanced our knowledge of Euphratine monasticism

in late antiquity. Graves discovered in the area of the
large courtyard south of the church have been dated
in the period of the fourth and fifth centuries (Strommenger, MDOG 123 [1991]: 1011).

Page 20
century asceticism, since it foreshadows conditions
known to have existed in the mid fourth century; but it
is thin evidence.
Anachronism reflects on the text in which it appears
but tells us little or nothing about the events the text
purports to describe. It is most likely that the schola
gentilium did not exist under the tetrarchy; but the
author's intrusion of this later entity into Sergius's martyrdom does not allow us to exclude the possibility
that Sergius was executed by a tetrarch. It merely
convicts the author (or some earlier participant in the
development of the oral tradition) of projecting his
knowledge of the scholae, and perhaps likewise of
Euphratine monks, into Sergius's story in order to lend
it a certain specificity.46
Particularly important from the point of view of the
Sergius cult's development in late antique Syria and
Mesopotamia is the attention paid in the Passio to topography, as Sergius and Bacchus travel from
castrum to castrum along a military road to their martyrdoms. Topographical and archeological details
seem correct for both fourth- and fifth-century Syria.

104/754

The road later called the Strata Diocletiana had been

marked with milestones from the time of Vespasian.47
Generally speaking, the military castra that serve as
the backdrop for the Passio's action were foci of
Diocletian's eastern fortification activity, and some existed, it now seems, even earlier.48 Archeological investigation confirms a Diocletianic date for the forts at
Sura, Rusafa, and also Khulla, another fort located
some 20 kilometers south of Rusafa, but not mentioned in the Passio. Excavation at Tetrapyrgium /
Qasr al-Sayla, Sergius's penultimate stop in the
Passio, has uncovered a Neronian or Flavian military
post followed by no material evidence of a fort or other settlement before ca. 320.49 Here again, it must be
understood that the
46. Hoffmann, Sptrmische Bewegungsheer,
28283, surveys the anachronisms related to institutions and interprets them as later additions to the
Passio.
47. For the Vespasianic milestone from Arak attesting
to road construction in A.D. 75 between Sura and
Palmyra via Rusafa, Tayyiba and Arak, see Seyrig,
Syria 13 (1932): 27072; AE = RA 1 (1933): 42122,

105/754

no. 205, with the discussion by Bowersock, JRS 63

(1973): 13340. For an overview of the Roman presence in the neighborhood of Rusafa from the first to
the sixth centuries A.D., see Isaac in Frontires, 110
and n. 12. For discussion of the Strata Diocletiana
from the second to the fourth century, see Bauzou in
Archologie, 21153. See also Konrad, DaM 6
(1992): 346, with nn. 166, 167; Mouterde and Poidebard, Limes, 13132; and Honigmann in RE, 2A:
1684. Rusafa's place in the frontier zone is discussed
further in chapter 3.
48. Konrad, DaM 9 (1996): 16376. For Diocletianic
fortification and Rusafa's place in it, see esp. Konrad,
DaM 6 (1992): 34750; Bauzou in Archologie,
21113; Ulbert in Actes, 1: 445, and in Archologie,
28396, esp. 28891; and Poidebard, Trace de
Rome, 7374. Joh. Mal., Chron. 12, p. 308, describes
Diocletian's erection of milestones marking the military
routes along the eastern frontier.
49. I am grateful to the excavator, Michaela Konrad,
for correspondence regarding this site. For the publication of the early Roman evidence, see Konrad, DaM
9 (1996): 16376, and

Page 21
author's inclusion of Tetrapyrgium in Sergius's martyrdom may well have been inspired by his personal
knowledge of the region in the fifth century. However,
considering the range of imprecision inherent in dating
archeological evidence, a date of ca. 320 for the earliest late antique evidence at Tetrapyrgium does not allow us to exclude the possibility that Sergius spent a
night in Tetrapyrgium in 312.
Evidence exists also for the presence at this time of
troops in the forts mentioned in the Passio. Barbalissus had served as a garrison for the Equites Dalmatae Illyriciani since the time of Diocletian.50 Rusafa
appears as a station on this route on Ptolemy's map
and the Tabula Peutingeriana, which is based in its
eastern part on a second-century archetype, as well
as in the late fourth-century Notitia dignitatum, which
records at Rusafa the Equites sagitarii indigenae, locally recruited cavalry.51 The Passio recounts that
many years after Sergius's martyrdom, the saint's relics were defended by fully armed soldiers who
emerged from within Rusafa's walls. The Syriac adds
that the Surans wished, by stealing the relics from
Rusafa, to venerate the bones "in a pacified place."52

107/754

If the martyrdom took place in 312, "many years later"

would push the date of the attempted grave robbery
well into the fourth century, when the martyr's fame
had grown sufficiently for his relics to merit theft, and
into the period of military organization recorded by the
late fourth-century Notitia dignitatum for the East. Not
long afterward, probably late in the second decade of
the fifth century, we are finally on firm ground with the
translation of the bones within the walls to the new
church built by the bishop of Hierapolis.
From this weighing up of the Passio's details against
the historical and topographical situation known from
other sources, the following picture emerges. With the
establishment of the martyr's bones in a new church
inside Rusafa's walls, someone was called upon, or
inspired, to write down the martyr's story. The author
would have collected all the evidence he could

for the late antique material, id., Resafa, vol. 5: Der

sptrmische Limes in Syrien. Also for Tetrapyrgium,
see Konrad, DaM 6 (1992): 34750; Poidebard and
Mouterde, AB 67 (1949): 10910; and Musil,
Palmyrena, 26364. Tetrapyrgium, unlike Rusafa,

108/754

was not heavily settled in late antiquity. Excavation

within the walls at Tetrapyrgium has uncovered a
sixth-century monastery similar in plan to that discovered at Tal
on the Euphrates near Raqqa (AA
1993: 730, and 1991: 672). Ceramics at the fortified
site of Umm al-Tlala near al-Kawm reveal settlement
phases similar to those at Tetrapyrgium, i.e., late first
to early third century A.D., followed by fifth- to
seventh-century occupation (Majcherek and Taha,
Cahiers de l'Euphrat 7 [1993] 107).
50. Not. dig., Oriens 33.25; Ulbert in Archologie, 284,
286.
51. Ptol., Geog. 5.15.24; cf. Ravenna Geographer,
Cosmogr. 2.15, p. 89.1; Not. dig., Oriens 33.27. Kollwitz suggests without good reason that the equites at
Rusafa were a corps of camel riders (Neue Deutsche
Ausgrabungen, 46).
52. Pass. syr. p. 320.

Page 22
for the martyr's life and death and life after death, and
then he would have been left to flesh out the story
himself where it seemed fit. He would have understood the price of time's forgetfulness bemoaned by
Prudentius in the Peristephanon: "Alas for what is forgotten and lost to knowledge in the silence of olden
time! We are denied the facts about these matters,
the very tradition is destroyed."53 Taking a minimalist
view of the author's raw material, we can accept that
he learned that a soldier (perhaps named Sergius)
had been martyred outside Rusafa's walls in the distant past and a cult had grown up focused on the site
of the martyrdom. He could certainly have been told
that the martyrdom happened during the Great Persecution under the tetrarchs, among whom Maximianus was a common name. Sergius and Bacchus
may have suffered along the Strata Diocletiana as the
Passio relates, with place-names added to bring it up
to date for contemporary listeners and readers; or,
perhaps the presence of another martyr's bones at
Barbalissus served as a catalyst for the idea of linking
up the two lives along the military highway.

110/754

What all students of the Passio must accept is that the

author's hand is clearly visible in the story he produced. Therefore, for the purpose of studying the
cult's spread, the date of the Passio's composition is
crucial. The earliest datable evidence for the use of
the written Passio is Severus of Antioch's homily on S.
Sergius. Severus delivered this sermon at the feast of
Sergius at Chalcis / Qinnasrin, celebrated in this case,
it seems, on 1 October 514.54
53. Prud., Perist. 1.7374 (tr. Thomson).
54. Sev. Ant., Hom. cath. 57, pp. 8394. On Severus's
background and the place of Hom. cath. 57 in his extant work, see Brire, Introduction, 5556. Menologia
reveal a variety of dates for the commemoration of
SS. Sergius and Bacchus, of which 1 and 7 October
are the most common. Neither S. Sergius nor S. Bacchus appears in the Calendar of Edessa of 411, dated
between 362 and 411, a poor Syriac translation of a
Greek calendar. The sources for the Greek are not
oriental, a fact that helps explain the absence of SS.
Sergius and Bacchus. The Passio unambiguously
names 7 October as the feast of S. Sergius at Rusafa,
and there is nothing to suggest that the date is a later

111/754

insertion. A late seventh-century non-Chalcedonian

menologion in the Syrian Calendar, 3135, records S.
Sergius and S. Bacchus on 1 October; a ninth-century
non-Chalcedonian menologion in the Syrian Calendar,
3548, is unique in placing the commemoration of
both saints on 3 March, 1 September, 1 October
(adding that on this day Bacchus was killed), and 7
October (adding that on this day Sergius was killed); a
tenthtwelfth-century non-Chalcedonian menologion
in the Syrian Calendar, 4853, has S. Sergius and S.
Bacchus twice, on 9 December and 1 September; a
twelfth-century non-Chalcedonian menologion in the
Syrian Calendar, 5356, records the illustrious SS.
Sergius and Bacchus on 31 October. The only date
consecrated to SS. Sergius and Bacchus in fourteen
other non-Chalcedonian menologia dating from the
twelfth to the fifteenth centuries in the Syrian Calendar, 56 and 6387, is 7 October. The same date is
also recorded for SS. Sergius and Bacchus in the
Chalcedonian Synaxarium ecclesiae Constantinopolitanae, which was formed in the tenth century. The
Coptic redaction of the Arab Jacobite Synaxarion
commemorates the death of S. Bacchus on 1 October
and the death of S. Sergius on 7 October. Some of
the manuscripts

Page 23
As patriarch, Severus spent most of his time in Syria
at Antioch, although we know of a few occasions
when he preached at churches outside the great city.
In another impromptu sermon, delivered at Chalcis on
an unspecified date, Severus addresses his audience
as an "assembly of the faithful and God-loving" and
"the living church of God," which confesses the true
Orthodox faith.55 He also describes Chalcis as the
true daughter of Antioch. Clearly, his visit must be
seen in the context of his efforts to rally support for his
anti-Chalcedonian views in Antioch's hinterland. The
sermon's explanatory introduction added by the compiler, James of Edessa, notes that this particular
homily was interrupted by a raucous crowd demanding that Severus arbitrate in some unnamed municipal
dispute.
Chalcis was also an important camp for the pastoral
Arabs of the region, who would have gathered for the
feast of the famous martyr.56 We can imagine how,
looking around at the tents of the nomad pilgrims,
Severus chose to open his homily with a comparison
of the generosity of the people at Chalcis with the
hospitality of Abraham, who emerged from his tent to

113/754

serve the angels at Mamre.57 After this brief, personalized introduction, Severus retells the story of the two
martyrs, and of Sergius's miracles at Rusafa.58 He
concludes his sermon with a cadenza inspired by
Sergius's pierced feet: We must be watchful against
Satan, the snake who with sleepless eye fixed on our
heels lies waiting to push us into the pit of sin by our
love of pleasuresuch as stuffing one's belly, which
Severus warns against early on. On the whole, Severus recycles rather predictable, spoilsport festal
themes.59
It is apparent from the opening of the sermon that
Severus was acquainted with the cult of S. Sergius
from the written Passio, and had no personal experience of the cult at Rusafa. With great faithfulness
Severus follows the sequence of events in the Passio,
but he tells a punchier story and dwells on certain incidents, such as the dressing up of the two soldiers in
women's garments, in order to hold the attention of a
lively audience. Both the Passio and Severus's homily
exploit the symbolic value of clothing and adornment
to stress the soldiers' transgression of culturally fixed
bound-

114/754

of this synaxarion also commemorate the consecration of the church of S. Sergius at Rusafa on 15
November, an otherwise unattested feast.
55. Sev. Ant., Hom. cath. 56, p. 72.
56. For Arab encampments in the area of Chalcis /
Qinnasrin and Beroea / Aleppo, see Shahd, BAFOC,
400402.
57. Sev. Ant., Hom. cath. 57, pp. 8385.
58. Ibid., pp. 8593.
59. Ibid., p. 93. The improvised tone of this sermon is
echoed in the later Hom. cath. 110, delivered in May
517 at the town of Aegae and known from James of
Edessa's annotations to have been ex tempore (see
Brire, Introduction, 61).

Page 24
aries, in this case their betrayal of a soldier's duty to
the emperor and the gods who bestowed authority on
him.60
Severus's homily is punctuated with quotations, often
in abbreviated form, from some of the same scriptural
passages cited in the Passio, along with near-verbatim rendering of speeches and events the Passio
relates, such as Maximianus's charge that Christ had
been the illegitimate son of a carpenter crucified by
the Jews for stirring up trouble, followed by an antiOlympian response by Sergius and Bacchus.61
Severus also calls the responsible emperor "Maximianus." Finally, since the feast at Chalcis was in honor
of S. Sergius alone, Severus had to reconcile this singular dedication with his own text-bound knowledge of
the joint martyrdom of Sergius and Bacchus. He does
this by underlining, as he begins to tell the story, that
Sergius was martyred with Bacchus and that those
whom the martyr's crown has joined together should
not be separated in the recollection of their shared
martyrdom.62 Severus remarks on their similar appearance and youthfulnessdetails absent from the
Passio, but inserted here by the Pisidian-born

116/754

patriarch in an attempt to flesh out his purely bookish

knowledge of the martyrdom for a crowd that in all
likelihood knew much better than him.
Severus's homily on S. Sergius is commonly cited as
a sign of the cult's spread. But this rather unremarkable sermon offers more than that. It provides a terminus ante quem for the Passio's composition.
Severus's dependence on it means that the Greek
Passio was in circulation at least by 514. The author
of the Passio does not for a moment pretend to have
been an
60. The public punishment, designed to return Sergius
and Bacchus to their proper place in society, deliberately inverted the soldiers' normal status through the
removal of military symbols worn by men who enjoyed
great freedom before the emperor and the substitution
of both women's clothing and chains, signs of dependence. Mithraism exhibits a similar attempt to confirm
desired social order by its temporary, controlled inversion, found particularly in the grade called Nymphus.
On the top register of a wall painting in the church of
Santa Prisca in Rome, a Nymphus wears a veil that
covers his face. On the bottom register he holds the

117/754

veil revealing his face. See van Essen and Vermaseren, Excavations, 157 (top register), 169 (bottom register). Cf. the Mithraeum of the Painted Walls at Ostia, where a young man, thought to represent a Nymphus, again wears female clothing (CIMRM 1:
13031, no. 268). See also Beck, Studies in Religion
9; Gordon, JMS 3 (1980): 4864, esp. 50. Greg. Naz.,
Or. 4.89, explicitly likens the public torture of Bishop
Marcus of Arethusa to the humiliations inflicted on the
body in Mithraic rituals. (I owe this reference to Garth
Fowden.) Naguib, Numen 41 (1994), esp. 23643,
discusses the martyr's body as a symbolic vehicle for
the expression and transgression of cultural norms in
her study of Victor, a Christian soldier-martyr in the
early fourth century who was exiled from Antioch to
the Egyptian desert.
61. E.g., Psalm 135 at Pass. gr. 6 and Sev. Ant.,
Hom. cath. 57, pp. 8687; for Maximinus's speech
and the response of Sergius and Bacchus, cf. Pass.
gr. 89 with Hom. cath. 57, p. 89; for Maximinus's
rage and the parade of the officers, stripped of rank
and dressed in women's clothing, cf. Pass. gr. 7 with
Hom. cath. 57, p. 87.

118/754

62. Sev. Ant., Hom. cath. 57, pp. 8586.

Page 25
eyewitness to the martyrdom. In describing the chasm
that opened "then" (
) to protect the sacred site of
Sergius's martyrdom, he writes that even "until the
present day" (
) the chasm was a visible
witness for the faithless. The author goes on to describe the attempted relic-robbery, which took place a
long time after the martyrdom, and the construction,
even later, when Christianity had spread throughout
the land, of a martyrium within the city walls by fifteen
bishops and their translation of the relics to the new
shrine. The last events mentioned in the Passio are
the translation of the relics within Rusafa's walls and
yearly celebrations of the martyr's feast at the new
churchevents well enough established to give an
edge to the author's observation that the site of
Sergius's execution still produced more miracles than
the new martyrium within the walls.
The verse homily on SS. Sergius and Bacchus by Jacob of Sarug (d. 521) also bears the imprint of the
Passio's influence.63 The date of the homily is not
known. The somewhat repetitive mimra is composed
of 152 couplets and, like Severus's homily, was probably intended for public performance at a feast. Jacob

120/754

took liberties to elaborate particular themes: Sergius's

service to two kings, the waywardness of polytheist
religion, the martyr's Christ-like endurance of humiliations. He omits details in the Passio that Severus
picked up on, such as the parade in women's clothing,
but makes much of other events in the story. Two
themes in particular were selected: the road that Sergius is forced to walk along with nails affixed inside
his shoes, and the burial of the martyr's body at
Rusafa.

The iron thorns did not hurt him, while he was running
to go to the place of Life where He is and to live with Him
For the sake of the nails which he saw fixed in the hands
he accepted blades in his own feet without grumbling. (111)64
The final twenty couplets turn from the martyr's sufferings to his victory, and in these few lines Jacob develops a theology of relics in the frontier region:

A region full of danger fell to this energetic soldier,

and his fame and his name made peace in it and he is its

121/754

63. This homily has received no scholarly attention as

far as I am aware. I wish to thank Andrew Palmer for
translating it for me. All quotations are from his unpublished translation.
64. Jacob of Sarug, Mimra on the Victorious Sergius
and Bacchus, AMS 6: 65061; cf. 10, 83, 96110.
Pitra, Analecta sacra, 28992 with n. 1, published a
Greek hymn to SS. Sergius and Bacchus, written by
Elias, patriarch of Jerusalem. There are two possible
candidates for this Elias, one of whom flourished in
the early sixth and the other in the mid eighth century.
On which Elias was the author, Pitra's comment was
"dictu arduum est." In contrast with Jacob's, this undistinguished hymn reveals no particular familiarity
with the Passio's story, or any knowledge of the saints
or their cult.

Page 26
A frightening post full of shuddering was assigned to him,
because he was fit and alert and tough and warriorlike.
The hoard of his bones, like that legion of champions,
guards the country from harms by the power of God.
Joseph's people acquired a wall in his bones;
Jerusalem, too, used to be preserved by the bones of David. (13740)
The martyr's protective powers became highly prized
in the course of the fifth and sixth centuries. Jacob's
homily is but a foreshadowing of Sergius's role as
guardian of the frontier zone.

ALEXANDER OF HIERAPOLIS
EARLY CULT OF S. SERGIUS

AND

THE

In a brief note, Baumstark asserted that the Passio of

Sergius and Bacchus must belong to the fifth century;
he arrived at this conclusion only by grouping the

123/754

Passio with other dated martyr accounts, but without

indicating what characteristics the Passio of Sergius
and Bacchus shares with them.65 Considered together with both the external literary evidence and the epigraphical and archeological evidence now available,
the Passio does indeed seem to date from the fifth
century, but after the 440s. To reach this more precise
date, we must turn to the activities of Alexander of Hierapolis, who flourished in the first decades of the fifth
century.
As we have seen, the Passio must have been composed in Greek by 514 when Severus relied on it for
his homily at Chalcis; and the Passio's account mentions the dedication of a martyrium within Rusafa's
walls by fifteen bishops. It is also known that, not long
before 431, Bishop Alexander of Hierapolis invested
three hundred pounds of gold in a church at Rusafa in
commemoration of the martyr Sergius.66 In 431, Alexander was part of the cohort of bishops who arrived at
the Council of Ephesus under the leadership of John,
patriarch of Antioch. Their journey had been delayed,
causing them to arrive only after Nestorius's condemnation, which had been carefully engineered by Cyril
of Alexandria. Alexander remained adamant in his

124/754

anti-Cyrillian stance even after 433, when John and

Cyril accepted the Formula of Reunion. After the reconciliation, John and his supporters made repeated
attempts to moderate the stance of Alexander and the
other bishops who held out against Cyril, including
many from the
65. Baumstark, Geschichte, 95 and n. 2. Brock in
Lapridge, Archbishop Theodore, 41, groups the
Passio of SS. Sergius and Bacchus with other Greek
passiones translated to Syriac in the fifthseventh
centuries. The earliest manuscript of the Passio is
sixth-century.
66. ACOec. 1.4, p. 185; cf. 1.4, pp. 16263.

Page 27
province of Euphratesia.67 In a letter to Theodoret,
bishop of Cyrrhus, Alexander asserted that he could
not bear to join in communion with John, even if the
latter were to grant Alexander the entire kingdom of
heaven (over which, Alexander added, John did not
rule), or even Rusafa and the other cities (civitates) in
the desert (solitudo).68 John resorted to diminishing
the influence of the rebellious bishopsthe bishop of
Doliche was replaced, and in about 434, without consulting Alexander, John traveled to the walled settlement that housed S. Sergius's shrine and consecrated
a bishop of Rusafa, which until then had fallen under
the direct control of the metropolitan of Hierapolis.69
Alexander's activities at Rusafa and John's separation
of the town from Alexander's direct patronage are
clear signs that Rusafa's stature as a pilgrimage site
had grown sufficiently to make it a prized possession.
The Passio also claims that the miracles of Sergius at
Rusafa drew many pilgrims.70 There is evidence that
the cult of S. Sergius had already begun to spread
from Rusafa by the early fifth century: Edessa / Urfa
and a village now called
, northeast of Edessa,
had churches dedicated to Rusafa's martyr from

126/754

around the time when Alexander showed an active interest in Rusafa in the 430s.71 John took the radical
step of wresting a potential seat of influence from a
defiant bishop's control. Alexander vainly protested
that the Antiochene robbers were after his gold.72 No
fifteen bishops are mentioned in Alexander's letter of
protest, where he includes among his losses his investment in the church at Rusafa. We can be confident that Alexander would have rallied as many bishops as he could, including even his friend Theodoret
of Cyrrhus, for the consecration of the miracle-working
Syrian martyr's shrine.73 But we should not expect
the metropolitan to have distracted the reader from his
loss with such details. Alexander took the controversy
very personally, and this no doubt only compounded
John's wariness of Alexander's direct control over
Rusafa. On 15 April 435, deaf to pleas for moderation
by fellow bishops, including Theodoret, Alexander
was exiled to the Egyptian mines. Rusafa's fame only
continued to increase.
67. Frend, Monophysite Movement, 23; Devreesse,
Patriarcat, 5153.
68. ACOec. 1.4, p. 171.

127/754

69. Ibid., pp. 16263, 185.

70. Pass. gr. 30; Pass. syr., p. 322. Ulbert in Actes,
445, suggests a link between the translation and the
growth of the cult's popularity.
71. See chapter 4.
72. ACOec. 1.4, pp. 18485.
73. Cf., e.g., the invitation of neighboring bishops, abbots, and priests by Bishop Perpetuus of Tours to the
feast of S. Martin in the 450s, at which Perpetuus consecrated the miracle-working saint's magnificent new
church (Greg. Tur., De virtutibus S. Martini 1.6).

Page 28
Discovery of a pivotal inscription originally located in
the structure called basilica B by the archeologists, together with excavations beneath this building, have
helped identify Alexander's church. The inscription
states that the stone-built structure was begun in
518.74 It also relates that the new building replaced
an older shrine built of mud brick; and indeed a brick
building, finds from which date it no later than 425,
has been discovered beneath the stone basilica.75
The new evidence allows Konrad, the excavator of the
levels beneath basilica B, to confirm that the early
fifth-century building is the church known to have
been built at Rusafa by Alexander, and the same one
consecrated by fifteen bishops, as the Passio
relates.76 Allowing a short period of time to justify the
observation by the Passio's author that the original
site outside the walls worked more miracles than the
new shrine, the composition of the Passio can be
placed safely in the mid fifth century.
The insistence of the Passio's author that the original
shrine was still in his day, that is, in the mid fifth century, the site of miracle-working deserves our attention. One way in which the memory of the traditional

129/754

place of martyrdom could have been kept alive would

have been if the feast of the saint included a procession linking the site of martyrdom with the new shrine
within the walls that held the relics. One can imagine
how, commencing from the shrine within the walls, a
procession might have passed from the new shrine to
the original tomb outside the walls. The relics would
have been carried in the procession, escorted by the
region's ecclesiastical and secular leadership, perhaps even a military guard. At the site of the martyrdom, the Passio would have been read, litanies performed and hymns sung in honor of the saint.77
There is no doubt that by the mid fifth century, Rusafa
contained within its walls the shrine of S. Sergius. But
no evidence, not even the Passio, suggests that S.
Bacchus too was buried there. In fact, the Passio
spends little time on Bacchus. No mention is made,
for instance, of his veneration after his body was
saved by the Euphratine monks. Bacchus does nonetheless play a part, if only a cameo role, in the story
related by the Passio, and the consequence of this
was that as S. Sergius evolved into one of eastern
Christianity's greatest defenders, he would sometimes
appear alone, and at other

130/754

74. Ulbert and Gatier, DaM 5 (1991): 16982. For discussion of the inscription and its place in the cult's development at Rusafa, see below, pp. 8091.
75. Konrad, DaM 6 (1992): 34344.
76. Ibid., 349 and n. 181; Kollwitz, Neue Deutsche
Ausgrabungen, 46; Ulbert in Actes, 447. Kollwitz was
tempted as early as 1959 to identify the church of the
fifteen bishops with Alexander of Hierapolis's church
(Neue Deutsche Ausgrabungen, 46).
77. For processions, vigils, miracula readings, and
banquets accompanying saints' feasts, see Maraval,
Lieux saints, 21321. For the probable location of the
tomb outside the walls, see below, pp. 14973, esp.
pp. 15759.

Page 29
times with S. Bacchus by his side. From Severus's
homily, it appears that, at least in 514, the feast at
Chalcis was held in honor of Sergius aloneSeverus
had to lobby for Bacchus's inclusion. Severus later
wrote three hymns in honor of Sergius, but only one
featured Bacchus.78 Severus may have insisted that
what God had joined in martyrdom should not be separated in pious remembrance. But those who came to
know the powerful martyr Sergius on his own at
Rusafa and at local feasts, may have thought it unnecessary to include Bacchus. Others welcomed him.
The result was not terribly uniform. In Syria and Mesopotamia, the predominance of Sergius was a given,
the appearance of his less-spectacular fellow martyr
only occasional.

IMAGES OF S. SERGIUS
Naturally, with the spread of the cult and the Passio
there arose also a demand for images of S. Sergius,
both with and without S. Bacchus. Two types appeared. The first closely reflects the martyr we encounter in the Passio, the standing soldier-saint with a
gem-studded maniakion prominently displayed around

132/754

his neck. The second is a rider saint with a flowing

cape and a martyr's staff. Only the first of these types
has been discussed in the scholarly literature, and
that briefly. None of the known images can be dated
earlier than the mid sixth century.
A good example of a representation of both Sergius
and Bacchus together as standing soldier-saints is to
be found on an offering dedicated at the church of S.
Sergius at "Kaper Koraon," a village somewhere in
the limestone hills of northwest Syria (fig. 1).79 A silver flask, possibly meant for oil, portrays four independent nimbed images in the following sequence:
Christ, a soldier-saint, Mary, a soldier-saint. The
saints, in the pose of an orans, wear chlamys and
chiton fastened at the shoulder by a fibula with prominent maniakia around their necks. The flask is
thought to date to the mid to later sixth century. Marlia
Mundell Mango has proposed that the soldiers represent SS. Sergius and Bacchus, basing her identification on the flask's inclusion in a hoard of objects,
many of which are inscribed with dedications to S.
Sergius, as well as by comparison with other images.

133/754

One of these is the encaustic icon from Sinai, although probably produced in Constantinople, which
shows the busts of two very young soldiers, each
crowned by a nimbus and dressed in a gold chiton
with a carmine
78. James of Edessa, Hymns of Severus of Antioch
14345, pp. 18991.
79. Mundell Mango, Silver, no. 15.

Page 30

Figure 1. SS. Sergius and Bacchus. Silver flask from

the Kaper Koraon treasure, Syria. The Walters Art
Gallery, Baltimore, no. 57.639.

Page 31

Figure 2. SS. Sergius and Bacchus. Encaustic icon

from the Monastery of S. Catherine, Sinai. City Museum of Eastern and Western Art, Kiev, no. 111.
clavus and a white chlamys held in place by a gold
clasp (fig. 2).80 Each of the pale figures holds a gold
martyr's cross and stares directly ahead with wide,
watery eyes. They give the impression of order and
total obedience, with their hair and clothes carefully
arranged. A tiny bust of Christ floats between them.
Later, the soldiers were clearly labeled with inscriptions, Sergius to the left, Bacchus to the right. But
their identity would have been immediately apparent

136/754

from their maniakia, here of gold, decorated with two

rectangular gems or medallions on either side of one
central oblong cabochon gem.81 The icon is believed
to date from the early seventh century.
80. Weitzmann, Icons, 1: 39, 2830, pls. 12, 52, 53.
Weitzmann notes that the maniakia distinguish the
martyrs from other military saints, but inaccurately
claims that they are "always depicted as a pair," even
though he compares the hieratic rendering of Sergius
and Bacchus in the Sinai icon to that of the S. Demetrius figures in the mosaics of the S. Demetrius church
of Thessalonica, where S. Sergius appears alone. As
a general but not absolute rule, the maniakion is distinctive of S. Sergius and S. Bacchus. But other military martyrs do occasionally appear with the gold
torque. For exceptions, see, e.g., the Coptic rendering
of S. Menas and S. Theodore illustrated by Nicolle in
War and Society, figs. 61, 62c.
81. Speidel,
sketches the
torques were
tention to a

Antiquit tardive 4 (1996): 23739,

court culture in which bejeweled gold
worn by the military elite. He draws atdescription of such torques in Const.

137/754

Porph., De cer. 2.52, p. 708, which closely coincides

with the depictions

Page 32
What is especially striking about this icon is the extremely youthful rendering of the martyrs.
A hardier portrait of S. Sergius, although unlabeled,
appears on the tondo of an elegant silver bowl, dated
by silver stamps to the reign of Constans II (64168),
and discovered in Cyprus (fig. 3). Here, a single military saint holds a martyr's cross and wears the chiton
and chlamys, the nimbus and the maniakion.82 With a
much bulkier body and less orderly rows of curls than
the Sinai Sergius, this figure presents a more convincing image of a primicerius scholae gentilium. The
maniakion he wears is studded with a large, round
medallion or gem surrounded by four smaller medallions. The configuration is nearly identical to that depicted on the Kaper Koraon flask. The prominence of
the saints' neckgear in these objects recalls the
Passio, where the first stigma is the removal of the officers' maniakia. This emphasis on the maniakia in the
Passio, which is repeated in the portraiture, helps
verify the identification of the two unlabeled imageson the Cyprus bowl and the Kaper Koraon
flaskas representations of S. Sergius.

139/754

Another example of the same type is the late seventhcentury full-length mosaic icon of S. Sergius that
takes a position of honor to the right of the apse in the
church of S. Demetrius, another soldier-saint, at Thessalonica (fig. 4).83 Sergius is depicted as an orans,
wide-eyed, staring directly ahead. He wears an elaborately decorated chlamys with a tablion and a chiton,
attached at the shoulder by a fibula. Around his neck
is the characteristic maniakion, very similar in design
to that worn by the soldier-martyr on the Cyprus bowl
and the Kaper Koraon flask: that is, attached to the
gold neckband are four small medallions or gems,
which encircle a large central medallion. As in the
Sinai icon, here too Sergius's face appears young,
austere, and, to the modern eye, effeminate. What appear today as feminine features were often used to
evoke youthfulness in portraiture of

in the images of Sergius discussed above. It should

be noted that in the Passio, the torques are described
simply as golden, so we may assume that the addition
of jewels reflects the early seventh-century milieu to
which the portraits belong. One may hazard the suggestion that the appearance of a gold torque

140/754

embellished with precious stones in a later Armenian

synaxarion was inspired by the visual representation
of the soldier saint in this guise (Armenian Synaxarion
of Ter Israel, 378).
82.
in Age of Spirituality, ed. Weitzmann,
54849, no. 493.
83. For color photographs of the seventh-century mosaics, see Bakirtzis, ed.,
, ills. 716. Soteriou,
, 19394 with pl. 65a. The church of
S. Sergius at Gaza, erected during the early years of
Justinian's reign, probably before 536, was described
in a laudatio by Choricius of Gaza. In the mosaic decoration of the church, S. Sergius is portrayed as a mediator between the building's benefactor and the
Mother of God and her Son (Chor., Laud. Marciani
1.31). Details of the saint's attire are not provided, although perhaps the Sergius mosaic at the church of
S. Demetrius in Thessalonica may offer some suggestion of how he was portrayed at Gaza.

Page 33

Figure 3. S. Sergius (?). Silver bowl from Cyprus. The

British Museum, M. and L.A. 99, 425, 2.
the period. Perhaps the most evocative attempt to describe this tendency in divine portraiture of the East
was penned by Michael Rostovtzeff:

142/754

Most of the Palmyrene gods and some of the goddesses are noble youthful figures with rich curled hair,
resplendent in their boyish beauty; and this is true
also of many of the heroized men, women, boys, and
girls represented in relief or in the round on the funerary monuments. Their features are always somewhat
effeminate, with peculiar languid eyestraits which
are both highly typical of the later Oriental art of the
Near East. . . . Despite their military dress, the military
gods of Palmyra are refined, elegant ephebes of the
Oriental type. This effeminacy is not, however, only
sensuous. The graceful figures of the boyish gods and
of their curly-haired attendants, the slim proportions

Page 34

144/754

Figure 4. S. Sergius. Mosaic, basilica of S. Demetrius,

Thessalonica. Ephoria of Byzantine Antiquities,
Thessalonica.

Page 35
of their bodies, the romantic eyes, their almost airy
appearance enable us to grasp at once, even without
the help of the halos and radiate crowns which surround the heads of the gods, their solar, ethereal, and
celestial nature.84
The last image to fall within this portrait type of S. Sergius is perhaps the most striking, although it is not
found in the few brief accounts of the saint's iconography that exist. It appears on a reused cameo dating from the late fourth century onto which the names
"Saint Sergius" and "Saint Bacchus" have been
etched. The cameo is believed to have been carved
as a portrait of the emperor Honorius with his bride
Maria, the daughter of Stilicho, on his left (fig. 5).85
The reuse seems less bizarre when the hair style and
general effeminacy of the young soldiers in other representations is taken into consideration. To make interpretation even more complicated for modern viewers, one recent discussion of the cameo proposes that
Maria's (later Bacchus's) hair style resembles that of
the men rather than the women of the House of
Theodosius.86

146/754

The second type of Sergius image represents the

saint with nimbus on horseback. The rider wears military attire, with a flowing cape and a long martyr's
staff. The horse does not gallop, prance, or rear up on
its hind legs, but rather holds a noble pose with its left
forefoot raised. A bronze attachment said to have
come from Palestine is decorated with an intaglio
image
84. Rostovtzeff, YCS 5 (1935): 237. Cf. also Walter,
REB 53 (1995): 320. At the beginning of the twentieth
century, icon painters on Mount Athos used photographs of women or of beardless boys or ancient
statues as models for martyrs. To his astonishment,
one visitor observed a portrait of Kyra-Basiliki, the
concubine of Ali Pasha, standing in as a model for the
Protomartyr Stephen (Moyses,
35; I owe this reference to Garth Fowden).
85. Delbrueck, Consulardiptychen, 26061, proposed
that reused Sergius and Bacchus cameo served as an
icon in the church dedicated to the saints in Constantinople, but this suggestion must remain conjectural. Even Boswell, Same-Sex Unions, does not suggest that Bacchus dressed up as a bride. Cf. a bust of

147/754

Antinous recarved in the third century to represent a

middle-aged woman with a Severan hair style
(Blanck, Wiederverwendung alter Statuen, 45 and pls.
1415). Another example is the gold reliquary representing the female saint Fides, which in its original
form, before the gem- and cameo-encrusted statue
was made to do duty as Fides, depicted a Roman emperoreither Marcus Aurelius or Julian (MacMullen,
Christianity, 128, with n. 87, and 129, fig. 3). I would
like to thank Christopher Walter for drawing my attention to another cameo relabeled in secondary use as
Sergius and Bacchus. This onyx gem depicts two soldiers en face wearing the cuirass and holding lances.
They have been identified as Caius and Lucius, adopted sons of Augustus: DACL 6.1.857, no. 5143.
86. Meischner, AA 1993: 61319, with photograph, although the later inscriptions appear only faintly. See
also Kiilerich, Late Fourth Century Classicism, 9294,
with fig. 46, according to whom Maria's hairstyle most
closely resembles that of Constantine's mother,
Helen, with echoes back to Livia. Neither study alludes to the subsequent rebaptism as Sergius and
Bacchus.

Page 36

Figure 5. The imperial couple Honorius and Maria,

later inscribed as SS. Sergius and Bacchus. Rothschild cameo. Private collection, Paris.
of the saint in this guise, while around the image an
inscription proclaims the allegiance of a camel-driver
to S. Sergius: "camelarius of Saint Sergius" (fig.

149/754

6a).87 One could imagine that its owner belonged to a

company of cameleers working under the banner of
Sergius and assuming responsibility for caravan trade
from their base at Rusafaas the Palmyrene companies once had. Or the camelarius could have belonged to a monastery of
87.
: Muse 7 (1973): 11. The camelarius would have been under the absolute authority of
the caravan leader, the
. For a vivid account of
the practicalities of a caravan, see Grant, Syrian
Desert, 18992.

Page 37

Figure 6. a: S. Sergius. Bronze attachment. Courtesy

Museum of Art and Archaeology, University of MissouriColumbia, no. 72.73.
b: S. Sergius. Bronze bracelet. On Beirut antiquities
market in 1960. Published by Mondsert, Syria 37
(1960), fig. 4.
c: S. Sergius (?). Lead seal. Courtesy of Mrs. Janet
Zacos, Athens.
d: "Holy rider." Silver armband. Royal Ontario Museum, Toronto no. 986.181.93.

151/754

S. Sergius at Rusafa, on the model of the Pachomian

camelarii.88 In the late 1950s, a bronze bracelet appeared on the antiquities market in Beirut and was
photographed by Henri Seyrig and published by
Claude Mondsert, but it has not surfaced in scholarly
discussion since that time (fig. 6b).89 Thirty-five mm
in diameter, the bracelet's medallion shows a nimbed
saint on horseback, wearing a flowing pallium and
carrying a cross staff. An inscription follows the edge
of the medallion: "camelarius of Saint Sergius of the
Barbarian [Plain]."90
88. Pall., Hist. Laus. 32.9.
89. Mondsert, Syria 37 (1960): 12325.
90.
. See now the comments of
Shahd, BASIC, 5078. On the "Barbarian Plain," see
pp. 12 above and pp. 6566 below, with notes.

Page 38
The image of a saint on horseback, a speedy guardian of cameleers but also of pilgrims, would have had
immediate appeal in the potentially hostile environment of the steppe. In addition to a flask of blessed oil
or water, pilgrims to the martyrium would have taken
away from Rusafa clay or metal medallions commemorating their visit to the martyr. Such tokens have survived from the busy shrine of Symeon the Younger,
the subject of a study by Gary Vikan.91 The tokens
bearing the image of Symeon were ceramic in the
sixth and seventh centuries and made of lead during
the revival of pilgrimage to the shrine in the tenth century. They were often inscribed with words such as
"blessing" (
) and "health" ( ), which crystallized
the nature of the bearer's rapport with the saint, particularly in the case of Symeon, who was famous for the
miraculous cures worked by his prayers.92 Such
tokens received at the saint's shrine were mementos
in a powerful sensethey were made to do more than
simply commemorate a visit, like an empty souvenir.
The token placed the bearer in continuing contact with
the image, the prayer and the healing power of the
saint. In the 1920s, the Aleppo collector Guillaume
Poche was shown a stone mold on which the

153/754

inscription

"the

blessing of Saint Sergius" (

) was carved around an image of
Sergius on horseback, with a cross resting on his
shoulder.93 What is more, the mold was said to have
been found at Rusafa, a believable claim. Most likely,
the object surfaced during clandestine digging and
found its way onto the antiquities market. Unfortunately, no illustration accompanies Poche's description as reported by Mouterde and Poidebard. But the
description is enough to relate the mold to the image
of the rider saint on the two bronze medallions owned
by cameleers. To these three nearly identical portraits
of Sergius on horseback should be added a lead seal
dated 550650 by Zacos and Veglery (fig. 6c).94 The
authors describe the seal's obverse as follows: "saint
riding horse to right, wears nimbus, chiton and himation; holds in right hand cruciform spear with which
91. Vikan, DOP 38 (1984): 6586, esp. 73, on the pilgrim token as both "distillation of and catalyst for" late
antique belief in miraculous healing through saints. A
seventh-century fragment of parchment shows S. Sergius as a Roman official, nimbed and, uncharacteristically, holding an orb in his left hand, while he blesses
with his right. The saint is clearly labeled in Greek

154/754

. A cord, part of which is still intact, was attached to the parchment image, so that it could be
worn as an amulet (Horak in Syrien, 19495).
92. Vikan, DOP 38 (1984): 6769.
93. Poidebard and Mouterde, AB 67 (1949): 115,
quote Poche's description from a letter dated 6 October 1924. Vikan, DOP 38 (1984): 74, with n. 47, describes molds used to make both lead and clay
tokens. Poche, in that portion of his letter quoted by
Poidebard and Mouterde, did not include such
specifications.
94. Zacos and Veglery, Byzantine Lead Seals, no.
2975, with photograph in plate vol. 1, pl. 202, no.
2975.

Page 39
he attacks a prostrate enemy (not visible)." In the light
of the three other images of Sergius on horseback, it
seems probable that the prostrate foe is invisible because he does not exist. Zacos and Veglery were induced to imagine the prostrate figure by assimilating
this image either to the commonly portrayed figure of
the "holy rider," sometimes named as Solomon, other
times anonymous, or to later icons of S. George or S.
Demetrius.95
The inscription on the seal's reverse supports the suggestion that the rider represented is S. Sergius. Abbreviated Latin characters proclaim the seal's owner:
Sergii illustris et commerciarii.96 It is relatively rare for
a rider saint to appear on a lead seal, and when a
saint is depicted, the image is not always identified by
an inscription. Often, as seems to be the case with the
seal of Sergius the commerciarius, the reverse will
simply bear the name and title of the seal's owner.
The saint shown may be the owner's patron saint, his
homonym, although this seems not to be the case
automatically. For example, one seal with a rider saint
on its obverse is marked on the reverse with what
must be the owner's name, Stephen, an unlikely rider

156/754

saint.97 But because of the close resemblance

between the image used by Sergius the commerciarius and that on the mold and the two bronze medallions, there can be little doubt that the lead seal portrays the commerciarius's patron saint, Sergius.
The similarity between the four portraits of Sergius as
a rider saint suggests that they represent an image
disseminated from the pilgrimage shrine at
Rusafaparticularly since the mold, devised for the
purpose of reproducing an image, shows Sergius on
horseback. The relationship between these four objects, not hitherto discussed, restores to the thoroughly robbed shrine a cult image popular with pilgrims. It also teaches caution. Nowhere in the Passio
is Sergius associated explicitly with a horse. The
schola gentilium was, of course, a horse guard, but
this fact failed to ignite the imagination of the author of
the Passio, who not once mentions his hero in connection with a horse. However, the popularity of the
rider image is confirmed by a Greek inscription from
the Hawran that refers to S. Sergius as a rider.98
There is also a vivid illustration of personal contact
with Sergius the rider

157/754

95. Walter in DCAE 15 (198990): 3342, focuses on

the armed rider who spears a prostrate figure, often a
woman or a monster. He traces the development of
this widely disseminated image, both pre-Christian
and Christian, through several transformations.
96. This is the only lead seal of a commerciarius in
Zacos and Veglery, Byzantine Lead Seals that uses
Latin characters. The commerciarius, first attested under Anastasius, levied taxes on imports, exports, and
commercial transactions (Hendy, Studies, 174 and
654 n. 436; Stein, Histoire, 2:214, with n. 1, and 750,
n. 1).
97. Zacos and Veglery, Byzantine Lead Seals, no.
1311, with photograph in plate vol. 1, pl. 103, no.
1311, dated by the authors to the second half of the
sixth century.
98. See below, pp. 11011.

Page 40
saint in the Syriac life of Mar Qardagh, datable most
likely to the late sixth or early seventh century.99 In
his first night vision of the martyr who was to become
a role model for this Iranian aristocrat, Qardagh sees
a young man armed and mounted on his horse. The
cavalier stands before Qardagh, strikes him on the
side with his spear, calls out his name, "Qardagh,"
and foretells his future martyrdom. When asked who
he is, the mounted messenger replies, "I am Sergius,
the servant of Christ."100 The vision reveals not only
the power of the rider iconography, but also how well
it travels, in this case to late sixth-century Adiabene,
due east of Rusafa beyond the Tigris. The story also
underscores the influence of established iconography
in mortal imaginings of divine figures.101
The mounted image of Sergius was inspired by the
martyr's role as guide and divine protector in the semidesert landscape.102 Armed immortals had long been
appealed to in Syria. Rider gods were particularly at
home in the Syrian steppe, and their divinity was further charged by the strength and speed of the muchvalued horse. Foes mortal or immortal often appeared
on horseback: just as the speed of the mounted Arab

159/754

allies was a tangible asset remarked upon by the historian Evagrius, so too S. Symeon, in his ascetic seclusion, waged war with a spiritual enemy who assumed the outward form of cavalry.103 A group of
stone relief carvings depicting an Arab rider god,
sometimes identified as
, show the divinity with a
flowing cape, lance in the right hand, shield in the other (fig. 7). The stately mount is represented in profile
with one foreleg raised, while the rider turns to face
the viewer.104 The resemblance to the image of S.
Sergius on horseback is striking. All of the reliefs have
been found in the area between the Hawran and
Aleppo, that is to say, in the area where the Sergius
cult put down its earliest roots. Ernst Lucius, who first
drew attention to the similarities between
99. See below, p. 120.
100. Acta Mar Kardagh, 7. In the mid eighth century,
SS. Sergius and Bacchus rode up on horseback to
the last Umayyad caliph, Marwan II, as he was torturing the patriarch of Alexandria (History of the Patriarchs 1.18, p. 428).
101. Cf. Mir. Theclae 12, with Dagron, Vie et miracles,
98; and Mir. Demetr. 10.

160/754

102. In nineteenth-century Cappadocia, Greek Orthodox children were still taught to call on a rider
saintusually George or Theodorewhen in need of
immediate rescue. See, e.g., Passios Hagioreitis,
, 33.
103. Evagr., HE 5.20; Syriac Vita S. Simeon. Stylit.
93.
104. Ronzevalle, CRAI 1904: 812, discusses a relief
discovered south of Damascus; Mouterde, MUSJ 11
(1926): 30722, compares two reliefs from south /
southeast of Aleppo; Rostovtzeff, JRS 22 (1932):
10910, comments on the iconography of a relief of
and
from Palmyra; Seyrig, Syria 47 (1970):
77100, collects evidence of armed Arab deities, often in pairs, that has been found especially in the regions north and south of Palmyra. See also
Rostovtzeff in Excavations at Dura-Europos, ed. Baur
and Rostovtzeff, 199200, pl. XXIV. 1, 2, 3, for a discussion of three comparable terracottas.

Page 41

Figure 7. Rider god. Stone relief found south of

Damascus. Published by Ronzevalle, CRAI 1904, fig.
1.
and S. Sergius, was not familiar with the relief images,
but had been impressed by the early appearance of
dedications to S. Sergius in Edessa and the Hawran,
where the cult of
had also left pronounced
traces.105 Lucius's observations have occasionally

162/754

been cited in passing and without comment.106 But

the close iconographic relationship between the Arab
military god and the soldier-saint speaks volumes for
the critical role played by the landscapeand the distinctive ways of life it gives rise toin the formation of
religious conception and expression. While attempts
to trace the origins of rider iconography have quickly
led to tenuous and almost
105. Lucius, Anfnge des Heiligenkults, 240.
106. See, e.g., Honigmann in RE, 4A: 1707.

Page 42
meaningless associations, the case of S. Sergius and
the Arab rider god is bolstered by the geographical
framework in which both images circulated.107 It requires no great leap of the imagination to Christianize
the second-century inscription that accompanies an
elegant Palmyrene relief showing the camel-riding
god
, followed by
on horseback:
To
and
, the good gods who reward,
has
made [this], son of Jarhibole, priest of
the good
and merciful benevolent god, for his safety and that of
his brothers.108
At issue here is not some crypto-paganism but rather
the instinctive response to particular environmental
circumstances unchanged, to a great extent, for millennia.109
Another image closely akin to the first four depictions
of Sergius as cavalier may confirm the link of the rider
Sergius with pilgrimage. In his study of "implements
for supernatural healing," Gary Vikan published a photograph of a silver amuletic armband with four medallions, one of a dozen such objects known to scholars.110 The armbands typically show scenes from

164/754

holy sites or from the life of Christ, and often include

intercessions for the owner's well-being. A figure Vikan calls the "holy rider," associated with healing and
the subjection of evil forces, also appears on many of
the bracelets, whose iconography, as Vikan argues,
originated in pilgrim tokens (fig. 7d).111 The Christian
"holy rider" is a nimbed figure on horseback who
wears a flowing pallium and holds a long spear,
sometimes topped with a cross, which he aims at the
prostrate figure shown below his rearing horse's
feet.112 The silver armband in question is typical,
among this category of objects, in its inclusion on its
four medallions of the women at the tomb of Christ,
the Theotokos and child, and the "holy rider." The
Trisagion appears on the fourth medallion, while
Psalm 90.1, an apotropaic verse that frequently appears on amulets, is inscribed around the band that
links the medallions. "Health" is carved beneath the
so-called holy rider, and around the image of the
Theotokos and child is written: "Theotokos help Anna
Charis." What is unusual about this excellently preserved arm107. On this problem, see Walter, ByzF 14 (1989):
1.65773 with 2. plates 24955.

165/754

108. Rostovtzeff, JRS 22 (1932): 109.

109. Weber, DaM 8 (1995): 20311, draws parallels
between iconographic representations of paired Arab
caravan gods, the twins Castor and Pollux, and the
Christian saints Cosmas and Damian, but with no reference to Sergius and Bacchus.
110. Vikan, DOP 38 (1984), fig. 10.
111. Ibid., 75.
112. Bonner, Studies in Magical Amulets, 21112,
discusses the pre-Christian antecedents of this image.
A recent study of amulets portraying the "holy rider"
confirms that the prostrate figure was part of the image in both pre-Christian and Christian tradition: see
Spier, JWI 56 (1993): 3338.

Page 43
band is the rendering of the so-called holy rider. Vikan
remarks that this figure "differs from the more common sort . . . insofar as he does not impale an evil
prostrate figure or beast." Vikan derives this more
"stately version" from adventus iconography, particularly from depictions of Christ's entry into Jerusalem.113 But in the images discussed by Vikan that
portray Christ's triumphal entry into the Holy City, the
mount is clearly a long-eared donkey. We do not have
to turn to Christ the holy rider to find a good parallel to
this unusual representation. The horse's stance in
particular, with the left front leg raised in a position of
controlled movement, but not of rearing attack, recalls
the image of Sergius on the three objects depicted in
fig. 6a, b, and c. If the rider saint Sergius is depicted
here, then the addition of "health" beneath Sergius's
image fuses the portrait of the mounted protector with
the martyr Sergius, whose healing powers were
demonstrated especially at the pilgrimage site of
Rusafa and acclaimed in both the Passio and
Severus's homily.
It is impossible to divine from the handful of Sergius
images that have survived when the two

167/754

representations first appeared. The class of images in

which Sergius is the standing soldier wearing a
maniakion, sometimes accompanied by Bacchus, was
inspired by the verbal Passio portrait and closely related to the story's circulation. It may, at least in the
early stages of iconographical development, have appealed more to those remote from the Barbarian
Plain, and perhaps to a more urban following. The
rider image must have arisen locally from the context
of the frontier zone, criss-crossed by a wide range of
people, including Arab pastoralists. The known rider
images seem to be more closely linked with Rusafa
and its environs than the standing soldier portraits
found at Sinai, Thessalonica, Cyprus, and Kaper
Koraon. It is surely significant that, especially in the
case of the Sinai and Kaper Koraon images, Constantinople is thought to have been their place of origin.114 Although much depends on the accidents of
survival, the high-quality seventh-century images from
Sinai, Cyprus, and Thessalonica may represent a
surge of interest in the soldier-martyr as Christian
Rome's patron at a time when Arabs, Avars, and
Slavs were harassing the empire's frontiers. Perhaps
it was his reputation as a celebrated military defender
in the East that inspired the Thessalonicans to honor

168/754

Sergius so prominently in their church of S.

Demetrius.
113. Vikan, DOP 38 (1984): 75 n. 57. See also
Vikan's more recent discussion of the "holy rider" image in
, 34146, with pls. 19798.
114. Grabar, Martyrium, 2: 26, conjectured the existence of a single prototype at the Sergius shrine at
Rusafa, on which five known images of Sergius dating
from the sixth to the eleventh centuries were based,
namely, the mosaic in Thessalonica; the fresco at S.
Maria Antiqua in Rome; the Sinai icon; the Cyprus
bowl; and the Sergius mosaic at Daphni near Athens.

Page 44
The composition of the Passio of Sergius and Bacchus in the mid fifth century gave impetus to, but was
probably also itself encouraged by, the spread of the
cult of S. Sergius. Until that time, the story of Sergius,
and of Bacchus, had been preserved at the martyr's
shrine and put into circulation by pilgrims. With the
cult's expansion, a written account as well as sacred
images were needed to make the martyr more widely
available to the many believers who would never visit
Rusafa but wished to place their faith in Sergius's miraculous powers.115 Dissemination of the cult owed
much to the mobility of the pastoralists, merchants,
and soldiers who traversed the Barbarian Plain. But it
must also be seen in the context of the heating up of
Roman, Iranian, and Arab relations in the shrine's vicinity, and of the role that was developing for martyr
cult more generally in the frontier zone. In the early
fifth century, Marutha, the bishop of Mayperqat and a
diplomat who traveled back and forth between the Roman and Iranian courts, was busy planting another
martyr cult in the frontier zone. The next chapter examines the martyr cult that was deliberately established at Mayperqat in northernmost Syria-Mesopotamia on the frontier between Rome and Iran, just

170/754

before Alexander's construction within Rusafa's walls

of the martyrium to the soldier-martyr Sergius. The example of Mayperqat not only helps to contextualize
the Sergius cult at Rusafa, but suggests ways in
which martyrs' bones could play a dynamic role in the
region's geopolitical developments.
115. Cf. Greg. Tur., Glor. mart. 63.81, commenting,
with reference to the cult of S. Patroclus of Troyes,
that "the men of that place had paid little reverence to
this martyr, because the story of his sufferings was
not available. It is the custom of the man in the street
to give more attentive veneration to those saints of
God whose combats are read aloud" (tr. Van Dam).

Page 45

TWO
Martyr Cult on the Frontier
The Case of Mayperqat
RELICS AND DEFENSE
Among eastern Christians, martyr cult had been
linked from the start not only to holy objects, the
bones and blood of martyrs, but also to holy sites. The
bones of martyrs were often enshrined at the reputed
site of their martyrdom. But if that spot was inconvenient for practical reasons, as we have seen in S.
Sergius's Passio, the relics could be translated to a
more appropriate place. The establishment of new
shrines or embellishment of old became the concern
of local groups, who disputed rights of possession
and, in doing so, might attract the interest of greater
powersbishops, shaykhs, and kings. Sacred sites in
the open spaces of the Syro-Mesopotamian landscape worked on multiple levels and were intimately
linked with the maintenance of order among the
region's diverse communities.

172/754

Relics could be moved around to deal with particular

problems. The translation to Constantinople of relics
of the Apostles Andrew and Timothy, either at the end
of Constantine's reign or in his successor's, was
praised by Paulinus of Nola in 405:
Indeed, when Constantine was founding the city
named after himself and was the first of the Roman
kings to bear the Christian name, the godsent idea
came to him that since he was embarking on the
splendid enterprise of building a city that would rival
Rome, he should also emulate Romulus's city with a
further endowment, by gladly defending his walls with
the bodies of the apostles. He then removed Andrew
from the Achaeans and Timothy from Asia. And so
Constantinople now stands with twin towers, vying
with the eminence of great Rome, or rather resembling the defenses of Rome in that God has counterbalanced Peter and Paul with a protection

Page 46
as great, since Constantinople has gained the disciple
of Paul and the brother of Peter.1
Old Rome fell to Alaric in 410 because, according to
opponents of the empire's new religion, the traditional
divine allies had not been evoked to defend the
Eternal City.2 When the Goths reached Athens in
396, belief in the old gods had still been robust
enough to rally Athena and Achilles to the walls and
save the city.3 But in the fifth and sixth centuries,
Christian martyrs and saints, usually represented by a
relic, were staking their claim as protectors of cities
across the empire. In direct criticism of the old gods,
Theodoret of Cyrrhus explained the omnipotence of
Christian relics:
The noble souls of the victorious traverse the heavens
and join in the dance of the immaterial beings. Their
bodies are not hidden away each in its single grave,
but the cities and villages that have divided them
among themselves call them saviors of souls and
bodies and doctors and honor them as protectors of
cities and guardians [
] and treat
them as ambassadors before the master of the

174/754

universe and through them receive divine gifts. And

even though the body has been divided, the grace
has remained undivided, and that minute relic possesses the same power as the martyr, just as if he
had never in any way been divided.4
The martyrs became champions (
) and allies (
).5 Severus of Antioch, in a hymn on the martyrs, proclaimed that "the festival of the holy martyrs is
the joy and pleasure of all churches, a strong wall for
all the inhabited earth, and the victory of kings, and
the glory of priests."6
When in the early sixth century King Theuderic was
besieging the city of Clermont, he lost heart when his
general attributed the magnitude of the city's defenses
to "the saints whose churches surround the city
walls."7 That saints and fortifications were not associated only in the minds of hagiographers is clear from
an inscription found at the southeastern corner of the
1. Paul. Nol., Carm. 19.32942; tr. Mango, after
Walsh, BZ 83 (1990): 53. On the possibility that it was
Constantine in 336 and not, as usually believed, Constantius in 357, who translated the relics, see Mango's
addendum, BZ 83 (1990): 434.

175/754

2. Zos., Hist. nov. 5.41.

3. Ibid. 6. Cf. Olympiodorus, fr. 27 (Blockley) = fr. 1.27
(Mller-Dindorf), where it is reported that during the
reign of Constantius III, three solid silver statues were
excavated in Thrace where they had been consecrated and buried to ward off barbarian invasions.
Shortly after the statues' removal, the region was inundated with Goths, Huns, and Sarmatians.
4. Theod., Graec. aff. cur. 8.1011.
5. Ibid. 39. Cf. S. Thecla, who dislodged Pallas
Athena from Mount Kokysion in Cilicia (Mir. Theclae,
2), and later appeared on the walls of Seleucia, letting
out a war cry to fend off bandits from the city (ibid., 5).
6. James of Edessa, Hymns of Severus of Antioch
216, PO 7.5.676.
7. Greg. Tur., V. Patr. 4.2.

Page 47
walls of Calama, modern Guelma in Algeria, commemorating their erection by Solomon:
Twelve and one towers altogether rose up in a row;
It seems a work of wonder, constructed so swiftly.
The postern gate behind the baths is fastened with iron.
No enemy could raise a hand against it.
No one could take by storm the work of the Patricius Solomon.
The protection of martyrs secures this postern gate.
The martyrs Clement and Vincentius guard this entrance.8
The Solomon of the inscription is Belisarius's commander, known from Procopius as an energetic fortifier of North African cities against the desert tribes.9
Solomon was a native of the region of Dara, a village
just west of Nisibis that had been fortified by the emperor Anastasius with walls and with relics of S.
Bartholomew, described as the guardian of the city.10
A relic was the perfect symbol to represent the unity
of the world's material and spiritual dimensions. By
participating in both, a relic held out to those who

177/754

possessed the holy object a hope of divine intervention in the wavering sphere of material existence.
Walls would become impenetrable, soldiers invincible.
The appeal of such a device for rulers grew in
strength, particularly from the reign of Theodosius II.
In 421, under the influence of his sister Pulcheria and
as part of Rome's preparation for war with Iran,
Theodosius II sent a gold, gem-studded cross to the
bishop of Jerusalem with instructions to install it on
Golgotha.11 The gift suggested the emperor's desired
fusion of Christ's victory with Rome's triumph over its
traditional enemy, also a persecutor of Christianity. In
return the emperor received relics of S. Stephen's
right armboth the martyr's name, which recalled a
victory crown, and his faithful right arm were ideally
suited to the occasion. When, on their journey from
the Holy City to the empire's capi8. "Una et bis senas turres crescebant in ordine totas;
/ Mirabilem operam cito constructa videtur. Posticius /
sub termas balteo concluditur ferro. Nu[ll]us malorum
poterit erigere man(um). / Patrici Solomon(is) insti[tu]tion (em) nemo / expugnare valevit. Defensio
martir(um) tuet[u]r posticius ipse. / Clemens et Vincentius, martir(es), custod(iunt) in[t]roitum ipsu(m)"

178/754

(CIL 8.5352). Cf. Durliat, Ddicaces, 1114. On the

fortification at Calama, see Pringle, Defence of Byzantine Africa, 18891.
9. Proc., Bel. 4.19.3, refering to activity in 53940;
PLRE, 3: 116777 (Solomon 1).
10.
: Joh. Diakrin. fr. 2.558, p.
157; see also pp. 6465 below. The association of
saints or martyrs with walls was widespread. Ephrem
viewed holy men as the wall and shield of Nisibis and
her countryside: see, e.g., Ephr. Syr., Carm. Nisib.
13.21 on Jacob; ibid. 17.4 on Abraham. Evagr., HE
1.13, identifies S. Symeon as Antioch's rampart and
fortress.
11. Theoph., Chron., A.M. 5920, pp. 8687. See
Holum, Theodosian Empresses, 1024; Holum,
GRBS 18 (1977): 16267.

Page 48
tal the relics arrived at Chalcedon, S. Stephen came
to Pulcheria in a dream to announce his imminent arrival. The protomartyr was received in an adventus
ceremony and installed in a martyr chapel at the imperial palace: Pulcheria seems to have orchestrated
the whole process of the translation. Between 434
and 446, with the aid of an old monk and a vision, she
also uncovered the bones of the forty martyrs of Sebaste, soldiers executed in the reign of Licinius and
destined to join the ranks of Rome's military martyrs.
Theodosius's empress Eudocia also had her part to
play in the cultivation of martyrs. At Constantinople
she built a church in honor of the previously little
known soldier-martyr Polyeuctus. After her fall from
favor, Eudocia turned her attention to Holy Land benefaction and became one of the most celebrated patrons of churches, hospices, and poor houses. She
also refortified the Holy City, an act that she commemorated in an inscription in which she grounded
her endowment in the prophecy of none less than
King David.12
During the course of the fifth century, martyr cult and
the shrines of martyrs came to play an important role

180/754

in the frontier arena. The construction of a martyrium

within Rusafa's walls not long before 425, and the
subsequent recording of the Passio, were signs of
Sergius's rising popularity. But Rusafa was not the
only place between the Roman and Iranian empires
where martyr cult was blossoming in the fifth century.
The accounts of the emerging cult at Mayperqat reveal an instructive variety of motives, and of angles
from which Syria and Mesopotamia might be viewed.
The cult of the Iranian martyrs at Mayperqat and the
collection of their passiones may also have provided
special stimulus for the development of the Sergius
cult only a decade later. Out knowledge of the situation at Mayperqat depends primarily on the Vitae of
the fifth-century bishop Marutha, the founder of Martyropolis on the site of Mayperqat. Marutha had also
served as Theodosius II's ambassador to the Sasanian emperor Yazdgard I. His career crystallizes both
the importance of geography in Syro-Mesopotamian
martyr cult, and the relationship between political
maneuvering and martyr cult that would be expressed
also at Rusafa.

181/754

AT THE WORLD'S CENTER.

As the chief city of Sophanene, Mayperqat served as
market and ecclesiastical center for a large and dispersed assemblage of villages whose inhabitants
made a living from agriculture and animal husbandry.13 Situated on undulating ground south of the
Hazro range of hills, Mayperqat overlooks
12. Mal., Chron. 35758.
13. See Fiey, AB 94 (1976): 57, on Mayperqat before
Marutha's appearance; see also Honigmann, Ostgrenze, 7 n. 5.

Page 49
the Tigris basin, which begins to decline some distance to the south of the settlement. Mayperqat (modern Silvan) is watered by the Farkin Su, which in turn
flows into the broad but shallow Nymphios (modern
Batman Su) 20 kilometers to the south. After draining
part of the Taurus mountains, the Nymphios finds a
course through the Hazro range and then debouches
into the Tigris basin some 40 kilometers south of
Mayperqat. The settlement lay at the intersection of
many routes, especially those leading southward to
the Tigris. Mayperqat was a halt on a route between
Bitlis and Amida (modern Diyarbakir), connecting the
Black Sea and adjacent Iranian territory with the SyroMesopotamian plain and the Mediterranean coast.14
Situated in the mountainous northeastern limit of Roman control, the city of Mayperqat, "un lieu d'osmose"
as Fiey aptly called it,15 lay at the convergence of the
Roman empire to the west, the Iranian empire to the
east, Armenia to the north, and Arab lands to the
south.16 Throughout its history, the region has often
been caught up in power struggles, not only because
of its frontier position, but because its rugged terrain
makes it especially difficult to dominate and thus a
haven of resilient tradition.17

183/754

"There is a country district called Sophanene, located

in the East between Armenia and Persia, which used
to be a wasteland and had not received the faith of
Christ," the older of two Lives of Marutha preserved in
Greek begins.18 In the fifth century, Sophanene lay
within the Roman empire, but was often a bone of
contention in Roman-Iranian conflicts.19 It was a region peripheral to Constantinople's calculations, and
in times of persecution, it became, like the desert
fringes of Syria-Mesopotamia further south, a refuge
for anti-Chalcedonian monks. The Vitae sanctorum
ori14. Sinclair, Eastern Turkey, 3: 295; Nogaret, REArm
18 (1984): 41516; Minorsky in EI2, 6: 920; Lewis in
EI2, 1: 1242. The topography of Mayperqat has been
much discussed because of the city's identification by
some scholars with Tigranocerta, a view that now
seems untenable. On the difficulty of interpreting ancient accounts of the region, see Syme in Armies and
Frontiers, 6171, and now Syme, Anatolica, 5865.
For a discussion that argues persuasively in favor of
identifying Arzan as Tigranocerta, see Sinclair,
REArm 25 (199495): 183253.

184/754

15. Fiey, Muson 89 (1976): 38.

16. See Sinclair, Eastern Turkey, 3: 16162, 37677;
Minorsky, EI2, 6: 920. On the diversity of cultural, especially artistic, influences, see Nogaret, REArm 18
(1984): 42124.
17. The advantages of this liminal position were
clearly perceived. For instance, the monastery of Bar
, located hard against the mountains north / northeast of Mayperqat, chose to have its bishops consecrated by the Armenians in order to retain its independence from Antioch: see Fiey, Muson 89 (1976):
3132.
18. Vita Maruthae (BHG 2265). When describing the
idolatrous state of Marutha's native Sophanene, the
Armenian Vita, which is a translation from Syriac, assumes the reader's familiarity with the Mesopotamian
districts and so does not define Sophanene's geographical position.
19. See Mango, T&MByz 9 (1985): 9193; Minorsky
in EI2, 6: 92021; Fiey, Muson 89 (1976): 510; and
Dillemann, Haute Msopotamie, 21617, on the changing political boundaries around Mayperqat.

Page 50
entalium by John of Ephesus is packed with accounts
of trying sojourns in and around Amida and Mayperqat. An episode recorded from the city of Mayperqat
involved the holy man Habib, who attracted the faithful
from east as well as west of the nearby political frontier.20 While from Constantinople Mayperqat appeared
to lie at the eastern periphery of the empire, one could
also see it as the center of the known world, holding a
delicate balance between the two great empires and
the considerable powers of Armenia and the Arabs to
the north and south. Some of our sources for
Marutha's activities saw with the eye of the outsider,
others with that of the insider. Linguistically the
sources, written in Greek, Syriac, Armenian, and Arabic, represent the cultural traditions that surrounded
Mayperqat on all sides.21
The Greek ecclesiastical historian Socrates wrote
around 439, within a generation of Marutha's death
ca. 420, and from the vantage point of Constantinople. Socrates briefly records the bishop's two
imperial embassies to the Iranian court.22 Theophanes, drawing on Socrates and Theodore Lector,
also attributes two embassies to Marutha.23 There

186/754

are also an early, perhaps late fifth-century Greek Vita

and an eleventh-century Vita based on the early
Greek account.24
Our knowledge of Marutha does not, however, depend exclusively on Greek sources. An Armenian Vita
exists, which concludes with a prayer to the martyrs of
Martyropolis for Gagik and his deacon Grigor, who
translated the book from the Syriac, and then continues with a brief account of Kawad I's siege of Mayperqat in 502.25 This could either be an addendum also
translated from the Syriac, or, as is more likely, a later
addition to the Armenian translation. There is no explicit indication of the date of composition or translation,
but the confusion of Yazdgard II with Yazdgard I in the
20. Joh. Eph., Vita sanct., PO 17, pp. 1112. For the
ascetic milieu of the Amida area, see Harvey, Asceticism, 5775.
21. The Vitae and other sources will be discussed individually. The many accounts of Marutha's career are
surveyed by Tisserant, Dict. de thologie, 10: 14249,
and Sauget, Bib.Sanct. 8: 13059. See also Labourt,
Christianisme, 8789, 93; Sako, Rle de la hirarchie
syriaque, 5961.

187/754

22. Soc., HE 7.8, mentions two undated embassies.

See the discussion in Sako, Rle de la hirarchie syriaque, 6061, on the varied ennumeration and dating
of Marutha's visits to the Iranian court proffered by the
sources.
23. Theoph., Chron., A.M. 5906, p. 82, and A.M.
5916, p. 85.
24. Vita Maruthae (BHG 2265 and 2266). Both are
translated with commentary by Noret, AB 91 (1973):
77103.
25. ew
(Vitae et passiones sanctorum)
(Venice, 1874), 2.1732. References below to the Armenian Vita (cited as Vita arm.) are to the English
translation by Marcus, HThR 25 (1932): 5570.

Page 51
account of Kawad's assault suggests that at least the
final section following the translators' names was
composed when such details could already have become clouded (after the sixth century?).26 The lost
Syriac original may have been considerably earlier.27
The historian Ibn al-Azraq, who wrote a history of his
native Mayperqat in ca. 117677, and the geographer
Yaqut, whose Kitab
al-buldan appeared ca. 1222,
derived their accounts of Marutha from a lost Syriac
source, very possibly the same one the Armenians
translated.28 In addition, an Arabic Nestorian ecclesiastical history was written by Mari ibn Sulayman in
Iran in the twelfth century and revised significantly by
ibn Matta in the fifteenth. Together with Bar
Hebraeus's thirteenth-century Chronicon ecclesiasticum, Mari's Kitab al-mijdal and
later revision relate
the main events in Marutha's career, and in greater
detail than the Greek historians.29 The Nestorian accounts belong to the Iranian ecclesiastical world
where Marutha earned his reputation as conciliator.
Noret has argued, on the basis of textual comparison,
that the early Greek Vita is also related to a lost

189/754

Syriac original.30 Onomastic evidence supports this

claim, but the primary evidence consists of parallels
between the Greek and Armenian Vitae, and also
Yaqut, the latter two certainly dependent on an earlier
Syriac source.31 The Greek Vita is very lean, particularly in comparison with the Armenian Vita, which
shows signs of considerable pious adornment. But the
Syriac skeleton is discernible in both accountsin the
narrative order and also in elements such as
Marutha's genealogy, the appearance of Jacob of Nisibis, the role of a woman named Mariam in the
26. Vita arm. 3132, pp. 6869 (tr. Marcus). See
Marcus's comments, HThR 25 (1932): 54.
27. On the lost Syriac account, see Fiey, AB 94
(1976): 35 n. 5; Noret, AB 91 (1973): 96 n. 5.
28. Ibn al-Azraq,
Mayyafariqin, claims to have relied on a manuscript from the Jacobite church at
Mayperqat. On the relationship of Ibn al-Azraq to
Yaqut and the Vita arm., see Fiey, AB 94 (1976):
3545, on which this study must rely, since no edition
of the early part of the history exists. Yaqut,
albuldan 5.23538, draws on Ibn al-Azraq's account.
See also Robinson, JRAS 6 (1996): 2227.

190/754

29. Mari ibn Sulayman, Kitab al-mijdal, fol.

151b153a;
ibn Matta, revision of Kitab al-mijdal,
fol. 151b153a; Bar Hebraeus, Chron. eccles., 3:
4652. Marcus translates the relevant passages of
Mari ibn Sulayman and
ibn Matta in HThR 25
(1932): 5054.
30. Noret, AB 91 (1973): 96103, discusses the supporting evidence, although he does not claim that the
Greek Life is a direct translation from the original Syriac. Fiey, AB 94 (1976): 45, accepts Noret's claims
and concludes that the Vita arm., Ibn al-Azraq and
Yaqut are all translations of an original Syriac that underwent modification over time. The Armenian translation was made, according to Fiey and Noret, at a considerably earlier stage of the Syriac text's elaboration
than was the Arabic.
31. Noret, AB 91 (1973): 97 n. 4, bases his comparisons with Yaqut on Tisserant's comments in Dict. de
thologie, 10: 14249.

Page 52
Christianization of the region of Mayperqat, and
Marutha's successful visits to the Iranian court. The
Armenian Vita and later Arabic accounts view
Marutha's activities from the Syro-Mesopotamian
world between Rome and Iran. Variations between
these sourcesparticularly the Armenian onesand
the Greek accounts are explicable in terms of cultural
and geographical orientation, and are useful guides in
the study of Syro-Mesopotamian martyr cult.

THE ECUMENICAL CITY OF MARTYRS

Marutha's success as imperial ambassador and Christian leader derived from his cultural flexibility, which
he owed in large part to the environment that nurtured
him. He was the only son of the governor of Sophanene, whose mother, Mariam, had been a Christian woman of noble Armenian blood.32 Marutha did
not follow in his father's footsteps, but instead studied
medicine and joined the priesthood. It was the fortunate combination of his medical and theological learning, his native command of the region's languageshe knew at least Greek, Syriac, and Pahlavi33and his increasing involvement with the

192/754

western Church during a period of nervous calm that

made possible his prominence in the shaping of SyroMesopotamian political and religious relations. Bishop
Marutha's earliest certain participation in Church affairs was in 382, when he took part in the anti-Messalian synod held at Side.34 His talent as a physician
no doubt boosted the success of the Roman embassy
to Ctesiphon in 399, when Marutha cured Yazdgard's
son of demonic possession or, according to Socrates
and
ibn Matta, re32. Vita arm. 1819, pp. 5657 (tr. Marcus); cf. BHG
2265.6 where Mariamne, again an Armenian Christian, is Marutha's mother. See Noret, AB 91 (1973):
9799, for a comparison of the genealogies in the Armenian, Greek, and Arabic accounts.
33. See Baumstark, Geschichte, 53, on Marutha's
translation into Syriac of the canons of the Council of
Nicaea; Synodicon orientale, 256, on his translation of
the western bishops' letter from Greek to "Persian";
and Sako, Rle de la hirarchie syriaque, 60. Chronicle of Seert 78, p. 198, calls Maruta a philosopher
versed in Greek, Syriac, and Hebrew.

193/754

34. Photius Bib. 52; Soc., HE 6.15; Soz., HE 8.16;

and Mari, Kitab al-mijdal, p. 31, fol. 152b (tr. Marcus,
52), record Marutha's attendance at the Council of
Constantinople in 381, but Baumstark, Geschichte,
53, with n. 6, points to the absence of Marutha's name
among the signatories, despite claims that this was
Marutha's first appearance at a Church council. Contemporary eastern accounts do not name Marutha's
see: e.g., Syn.or., 1822. 35, 39 (tr. Chabot, 25562,
274, 280). Only the headings of the acts of the synods, which were probably added later, call him bishop
of Mayperqat: e.g., Syn.or., 17 (tr. Chabot, 253). See
also Fiey, Muson 89 (1976): 5. In the Greek authors,
Marutha's title is bishop of Mesopotamia (Soc., HE
6.15 and 7.8). In the fifth century, the bishop of Mesopotamia and of Sophanene were one and the same
(Honigmann, Ostgrenze, 8).

Page 53
lieved the king of kings of a chronic headache.35 But
although cures played their part in the Roman visit,
the embassy's main concern was Yazdgard's accession in 399 and the Romans' desire to reconfirm
peace between the empires and tolerance of Christians under the new shah's rule. All accountsGreek,
Armenian, and Arabicrecognize a good story and
linger on the neutral ground of Marutha's successes at
courthis miraculous cures of the king and his son
and the Christian sage's humiliation of the Zoroastrian
priests. Iranian Christians' increased freedom to worship and to build churches during the reign of
Yazdgard may have resulted from Marutha's efforts at
the bidding of the Roman emperor. But it also owed
much to the Iranian monarch's cultivation of minorities
as a counterbalance to the Iranian nobility and his
wish to avoid conflict with Rome while preoccupied
with internal disaffection.36 Theodosius II sent
Marutha in 408 on a second mission, this time in the
company of another frontier bishop, Acacius of Amida,
most likely in order to proclaim Theodosius's accession to the throne, ensure peaceful relations between
the two empires, and refresh the Roman appeal for
the welfare of Iranian Christians.37

195/754

The Synodicon orientale records Marutha as "the mediator of peace and concord between East and
West."38 During the Synod of Seleucia-Ctesiphon
convened at Epiphany 410, for instance, Marutha revealed the contents of a letter written by the bishops
of Roman Syria and Mesopotamia, in which they state
their decisions concerning the election and authority
of bishops,
35. For Yazdgard's headache, see Soc., HE 7.8;
ibn Matta, Kitab al-mijdal, p. 23 (tr. Marcus, 53). The
accounts of Marutha's successes at Yazdgard's court
vary in chronology and detail. For instance,
Yazdgard's son in Soc., HE 7.8; Theoph. Chron., A.M.
5916, p. 85; Vita arm. 23 (tr. Marcus, 61); and Mari,
Kitab al-mijdal, p. 29, fol. 151b (tr. Marcus, 51), is a
daughter in Vita Maruthae BHG 2265.7 and BHG
2266.7. See Sako, Rle de la hirarchie syriaque 15,
6061 and 63, esp. nn. 14 and 15, on the diverse accounts of the illnesses and the motivation for
Marutha's visit to Yazdgard's court.
36. Yazdgard was criticized by the Iranian nobles for
his cultivation of an alliance with the Lakhmids of Hira.
On this development, see Bosworth, CHIran,

196/754

593612. His non-Christian subjects remembered

Yazdgard as "the sinner": Christensen, Iran, 269; Labourt, Christianisme, 9192. The reasons for RomanIranian conflict in the fifth century are still under debate in modern scholarship. Holum, GRBS 18 (1979):
15372, focuses on the impact of Roman court intrigue on the religious policies of Rome in frontier
politics, at the expense of wider political and defense
considerations. Rubin in Defence, 67795, and MHR
1 (1986): 3362, discusses the motivation behind
Yazdgard's policy toward Rome, but overlooks the
role of religious embassies, including Marutha's.
Blockley, East Roman Policy, 5051, points to the role
of minorities in Yazdgard's macro-politics as a balance to the Iranian nobility and the priestly class of
mobads. The question of Iranian cultivation of cultural
diversity in Syria and Mesopotamia is discussed further in the context of Khusrau II's involvement at
Mayperqat, below, and at Rusafa, pp. 13341 below.
37. On Acacius of Amida, see Mari, Kitab al-mijdal, p.
31, fol. 153a (tr. Marcus, 52).
38. Syn.or. 18 (tr. Chabot, 255).

Page 54
the observance of Church feasts, and the canons of
the Council of Nicaea.39 After brief discussion, the
decisions were accepted by the forty eastern bishops
who, according to Mari ibn Sulayman, "made known
to him (Marutha) that the Westerners were their
brethren and comrades," acknowledging the place of
Iranian Christians in the Church of Christ that is in the
four parts of the world.40 Grasping at once the power
latent in the association of Christians across political
borders, Yazdgard is recorded to have responded to
the conciliatory efforts of Marutha with the claim that
"the East and West shall be one empire under the authority of my rule."41 In the light of the doctrinal and
ecclesiastical interests that the eastern accounts attribute to Marutha on his visits to the Iranian empire, it
is significant that the Greek sources focus exclusively
on Marutha's mission to convert the king, or at least to
dispose him favorably toward Rome and the Christians on Iranian territory. No mention is made of
Marutha's efforts to harmonize the western and eastern Churches, a detail that would have diminished
Rome's claim to be the sole Christian empire at a period of dynamic missionary activity in Iranian territory.42

198/754

On his second visit to Yazdgard's court, Marutha

made a momentous request of the shah.43 With the
expected worldly detachment of a holy man, Marutha
declined all offers of gold and luxuries. His one desire
combines the disarming humility of a spiritual leader
with the political savvy of an inveterate diplomat: he
asked to collect the bones of all Christian martyrs
from Iranian territory and take them home to Mayperqat. Without pause, we are told, Yazdgard granted
this wish.44 When the bishop returned with his relics
to Roman soil and report of his successes reached
the court in Constantinople, Theodosius, deeply impressed, in turn put his resources at the service of
Marutha. Not surprisingly, this offer did not fall on
stony ground. Like many former diplomats who transform retirement into a second phase of influence,
Marutha's next project was to build the village of
Mayperqat into a sacred fortress with the emperor's
blessing and financial
39. Ibid., 1936 (tr. Chabot, 25775). See also Chronicle of Seert 66, p. 206; Mari, Kitab al-mijdal, pp.
3031, fol. 152b (tr. Marcus, 5152);
, Kitab almijdal, pp. 2425 (tr. Marcus, 54); Bar Hebraeus,

199/754

Chron. eccles., 3: 4952; Sako, Rle de la hirarchie

syriaque 6870; Labourt, Christianisme, 9299.
40. Mari, Kitab al-mijdal, pp. 3031, fol. 152b (tr. Marcus, 52). For the number of eastern bishops in attendance, see Syn.or. 19 (tr. Chabot, 257, with n. 1).
41. Syn.or. 19 (tr. Chabot, 256).
42. For fifth-century Christian proselytism in eastern
Syria-Mesopotamia, see Rubin, MHR 1 (1986):
3336.
43. Vita gr. 8; Vita arm. 29 (tr. Marcus, 6667).
44. In the light of the long-standing friction in Iraniancontrolled areas caused by Christian burial practices,
which Zoroastrians considered an impious contamination of the earth, Yazdgard's permission to take the
bones might have been inspired by more than his admiration for the holy man, as Nicholson, AJA 89
(1985): 668, has suggested.

Page 55
support. Between 410 and 420, Mayperqat was fortified with impressive walls and adorned with churches,
at whose altars and in whose walls the bones of the
martyrs were deposited. The city was christened Martyropolis and elevated to the dignity of an episcopal
see.45
From the Syro-Mesopotamian perspective, it was not
sufficient for Martyropolis, as a fortress at the center
of the Christian world, to be protected by the holy relics only of eastern martyrs. According to the Armenian
Vita, Marutha also received permission from
Theodosius to collect "relics from all the saints in his
empire."46 "He went out," the Armenian Vita continues, "to Rome and to all the cities and provinces and
villages and monasteries and hermitages, and
gathered relics." His harvest was impressive120,000 from the Roman empire, 20,000 from
Asorestan (the Armenian name for Beth
, northcentral Syria-Mesopotamia), 80,000 from the Iranian
empire, and 60,000 from Armenia, a total of 280,000
relics. Putting aside the hagiographer's enthusiastic
hyperbole, what remains is a cult truly centered in
Syria-Mesopotamia. Sozomen and Theophanes,

201/754

Greek sources, are silent about Marutha's collection

of relics. The Greek Vita mentions only Iranian Christian bones, emphasizing Iranian persecution of Roman
reli45. One large and two small dark purple marble
reliquaries, now in the Diyarbakir museum, were photographed by Gertrude Bell at Silvan (Mayperqat) in
1906: see Mundell Mango's discussion in Bell,
Churches 129, and Mango, T&MByz 9 (1985): 95,
who is tempted to believe that they held the bones of
Marutha's martyrs. See Fiey, AB 94 (1976): 4243, on
pious constructions in the area of Mayperqat according to Ibn al-Azraq in the twelfth century. See also
Fiey, Muson 89 (1976): 2438, esp. 25, on the transfer of the relics from the Melkite church on the city acropolis to another church in 1012/13 by the Marwanid
Nasr al-Dawla. Local tradition knew the citadel church
as the "fortress," a fitting name in the light its location,
but also because of the relics it housed.
46. Vita arm. 29 (tr. Marcus, 6768). Oddly, the Armenian variation has not been remarked on by any
commentator. The Invention of the Relics of Saint
Bartholomew (BHO 159) also records that Marutha

202/754

assembled 270,000 relics from Iran, Syria and Armenia. Van Esbroeck, REArm 17 (1983): 19091,
does not date the Invention, but conjectures that it
had already undergone elaboration by the early eighth
century. According to Ibn al-Azraq, Marutha also
brought from Constantinople the blood of the Prophet
Joshua in a phial, which was kept in a black marble
reliquary. Ibn al-Azraq,
Mayyafariqin, records that
after curing the king's son, Marutha took advantage of
his favorable position to negotiate peace between the
Iranian king (named Shapur) and Roman emperor
(named Constantine). He requested from the shah
"the bones of the martyrs and the monks killed by
your followers in our country" (summarized by Fiey,
AB 94 [1976]: 42). Fiey supposes that the reference to
"our country" by Marutha, while still in the Iranian empire, implies that the legend recorded by Ibn al-Azraq
was written in Iran (AB 94 (1976): 42 n. 1). Despite its
historical muddle, Ibn al-Azraq's history is closely related to the Armenian Vita, except that the native of
twelfth-century Mayyafariqin has a habit of telescoping his material. For instance, two generations merge
when Marutha becomes both governor and bishop
(Fiey, AB 94 [1976]: 44). According to the Armenian
Vita, Marutha placed two requests for martyrs'

203/754

bonesone to Yazdgard from Iranian territory and

one to Theodosius from Roman territory. It may be
that Ibn al-Azraq here has fused the two into one
request.

Page 56
gion and subsequent Roman recuperation of their
own martyrs' relics to use as protection against
Iran.47 But the accounts closest to home, those preserved in Armenian and Arabic, emphasize the universal Christian intent of Marutha's deeds. Beth
stands distinct in its own right between the empires of
Rome and Iran. By assembling the relics at Martyropolis, Marutha not only did his part to ensure that God
was constantly worshipped at the martyrs' graves;48
according to the Syro-Mesopotamian point of view,
the effect was also to make present at the core of the
Christian world the unity of Christian martyrs and believers across space as well as time.

THE IRANIAN VIEWPOINT

It was in peacetime that Marutha translated the bones
of Roman, Syrian, and Iranian Christians to a fortress
in the mountains above the Syro-Mesopotamian plain.
According to all accounts, the collection of relics was
part and parcel of the building scheme for "Martyropolis," undertaken thanks to the patronage of
Theodosius II, who would certainly have appreciated
the strategic location of Mayperqat and welcomed any

205/754

additional assistance the martyrs could lend.49 But,

as one would expect, the Armenian Vita offers the
view from between Rome and Iran, and reveals the
constant need for balance. Rome's buildup of this
martyr cult at a strategic fortress was bound to elicit
an Iranian response. Yazdgard could not afford to be
left behind at a critical moment. In the Armenian Vita,
he too patronizes Martyropolis with an elaborate gift in
honor of Maruthaan inscribed gold cup filled with
gold to subsidize the building project.50 This benefaction should be seen as an encouragement of
Marutha's efforts to consolidate relations between
western and eastern Christians. The usefulness of
Christianity for Yazdgard's domestic and foreign policy
was obvious, and the Sasanian king could not risk alienating the crucial heartland between the two
empires.
47. In the early sixth century, Severus wrote hymns
commemorating the Iranian martyrs and other hymns
that stress the martyrs' role in the fight against the idolatrous Iranian empire (in James of Edessa, Hymns
of Severus of Antioch 152, 265, 266, pp. 19899,
3012). Not long after Marutha's death in 420,
Theodoret, HR 24.2, attests the inclusion of Iranian

206/754

martyrs among the yearly feasts. His language suggests this was a recent development"Now there are
also martyrs under the same roof who suffered at the
hands of the Persians and who are honored by us in
annual
feasts"
.
48. Vita arm. 26 (tr. Marcus, 6364).
49. On the gradual buildup of Mayperqat as a fortified
frontier post, see Sinclair, Eastern Turkey, 3: 37475,
381, and 42829, on Aqbas, an Iranian fort built-up in
the sixth century across the river Nymphios.
50. Vita arm. 30 (tr. Marcus, 67).

Page 57
The Iranian martyrs' stories traveled with, or at least
close behind, their bones. Accounts of Iranian martyrs
were written at different times and places by different
authors. Many were composed by eyewitnesses from
the Great Persecution under Shapur II and helped to
increase the Iranian martyrs' fame outside Iranian territory. A fourteenth-century Nestorian,
of Nisibis, attributed a collection of Iranian martyr accounts
to Marutha himselfnot a surprising claim, but one
that today seems doubtful.51 However, even if
Marutha played no part in the collection of the martyr
acts, it would have been surprising if, when he returned to Mayperqat with the relics, he did not also
bring their storiesat least in his memory. Marutha's
project to install the relics must have been accompanied by a translatio ceremony where the most spectacular martyr accounts would have been narrated, even
if they had not yet been recorded in writing. Certainly,
Iranian martyr acts were in circulation in the fifth century in the original Syriac, as well as in Greek and Armenian translations.52 From his see in northern
Syria-Mesopotamia, Marutha caught the attention of
political and religious observers on all sides: sowing
peace between Rome and Iran, fortifying his frontier

208/754

city with walls and relics, probably also involving himself in the verbal dissemination of the martyrs' stories.
This milieu where martyrs werenot so paradoxicallylinked with dtente and, at the same time, defense, also witnessed the composition of Sergius's
Passio and the construction of a magnificent church
and other buildings within Rusafa's walls.53
In 502, Kawad I (488531) besieged the city of
Mayperqat and eventually took it. The capture of such
a strategically located fortress was a great boon for
his westward campaign. As a last-ditch effort, the
city's inhabitants filled with gold and offered as their
ransom the gold cup given to Marutha by Yazdgard I,
Kawad's grandfather. According to the Armenian Vita,
which alone records this story, Kawad, in honor of his
forefather's esteem for Marutha, withdrew and left the
region in peace. Other sources relate that
51. Wiessner, Zur Mrtyrerberlieferung, 1139. Labourt, Christianisme, 5253, argued strongly against
Assemani's attribution to Marutha of the collection
Assemani translated in Acta martyrum orientalium.
Nevertheless, Labourt conceded that Marutha's role in
the collection of acta cannot be rejected only on the

209/754

grounds of the late date of

. Chabot, Littrature
syriaque, 4043, and Baumstark, Geschichte, 5357,
did not dismiss Marutha's authorship. Even Peeters
could not reject the possibility of Marutha's involvement absolutely (AB 56 [1938]: 21, esp. n. 3 and 122).
52. Soz., HE 2.14.5; Martyrum Persarum acta. Names
of Iranian martyrs appear as early as 411 in the Syriac
Calendar of Edessa. The manuscript (British Library
MS. Add. 12150) is the "earliest dated Christian literary manuscript in any language" (Brock, in Cambridge
Ancient History, 13: 718, with n. 22).
53. ACOec. 1.4, p. 163, mentions not only a basilica
at Rusafa, but also "illi altissimi muri et alia intra eandem munitionem aedificia."

Page 58
the city indeed soon reverted to Roman possession
after its capture in 502. Later, as part of Justinian's
building program, Martyropolis was refortified and
named Justinianopolis.54
Again at the end of the sixth century, in 589, Mayperqat fell into Iranian hands, only to be returned by
Khusrau II in 591 out of gratitude for the emperor
Maurice's support of his claim to the Iranian throne
against the usurper Bahram.55 From his position of
weakness, Khusrau attempted to maintain a presence
at the frontier settlement by recording in a Greek inscription on the city wall Mayperqat's shared Roman
and Iranian history and his voluntary cession of it to
Roman control.56 Khusrau returned the city to Roman
control, but by emphasizing his generosity at that particular frontier, famous to Christians as the resting
place of Iranian martyrs, Khusrau kept his hand in the
intricate game of frontier politics. His return of
Mayperqat should be seen as a foreshadowing of his
gestures at Rusafa, discussed in chapter 5.
Khusrau's maneuvering was immediately countered.
When, in 591, Mayperqat was returned to Rome, the

211/754

bishop of Melitene, Domitianus, inaugurated a feast to

celebrate the city's salvation and honor the martyrs.57
Theophylact relates that at the martyrs' feast
Domitianus sang a triumphal hymn composed for the
occasion, in which he praised the Iranian martyrs as
champions against Iran: "This, martyrs, is your offering from the Babylonian tyrant and foreigner, the fugitive from his own kingdom who is now obedient to the
Romans rather than hostile: for such great deeds
have you executed against your enemies."58 Book 4
of Theophylact's Historia Universalis closes with this
episode about the Iranian martyrs of Martyropolis. At
the opening of book 5 we encounter Khusrau before
his victory over Bahram, supplicating Sergius, another
martyr positioned on the frontier.59
The various accounts of Marutha's project illustrate
the possibilities of martyr cult in the Syro-Mesopotamian frontier zone. We glimpse in Marutha's career, especially as it is preserved in the Syriac, Armenian, and
Arabic traditions, how that zone became a center from
which the martyrs' power
54. Proc., Aed. 3.2.414; Malalas, Chron. 18, p. 427.
See also van Esbroeck, REArm 17 (1983): 183.

212/754

55. Theoph. Sim., Hist. 3.4.15, 3.5.116.5,

4.12.615.13; Evag., HE 6.1419; Mich. Syr., Chron.
2, pp. 36061; Agap., Kitab
, pp. 44047;
Sebeos, Hist. 9 (tr. Gugerotti, 59).
56. Although the inscription, now lost, was transcribed
in fragmentary condition, blocks 6 and 7 preserve
enough to reveal Khusrau's vaunting rhetoric. Mango,
T&MByz 9 (1985): 99100, republishes the text.
57. Theoph. Sim., Hist. 4.15.1816.28. Garsoan,
REArm 10 (197374), esp. 12022, discusses the
role played by eastern bishops in the face of siege or
attack. Cf. Theophyl., Hist. 3.5.816; Evag., HE
6.1419.
58. Theop. Sim., Hist. 4.16.13 (tr. Whitby and Whitby).
59. Ibid. 5.1.7.

Page 59
radiated, linking Christians whether Roman, Iranian,
or from somewhere in between. It was, after all, during Marutha's lifetime that Symeon ascended his
column and became a "great wonder of the oikoumene," attracting visitors from all directions"Ishmaelites, Persians, Armenians, Iberians, Himyarites,
Spaniards, Britons, Gauls, and Italians."60 The absence from the Greek sources of Marutha's relic-collecting activities west of Iranian territory makes it clear
that Rome's leaders aspired to mold martyr cult as a
bulwark against Iran. Yazdgard and Khusrau II, on the
other hand, through their involvement with Christian
communities in Iranian territory and in the frontier
zone at Mayperqat, played up the potential unity of
Syrian and Mesopotamian martyr cult in their attempts
to weaken a dangerous Roman monopoly. In the sixth
century, leaders on both sides of the frontier did their
best to court the favor of martyr-patrons. It was in this
tug-of-war between the martyr cult's natural impetus
toward cross-border expansion, and even cohesiveness, and the two empires' impetus toward cultural
monopoly that the cult of the martyr Sergius was
forged.

214/754

60. Theod., HR 26.11.

Page 60

THREE
Rusafa
THE FRONTIER ZONE
In about the year 570, the Piacenza Pilgrim traveled
from Antioch across northern Syria and Mesopotamia
to Chalcis, Harran, and Barbalissus, where he visited
the tomb of Bacchus. At Sura, his final stop, he was
told that the neighboring region, wherein lay the
shrine of Sergius, was "Saracen country."1 Judging
by his report that Sergius's tomb was to be found at
Tetrapyrgium, the pilgrim seems to have made do
with Sergius's fellow martyr at Barbalissus.2 Arab
raids posed a threat to travelers, merchants, and settlements in the area of Rusafa long before the late
sixth century. Most raids would have been small-scale
and aimed at the increase of flocks at the expense of
other pastoralists. The vast majority of these would
have passed unrecorded.3 Their impact on individuals
could nonetheless be dramatic

216/754

1. Piacenza Pilgrim, Itin. 47, p. 191, "inter Saracenos"

(tr. Wilkinson, Jerusalem Pilgrims, 89).
2. The Piacenza Pilgrim reported that both Sergius
and Bacchus had been martyred at Sura, and that
Bacchus was then buried at Barbalissus and Sergius,
whom he calls Bacchus's brother, at Tetrapyrgium,
located "in the desert." Wilkinson, Jerusalem Pilgrims,
89 n. 58, 174 (in the good company of Dussaud, Topographie, 254) takes the Pilgrim's report at face
value and erroneously identifies Tetrapyrgium with
Rusafa.
3. Macdonald, Syria 70 (1993): 31314, 328, cautions
against confusing brigands and nomads. Accounts of
nomad Arab raids on the settled population must be
weighed against imperial interest in propagandizing
against the Arabs, who were increasingly unavoidable
participants in frontier politics: see Macdonald, Syria
70 (1993): 32627, with nn. 154, 155; Sartre in
Dsert, 14348, 160. That is not to deny that the
threat of Arab raids existed: see, e.g., CIL 3.128, a
Severan building inscription at Khan al-Qusayr on the
road between Damascus and Emesa, constructed "for
the sake of public security and against the terror of

217/754

Scenitae Arabs"; with Rey-Coquais, JRS 68 (1978):

70 and n. 356. Proc., Aed. 2.9, claims that a mudbrick wall provided sufficient deterrence against Arab
raids, which also suggests small-scale

Page 61
as in the case of a camel-herder discovered in the
mountains southeast of Sinjar by the Nestorian bishop
of Haditha. The bishop approached the figure, whom
he heard reciting a Nestorian hymn of resurrection in
Syriac. At first the camel-herder responded only in Arabic. But when the bishop bound him with oaths, the
man reluctantly explained that in his former occupation as a priest he had been taken hostage forty years
earlier in an Arab raid, while praying for rain with his
congregation outside Damascus.4
In 498, Arabs invaded Syria under the leadership of
Iran's ally, the Lakhmid
. They were defeated at
Bithrapsa, "the first district of Syria," by Eugenius, the
strategos of Syria and Euphratesia.5 The identification
of this site eluded even Alois Musil and Ernst Honigmann. However, Irfan Shahd seems to have solved
the enigma: the Greek
, Bithrapsa, is a corruption of what Shahd calls "the Semitic form . . . Beth
Resapha."6 The shrine of Sergius at Rusafa had become a desirable object of plunder. But there is reason to suspect that the presence of
force near
Rusafa was part of a broader trend. The renowned

219/754

fortified camp, or hirta, of

est of Rusafa at Hira, be-

was located southw-

assaults. Cf. also Teixidor in Frontires, 95103;

Whittaker, Frontiers, 13639, 24546, with n. 4;
Sartre, Trois tudes, 121203, esp. 153203; Rets
in Aspects of Late Antiquity, 3141; and Graf in
L'Arabie prislamique, esp. 352, on traditional interpretations of nomads as brigands. For the modern
period, see Lancaster and Lancaster in PSAS 18
(1988): 5556.
4. Thomas of Marga, Book of Governors 2.41, with n.
3, p. 275. Fiey, Assyrie chrtienne, 1: 108, identifies,
with some hesitance, Maran Zha in Thomas of Marga
with Mar Zaha, who was bishop of Haditha after 740.
See also Magoulias,
48
(199093): 300310, for Arab raids on monasteries.
5. Theoph., Chron., A. M. 5990, p. 141. See PLRE, 2:
417 and Shahd, BAFIC, 122, for the identification of
Eugenius's jurisdiction. Three encounters between
Rome and the Arabs are recorded for the year 498.
Only the defeat at Rusafa is of direct relevance to the
discussion here. For the others, see Shahd, BAFIC,
12131; Sartre, Trois tudes, 15960.

220/754

6. "[T]he second part of the compound . . . has undergone metathesis, which is common in the (Greek)
transliteration of Arabic and other Semitic names"
(Shahd, BAFIC, 123). "Beth" has been added to signify the area around Rusafa. Additional evidence in
support of Shahd's suggestion can be found in the
Syriac Hist. Ahud. 4, p. 29, which refers to Rusafa as
Bet Rsapha. Shahd notes the local parallel of
Bizonovias for "Beth Zenobia" on the middle Euphrates (Musil, Palmyrena, 268). To this we can add
Betproclis for "Beth Forklos" southeast of Emesa
(Grimme, Palmyrae, 21 n. 8). In the thirteenth century,
Yaqut,
al-buldan 4.255 (tr. Le Strange, Palestine,
441) observed the foreign character of the name Betproclis (modern Furqlus). A village on the middle Euphrates refered to in Greek as "Beth Phrouraia" is frequently mentioned in the third-century papyri that appeared on the antiquities market not long before January 1988 and were published by Feissel and Gascou,
CRAI 1989, esp. 54045. For further examples from
the region, see Morony, Iraq 117, with n. 48. Contrary
to what Herzfeld suggests in Sarre and Herzfeld,
Archologische Reise, 1: 136, with n. 3, in Assyrian
cuneiform records of the second half of the ninth century B.C., Rasappa refers not to Syrian Rusafa but to

221/754

the region between the Khabur river in the west, the

Wadi Tartar in the east, the Jabal Sinjar in the north,
and the Euphrates in the south (Bernbeck, Steppe als
Kulturlandschaft, 119).

Page 62
tween the fertile riverine zone and the desert's fringe.
Shapur I and Shapur II developed Hira as a buffer to
control Arab incursions eastward, while Yazdgard I
had cultivated the Sasanian alliance with the Arabs of
Hira as a counterbalance to internal political threats.7
The sixth century saw the apogee of Hira's prosperity.
Its famously salubrious climate and convenient position for trade from the Arabian peninsula attracted a
culturally diverse population.8 Hira was an agglomeration of walled enclosures, towers, gardens, and permanent structures, including churches, surrounded by
camps. It is best understood as a permanent camp at
the steppe's edge; and the interconnection between
the settlement and the steppe is well illustrated by an
incident in the early fifth century. After the dux Timostratus routed Iran-allied Arabs near Callinicum,
Arabs allied with Rome launched a successful raid on
a caravan traveling to Hira. The raiders made off with
the camels, but did not linger at Hira, since the inhabitants had taken refuge in the "inner desert."9 Local
knowledge of the landscape played a critical role in
martial conflict in the region, and was understood by
both Roman and Sasanian leaders as a military
advantage.

223/754

Traditionally, Iran had launched attacks on Rome up

the Tigris or Euphrates, then westward across the
Syro-Mesopotamian plain south of the Taurus Mountains, with the goal of taking Antioch.10 This pattern
had led to Roman fortification of the major settlements
just south of the Taurus foothills. In 498,
followed the Euphrates northward, but instead of continuing as far as the fortified cities below the Taurus,
he kept the river on his right and the steppe on his
left, and then cut northwestward in a direct attempt on
Antioch. The first district of Syria he entered on this
trajectory was that of Rusafa, where he came up
against a Roman commander with a stronger force. It
was not the first, nor would it be the last time that the
steppe approach was taken. The Lakhmid al-Mundhir
III, who ruled 50554, recommended the route to the
Sasanian king Kawad as an alternative to the heavily
defended northern route, and it continued to be used
throughout the sixth century.11 According to
Procopius, al-Mundhir closed his remarks to Kawad
with the following advice: "As for lack of water or any
kind of provision, let no such thought worry you, for I
shall lead the army

224/754

7. Brunner in CHIran, 3: 757, on Shapur I; Eilers in

CHIran, 3: 485, 487, on Shapur II and Bahram V.
8. On the diversity of Hira, see Morony, Iraq, 218,
221; Bosworth in CHIran, 3: 597602; Donner, Conquests, 16773; Musil, Middle Euphrates, 1024, n.
57; Rothstein, Dynastie, 1233.
9. Chron. Edessa of 506 283, p. 208; cf. 29192, p.
215, for another encounter near Callinicum.
10. For the northern Syro-Mesopotamian route, see
Mouterde and Poidebard, Limes, 20.
11. Proc., BP 1.17.2939.

Page 63
wherever it seems most suitable."12 Procopius also
describes the Iranian invasion led by Azarethes,
which avoided the fortified northern route and, on the
Lakhmid al-Mundhir's advice, cut across Syria-Mesopotamia further south, thereby catching the Roman
forces by surprise. Belisarius was stationed in Mesopotamia, and before he was informed of their presence in Roman territory, the Iranians with their Arabs
allies were camped at Gaboulon, east of Chalcis.13
Again, in 540, Khusrau I invaded Syria, traveling from
Circesium to Zenobia / Halabiya and Sura, and thence
to Rusafa.14 Later, Procopius describes how Khusrau
I took the same route in 542, keeping the Euphrates
on his right, and then immediately relates the story of
Khusrau's siege of Rusafa.15 Procopius is not the
only source for this use of the steppe routes. John of
Epiphania records that in 573 the Iranian general
Adarmaanes crossed through the desert from Ambar,
south of Circesium, with Iranian and nomad Arab
forces.16 And Theophylact records that Adarmaanes
made a surprise invasion through the steppe northwestward from Circesium as far as Antioch, taking
Apamea on his way home.17 This approach expanded the arena of conflict between the Romans, the

226/754

Iranians, and their Arab allies to include the steppe

zone between Ctesiphon and Antioch, the region in
which Rusafa lay.
In addition to the route, the timing of Arab attacks at
the end of the fifth century is also significantan incursion into Phoenicia in 491 coincided with the accession of a new Roman emperor, Anastasius; while
in 498, the confrontation near Rusafa may have
marked the Iranian king Kawad's reestablishment on
the throne after a period of internal disruption. One of
the major sticking points in Roman-Iranian relations in
late antiquity was the guarding of the Caspian Gates
undertaken by Iran with Roman subsidy.18 Kawad
made an issue of delinquent Roman payment. But he
had failed in 491 to extract payment from Anastasius.19 It would have been in keeping with traditional
Iranian policy to approach the Roman emperor once
more when Kawad regained the throne in 498, after a
two-year exile. In 491 and 498, the Iran-allied Lakhmids may have offered Kawad a means
12. Ibid. 17.39 (tr. Dewing).
13. Ibid. 18.115.

227/754

14. Ibid. 2.5.133; cf. Proc., Aed. 2.6.111, on

Circesium; 2.8.925, on Zenobia; 2.9.110, on Sura
and Rusafa.
15. Proc., BP 2.20.116.
16. Joh. Epiph., Hist. 1.4 in Mller, FHG 4, p. 275.
17. Theoph. Sim., Hist. 3.10.610.
18. See Rubin in Defence, 68388, on the role of the
Caspian Gates in Roman-Iranian relations in the fifth
through sixth centuries.
19. Chron. Edessa of 506 248, pp. 18384, where
Kawad's gift of an elephant failed to induce payment.
See also Blockley, East Roman Policy, 8890;
Shahd, BAFIC, 12225.

Page 64
of expressing his hostility toward the uncooperative
Anastasius without overtly violating the treaty of 442.
The two Arab incursions took place after Kawad's diplomatic failures and were designed to remind Rome of
the vulnerability of her eastern frontier. Kawad formally broke the treaty when he led a force against
Amida four years later, in 502. Procopius explicitly describes this attack as a response to Anastasius's refusal to pay the Caspian Gates subsidy.20
Anastasius was not slow to react to modifications in
frontier politics. He began to adjust Roman fortification
of the frontier zone, and, in 502, concluded a treaty
with Arabs who had been raiding Roman territory
between 497 and 502. The verdict is still out on the
identity of
, whom Theophanes gives as
Anastasius's partner in the treaty.21 It seems that this
Arethas should be identified either with al-Harith the
, possibly related to the Jafnid dynasty of the
Ghassanid tribe, or al-Harith of the Hujrid dynasty,
commonly referred to as Kinda.22
Major armed conflict at the turn of the fifth century
was still drawn to northern Syria and Mesopotamia.23

229/754

Anastasius's answer to Jovian's disastrous cession of

Nisibis a century and a half earlier was to fortify the
village of Dara in defiance of the treaty of 442 and Iranian protests.24 During con20. Proc., BP 1.7.13. Cf. Chron. Edessa of 506 276,
pp. 203, where, having received the out-of-date news
of Kawad's preparation for an attack on Roman territory, Anastasius sends Rufinus (PLRE, 2: 95457) to
deter Kawad by payment in gold. Rufinus leaves the
gold at Caesarea when he learns of Kawad's progress
into Roman territory.
21. Theoph., Chron., A.M. 5990, p. 141; cf. A.M.
5995, p. 144. The view espoused by Shahd, BASIC,
1: 312, that Anastasius adopted a formal relationship
with the Ghassanids as early as 502 (as opposed to
Justinian in 529, the date universally accepted), remains controversial. The most recent discussion of
this knotty problem is by Robin, CRAI 1996: 665702,
esp. 69799, who brings South Arabian epigraphy into the discussion. For the situation in South Arabia
during the reign of Anastasius, see also Rubin in
Eastern Frontier, esp. 399403.

230/754

22. Various Arab groups, including both longtime

rivals and newcomers like the Jafnid dynasty of the
Banu Ghassan, were at this time jockeying for position in Syria, Mesopotamia, Palestine, and Arabia,
creating a somewhat blurred impression in the
sources and no doubt on the ground. It is likely that in
498 when the Lakhmid
made his foray into
Syria, his timing was influenced by the preoccupation
of rival Arab groups elsewhere. See, e.g., Malalas's
report that the Lakhmid al-Mundhir learned of the
quarrel between the Hujrid phylarch al-Harith and the
Roman dux of Palestine, Diomedes. When al-Harith
fled to the southeast into the desert, al-Mundhir met
and killed him, seizing the opportunity to resolve an
inherited enmity toward the Hujrid dynasty (Mal.,
Chron. 18.16). On Tanukh and Lakhmid migration into
south central Syria-Mesopotamia, see Bowersock,
Roman Arabia, 13233, esp. n. 42, on Tabari's association of the Tanukh with the frontier zone. For further discussion of the Banu Ghassan, see chapter 5.
23. See Whitby in Defence, 71726.
24. Chron. Edessa of 506 300301, p. 228. According
to Proc., BP 1.2.11, the treaty required that neither

231/754

party construct new fortifications near the boundary

between the two empires. See Blockley, East Roman
Policy, 92 n. 46, for the sources; also Whitby in Defence, 73783; Croke and Crow, JRS 73 (1983):
14359.

Page 65
struction, S. Bartholomew came to the emperor in a
dream, offering to protect the city, and in due course
the saint's relics were translated to one of Dara's new
churches.25
The emergent popularity of the arid steppe approach
exerted some pressure further south, where the route
left the Euphrates in the neighborhood of Circesium
and cut into the eastern extreme of Roman territory. In
response, Anastasius turned his attention to Rusafa
and to another saint. As we shall see, it is difficult to
determine from the sources whether Anastasius
played a role in the construction of the impressive
gypsum walls that protected the shrine. But certainly
Anastasius's elevation of Rusafa's ecclesiastical rank
to metropolitan status in the 510s and his translation
of Sergius's thumb to Constantinople asserted Roman
interest in the well-placed fortress in the frontier zone
between the two empires.
By the early sixth century, Rusafa was more than a
stronghold in a frontier region. It was a wealthy shrine
and pilgrimage center "in the so-called Barbarian
Plain." As the name suggests, the region was in the

233/754

hands of Arabs, commonly known in Greek and Syriac literature as barbarians

in Greek, barbaroye in Syriac.26 Procopius describes Rusafa as "in
the Barbarian Plain."27 Theophylact Simocatta uses
the short-hand "Barbarikon," as do the authors of the
Georgian and later Greek recensions of the original
Syriac Vita of the Iranian martyr Gulanducht, who visited the shrine from northern Mesopotamia in the sixth
century.28 The name suggests a Roman understanding that Rusafa was set in a region distinct from
25. Joh. Diakrin., fr. 2.558, p. 157, with the editor's
note on l. 9. Anastasius's younger contemporary John
Lydus, De mag. pop. rom. 47, also emphasizes the
role of divine intervention with reference to Dara's
place in frontier defense: "[U]nless God by the
former's [Anastasius's] hand had heavily fortified it at
the throats of the Persians, long ago the Persians
would have seized the domains of the Romans inasmuch as these are adjacent to them" (tr. Bandy).
26. Kugener, OC 7 (1907): 40812, judiciously corrects Duval's emendation of the Syriac in Severus's
homily on S. Sergius, Hom. cath. 57, p. 93. Duval
prefers to read "men," "population" (bnaynosha),

234/754

instead of "barbarians" (barbaroye). Kugener leaves

the text as it stands, reading "barbarians." Late antique Syriac and Greek texts that refer to the Arabs of
this region as "barbarians" include Tab. Peut. 11
("fines exercitus syriaticae et commertium barbarorum"); Hist. Ahud. esp. 3, pp. 2425; Evag., HE, 5.6 (
); and cf. 3.2, 5.9, 5.20 Steph. Byz., Ethnica s.v.
, p. 158, also attests the use of barbaros
to refer to Arabs. For further comment on the use of
"barbarian" in Syriac with reference to Arabs, see
Shahd, Martyrs of Najrn, 9899.
27. Proc., BP 2.5.29:
us.,
Contra
monophys.,
.

; cf. Leont. JerPG

86.1900A:

28. Theoph. Sim., Hist. 5.13.3, reports that the votive

gifts of Khusrau II to S. Sergius at Rusafa were taken
to the "so-called Barbarikon" (
), meaning, no doubt, the site of Rusafa in the "Barbarian
Plain," pace de Boor's edition of Theophylact, index
321, where it is suggested that the reference is to a
part of the shrine. Whitby and Whitby, History of
Theophylact, 150 n. 66, follow de Boer; while Schreiner, Theophylact, 316 n. 755, goes so far as to suggest

235/754

that the "barbarikon" was a building where the gifts

dedicated to the martyr by

Page 66
more hellenized territory further west. The first part of
the name establishes a distinction between Greek and
barbarian and qualifies the second partthe plain.
This second part raises important questions about the
frontier context of Rusafa. Rusafa does indeed dominate a plain, a wide expanse of stony, monotonous
steppeland that continues to the Euphrates and beyond into Iranian territory. But a plain is a poor frontier.
A plain is exposed, indefensible, subject to uncontrollable influences. This in part explains why Rome and
Iran were constantly fighting over the areathe SyroMesopotamian plain, what we might call the greater
Barbarian Plain, was not a natural frontier. Situated as
it was in this plain, the river Euphrates could not act
as a serious frontier. But the plain was, in practice, a
frontier zone, and it was given what cohesion it had by
the Arab pastoralists who inhabited both Roman and
Iranian territory in the region.
The name Barbarian Plain implies recognition that the
population of the area was significantly Arab in composition, a fact well illustrated by three incidents spanning over a century. In 504, a Roman officer named
Constantine, who two years earlier had joined

237/754

Kawad's forces as a general, returned to Roman Syria

taking with him two wives and a small retinue. They
traveled through the desert night and day for a fortnight without meeting a soul, until finally they encountered some Arabs allied with Rome, who took
them to the nearby castrum at Sura.29 When not
shepherding wayward generals, Rome's allied Arab
tribes in the region might occupy themselves with the
more mundane business of nomadism. This is clear
from the so-called Strata dispute between the Ghassanids and Lakhmids, which probably took place in
539, and nearly drew their imperial allies into armed
confrontation.30 Several particularly dry seasons had
forced Lakhmid Arabs beyond their traditional grazing
territory into an area controlled by the Ghassanids.
The name Strata, by which the disputed region was
commonly known, eventually helped to resolve the
conflict. The name derived from the Roman road that
traversed the grazing lands and linked Damascus, via
Palmyra and Rusafa, to the Euphrates. That the entire
region through which the Roman road passed could
absorb the road's Latin name is strong testimony, not
only to the usefulness of artificial markers in the barren steppe, but also to the power of such foreign intrusions over the imagination of the indigenous

238/754

population. The tenacity of the name Strata is particularly striking in the light of the Arabic origin of much of
the region's toponymy, which is attested already from
the fourth century.31 The Roman

the Arab tribes were stored.

20.2122:
the Georgian recension, XV.5.1.

Vita

Gulanducht
; and cf.

29. Chron. Edessa of 506 29798, p. 219. See PLRE,

2: 31314.
30. Proc., BP 2.115. For a recent account of the
sources, see Shahd, BASIC, 1: 20918.
31. Bauzou, Syria 70 (1993): 48.

Page 67
word strata, meaning clearly marked route, was eventually transformed into the
word sirat, meaning
the way, more specifically, the straight way, the right
way.32 Early Arabic poetry acknowledges this suggestiveness of man-made presences in the desert, as,
for example, in the comparison of a favorite camel
with a fine Roman bridge, or of the outlines of abandoned camps with the mysterious shapes of Arabic
letters.33
An incident that occurred about the year 630 again
confirms Arab control of the Rusafa region. At that
time a monk was traveling from Jerusalem to Dastagird in search of his master's bones. The monk had
come from the monastery near Jerusalem that had
been the spiritual home of the Iranian refugee goldsmith Magoundat, who became a monk and was renamed Anastasius. Once he had stealthily acquired
the relics of Anastasius, the monk was guided across
the desert as far as Palmyra by an unnamed Arab
phylarch.34 It was at very much the same time and
with regard to the same region that Theophylact commented on the enthusiasm of the Arab nomad tribes
for Rusafa's martyr.35

240/754

A PLACE OF CONVERGENCE
The shrine at Rusafa had the potential to provide
Christian emperors with a means of manifesting the
Roman presence in a culturally mixed frontier
zonea safety device that had been wanting from
Rome's relationship with Palmyra in the third century.
At Rusafa, diverse interests converged: it was a watering point for long-range transhumants, regional
pastoralists, and caravans.36 It also lay in the southeastern extension of an area of marginal agricultural
land that in the fifth and sixth centuries enjoyed tremendous economic prosperity.37 And a dense network of well-worn routes and tracks linked the
steppe's watering points.38 The confluence at Rusafa
of these various factors helps to explain why it had
been the site of a Roman castrum since the tetrarchic
period.
In the sixth century, the Arabs' role in Roman-Iranian
frontier relations
32. Ibid., 36.

241/754

33. On the bridge, see Tarafa,

, p. 84 (tr. Arberry), and Caussin de Perceval, Essai sur l'histoire
des Arabes, 2: 354. For the letters, a common image
in classical Arabic poetry, see al-Isfahani, Kitab alaghani 2.95 l. 5; and Montgomery, JSS 40 (1995):
283316.
34. Translat. reliq. Anast. 5.
35. Theoph. Sim., Hist. 5.1.7.
36. Herzfeld compares Rusafa with Hatra: Sarre and
Herzfeld, Archologische Reise, 1: 136; see also
Pigulewskaya, Byzance, 193; Charles, Christianisme,
33.
37. See Tate, Campagnes, 34350.
38. Bauzou in Archologie, 205, 213, 216. As we
should expect in a context where beasts of burden
provided the most suitable form of transportation,
most routes were roughly paved or unpaved.

Page 68
was intricately meshed with their place in pastoralism,
agriculture, and trade in the region. In the treaty of
561, the third clause reasserts the exclusive rights
over Roman-Iranian trade of the established customs
posts: from north to south, at Artaxata, Nisibis, and
Callinicum.39 Clause 5 also treats customs posts, but
the specific purpose of this clause is to control Arab
merchants in northern Syria and Mesopotamia,
namely, at Nisibis and Dara.40 From Callinicum, trade
in central Syria and Mesopotamia was directed westward through Rusafa via the steppe caravan routes.
Excavation of rows of shops along a major street at
Rusafa has helped to fill in our picture of this important trade center in the Syro-Mesopotamian steppe.41
Rusafa's commercial prosperity also played its part in
the formation of the famous Nestorian holy man Rabban bar
(d. 611). The son of wealthy Christian
merchants in Rusafa, the young Rabban bar
came under the care of his half-sister after his parents
died and left the children an enviable inheritance,
making them obvious targets for Rusafa's
opportunists.

243/754

Certain men among her kinsfolk were looking to take the chaste woman
to wife,
One for the sake of her great beauty, and another for her riches;
But when she learned this she left her country altogether.
Having sold everything which their parents had left them,
She distributed [the money] among the poor, and the widows, and the
monasteries, and the churches.42
The two orphans later settled in Nisibis. The report of
the physician Ibn Butlan, who visited the city in the
mid eleventh century, confirms the tenacity of
Rusafa's cultural and mercantile dynamism, which
continued into the mid thirteenth century: "[T]he inhabitants of this fortress are steppe folk [badiya], most of
them Christians, who earn their living by guarding the
caravans and bringing in goods and merchandise.
There are also vagabonds, as well as thieves."43

244/754

The few surviving images of the rider saint are potent

reminders that, while Sergius's miracles and healing
powers made him a desirable patron,
39. Codex Justinianus 4.63.4. (A.D. 408 or 409); Men.
Prot., fr. 6.1 (Blockley), fr. 11 (Mller).
40. Blockley, History of Menander, 71 n. 51; 73 n. 54.
41. Kollwitz in Neue Deutsche Ausgrabungen, 5153.
In the Islamic period, at least, shops encroached on
the street.
42. John the Persian, History of Rabban bar
pp. 16769 (tr. Budge).

1.115,

43. Yaqut,
al-buldan 3.48 (tr. Conrad in Jabbur,
Bedouins, 466. I translate badiya as "steppe folk," instead of Conrad's "desert folk"). For sources on
Rusafa after the Islamic conquest, see KellnerHeinkele in Sack, Resafa, 4: 13354.

Page 69
his great influence in the sixth century owed much to
his shrine's location at a watered crossroads in central
Syria-Mesopotamia.44 With the control of Rusafa
came, at least in theory, the possibility of monitoring
movement across central Syria and Mesopotamiathe movement of armies, raiding parties, caravans, pastoralists, and pilgrims. Nearly a millennium
before Sergius was executed outside Rusafa's walls,
according to tradition, the settlement had held a
pivotal position on the political and economic map of
Syria-Mesopotamia.45 The topography of the middle
Euphrates region funneled caravan traffic past the site
of Rusafa. At Barbalissus, the Euphrates quits its meandering northsouth course and turns to flow more
consistently southeastward to the Persian Gulf.46
From near Barbalissus as far south as Dayr al-Zawr,
the steppe meets the river in high cliffs.47 The Euphrates is forced into a narrow gorge by Jabal Bishri,
and the river continues past steep banks on the west,
which make progress difficult. The Roman and Iranian
riverine routes served armies whose supplies traveled
by river; but caravan traffic kept a distance from the
Euphrates,48 parallel, but to the west, along the
steppe's watering pointsboth for convenience, since

246/754

the river banks would be frequently soft on account of

rain and flooding between November and June, and
to avoid the charges with which the settled population
often taxed caravans, significantly diminishing their
profits.49
44. See Whittaker, Frontiers, 136, and Matthews, JRS
74 (1984): 163, who observes, with regard to the region east of Palmyra in the third century, that "in this
sector the whole concept of territory and its definition
and control should be formulated rather differently. It
consisted in command of water-points and their associated settlements in an area in which no permanent
control or occupation was envisaged."
45. Honigmann in RE, 2A: 1684; Musil, Palmyrena,
262; Musil, Middle Euphrates, 210.
46. The importance of Barbalissus / Balis continued
and is underlined in the mid to late tenth century by
the merchant-geographers Istakhri and Ibn Hawqal,
who describe the prosperity enjoyed in the region before the East Roman raids in the early tenth century.
Istakhri (Kitab al-masalik 62, tr. Le Strange, Palestine,
417) writes: "[Balis] is the first Syrian town you come
to from Iraq and the road to it is much frequented and

247/754

from Balis go many highways. It is, as it were, a port

to the Syrians on the Euphrates. However, since the
days of Sayf al-Dawla, its buildings have gone to ruin
and caravans and merchants go there much less than
of old. The city has strong walls and gardens in the
lands lying about it and the Euphrates. Its chief crops
are wheat and barley."
47. Musil, Palmyrena, 26061. Maps 3 and 4 in Chesney, Expedition, well illustrate the relationship
between the cliffs and the Euphrates.
48. See Gawlikowski, Iraq 56 (1994): 3031, on the
use of the Euphrates for trade as well as for transportation of military supplies; Bauzou in Archologie, 205,
on vestiges of the Roman riverine route; Mouterde
and Poidebard, Limes, 12, on the Roman routes, and
12736, on the traces of the Iranian Royal Road along
the Euphrates; Poidebard, Trace de Rome, 89, and
Musil, Middle Euphrates, 227, on the Iranian Royal
Road.
49. See Musil, Palmyrena, 261, on caravans' avoidance of the riverine route; Grant, Syrian Desert,
4243, 199, also 14856, with map opposite 258. And
see Mouterde and Poidebard,

Page 70
The caravan route from Babylonia followed the edge
of the steppe on either side of the Euphrates, and
then one branch found its passage at the level of
Circesium through the well-watered saddle between
the western spur of Jabal Bishri and the AntiLebanon's furthest outlier, Abu Rujmayn.50 North of
the pass and 30 kilometers south of the Euphrates lay
Rusafa, at the intersection of the following routes: (1)
the steppe caravan route, which cut inland at Circesium, near where the Khabur meets the Euphrates, or
further north at Callinicum, near the intersection of the
Balikh and the Euphrates; (2) the major route between
the Black Sea and the Gulf of
, running along the
Euphrates as far as Sura, where it linked up with the
Strata Diocletiana via Rusafa to Damascus, and then
carried on through the Hawran plain to Bostra to join
what we call the Via Nova Traiana to
;51 (3) the
route to Damascus that runs northwest of the Jabal
Rawaq, parallel to the Strata Diocletiana;52 (4) the
Euphrates-Emesa route that passes north of the Jabal
hills;53 and (5) the arid steppe route, or more
likely routes, between Ctesiphon and Antioch, which
cut into Roman territory around Circesium and

249/754

thereafter would have been variable according to season.54

Despite the topographical features that contributed to
Rusafa's importance, it must be emphasized that
Rusafa is not, and never was, an oasis.55 It stands
apart from the natural, spring-fed oasesal-Kawm,
Tayyiba, Sukhna, Arak, Palmyra, Qaryataynthat
cluster around the mountain chain running northeast
from the Anti-Lebanon to the Euphrates north of Dayr
al-Zawr. In this sense, Rusafa is exceptional among
the settlements in the

Limes, 12732, on the caravan route that diverges

from the riverine route north of Barbalissus and
passes through Rusafa. Cf. Brunner in CHIran, 3:
759, on the arid route west of the Tigris controlled by
Arabs, which branched off from Takrit as far as Sinjar,
offering an alternative to the populated riverine route.
On the dangers of caravan trade, including the imposition of tariffs, see Lukonin in CHIran, 3: 74041.
50. On recent late antique finds at Tayyiba, the wellwatered settlement in the saddle between Jabal Bishri

250/754

and Abu Rujmayn, see Taha, Cahiers de l'Euphrate

56 (1991): 61, 65.
51. Mouterde and Poidebard, Limes, 127, 132, 213;
Tabula Peutingeriana 10 and 11, and Miller, Itineraria
romana, 815. See also Bauzou, Syria (1993): 2736;
Bauzou in Archologie, 21213; Dussaud, Topographie, 25155; Poidebard, Trace de Rome, 7384.
Archeological finds at Tal Ramad, near Qatara, indicate the existence of trade links between Armenia and
Cappadocia and Palestine by way of Mesopotamia
and the Rusafa-Sukhna route as early as the sixth
millennium B.C. (Contenson, AAS 19 [1969]: 2930).
52. Bauzou in Archologie, 212; Musil, Palmyrena,
24950.
53. Dussaud, Pntration, 8285; Mouterde and
Poidebard, Limes, 137, 14346; Herzfeld in Sarre and
Herzfeld, Archologische Reise, 1: 137.
54. Dussaud, Pntration, 8687, and map on 82. In
the context of the development of the semi-desert
route during the sixth century, note that Justinian established a dux at Circesium (Proc., Aed. 2.6.9).

251/754

55. Pace Gregory, Roman Military Architecture, 2:

174.

Page 71
Syrian steppe.56 Only the establishment and growth
of the cult of S. Sergius at Rusafa can explain the size
and sophistication of the settlement, which would otherwise have continued as a modest watering point.
The site of Rusafa dominates the expansive plain
west of the middle Euphrates at the point where the
land that rises gently from the river valley begins to
undulate.57 In both Syriac (Rsapha) and Arabic
(Rusafa), the settlement's name suggests compactness, often meaning "pavement," presumably referring to the site's geological configuration, which
makes it a natural gathering place for waters.58 Although it possesses neither spring nor well, the site
lies at the convergence of wadis, most notably the
Wadi al-Sila, and is watered by seasonal pools, which
gather in depressions to the west after particularly
heavy rains.59 Rainwater can be stored for long periods in dolines, cavelike hollows in the ground that lie
northeast, east, and south of the settlement and also
run beneath it. Numerous cisterns within Rusafa's
walls also preserved rainwater.60 Small bottle-shaped
cisterns sufficed from the fourth and fifth centuries

253/754

until the city's cavernous cisterns were built in the

sixth century.
56. Wirth, Syrien, 442.
57. Approached from the west, Rusafa in its ruined
state is visible from two hours away by camel (Musil,
Palmyrena, 15455). This assumes a clear field of
vision. In the morning of 26 December 1992, after a
driving snowstorm the previous day, mist obscured
the view on my approach from the pass between
Jabal Bishri and Abu Rujmayn, almost until I arrived at
Rusafa.
58. On the meaning of "Rusafa" in Arabic, see Haase
in EI2, 8: 629b, and Stetkeyvych, Zephyrs of Najd,
109: "On the one hand it [the word 'al-Rusafa'] implies
the compactness and firmness of stone; but that stone
is necessarily associated with water,
being
'rock-water.' Furthermore, it brings to mind the idea of
the building of a pier or embankment. As such it carries imagist sensibility to the very edge of that which is
its semantic opposite. From there, too, it implies the
anticipation and the longing for the other bank of the
river.
, in brief, was to the poet a name and a
word that implied semantic tension and yearning for

254/754

its opposite." I would like to thank Sebastian Brock for

commenting on the range of meanings in Syriac.
59. Musil, Palmyrena, 69. Ulbert in Ruprechtsberger,
Syrien, 113, ill. 2; Grant, Syrian Desert, 160: the top
photograph shows a caravan at a water hole between
Sukhna and Rusafa.
60. On the cisterns and water supply, see Brinker,
DaM 5 (1991): 11946. For pools and dolines, see
Musil, Palmyrena, 69, 169, 172, 260. Musil recounts
how on his second visit to Rusafa, in late March and
early April 1912, his party was forced to leave the site
for the Euphrates after only two days owing to the inadequate water supply, even for camels (ibid., 65,
167). Heavy snows pose a serious threat to flocks and
shepherds, while the ensuing spring flash floods that
swell the region's wadis could easily overwhelm shepherds driven to seek shelter there from the harsh
northern winds. After the winter of 1912, when snow
had covered the region from Rusafa to Jabal Bishri,
Musil found the mountain landscape littered in early
April with the rotting sheep of shepherds who had
failed to cross to the Euphrates before the snows in
early January (ibid., 176, and cf. 153, where a heavy

255/754

snowfall on the route west of Rusafa froze another

flock of sheep in late February). On the perils of flooding, see ibid., 176.

Page 72
Aerial photographs of Rusafa show traces of gardens
with enclosures, especially to the south of the site,
and built basins and barrages to the west.61 South of
Rusafa the plain continues as far as the limestone
Jabal Bishri, whose ravines and gullies also provide
pasture for camels and sheep after rain.62 Arabic
writers frequently describe the surroundings of Rusafa
as badiya, often mistranslated as "desert."63 The Arabic term describes, not the aridity of a desert, but
rather the wide open expansiveness characteristic of
the Syrian steppe. The presence of vegetation, as in
all steppe and even desert landscapes, depends upon
seasonal rainfall, as the following passage referring to
the Umayyad Caliph Hisham, most likely outside
Rusafa, vividly illustrates: "[Hisham] had gone out with
his family, his retinue, his servants, and his companions. And he camped on a barren, stony plain that
was elevated and extensive, in a year when the rain
had fallen early and abundantly and the land had
taken its adornment from the variety of the colors of
its vegetation of pretty spring blooms, which were
beautiful to behold, and luxuriant, and well watered on
a plateau whose soil seemed like pieces of camphor."64

257/754

With the help of collected rainwater, the area around

the walled settlement of Rusafa could conceivably
have supported orchards, olive and figtrees, and even
grain, but no evidence survives to bear witness to
such industry.65 Seasonal migrations intersected at
Rusafa, where the underground reservoirs attracted
the region's pastoral Arabs.66 Fields further to the
west can be cultivated only at high risk. For example,
those around the al-Abjaz range north of Palmyra
flourish or parch depending on the rainfall of March
and April.67 Good pastureland continues from the
plain of Rusafa
61. Tchalenko, glises syriennes, 203; Poidebard and
Mouterde, AB (1949): 110. In 1912, Musil found "old
deserted gardens" and a dam southeast of Rusafa
(Palmyrena, 169). See now Ulbert, DaM 7 (1993):
226, with fig. 1.
62. See Musil, Palmyrena 67, 69, and 1734 on Jabal
Bishri; Mouterde and Poidebard, Limes, 134, on Jabal
Bishri's pasture.
63. For the Arabic sources on the region around
Rusafa, see Musil, Palmyrena, 27072. Conrad in
Quest for Understanding, 27577, comments on the

258/754

variety of meanings available for words such as

badiya, bariya, and sahra. Groom, Dictionary of Arabic Topography, is a fascinating witness to the subtle
descriptiveness of the Arabic language of landscape.
See also Grant, Syrian Desert, 11, on the badiya.
64. al-Isfahani, Kitab al-aghani 2.129.
65. Musil, Palmyrena, 260.
66. The Taghlib tribe, which remained Christian longer
than other Arab tribes, was attacked in 692 while
camping at the watering place al-Rahub, 36 km southeast of Rusafa (al-Isfahani, Kitab al-aghani 12.236). In
the thirteenth century, Yaqut records that Taghlib territory included the ridge of Jabal Bishri from the village
of Tayyiba (controlling the pass between Abu
Rujmayn and Jabal Bishri) as far as the Euphrates (
al-buldan 1.426). Early in this century, the Jijar
tribe spent the winter at Rusafa and the summer north
of Aleppo (von Oppenheim, Beduinen, 1: 303).
67. Musil, Palmyrena, 147.

Page 73
west along the northern foothills of the Abu Rujmayn
and al-Abjaz ranges as far as the fields east of
Emesa. This zone is punctuated by wells, many of
them perennial.68 The moisture level in this zone of
pasture between Emesa and Rusafa normally remains sufficient to sustain grasses into the autumn.69
The belt of pasture that the shepherds exploited
between these foothills and the stony plain of alMatyaha to the north is the same followed by the
Euphrates-Emesa road.70 Emesa lay at the juncture
of the route from the Euphrates and the major road
that followed the Orontes from Antioch to Damascus,
the Hawran plain, Bostra, and on south to
. Especially along the Euphrates-Emesa route, and the wide
band of pasture along its course, traders and breeders, merchants, soldiers, farmers, and part- and fulltime pastoralists would have met at the watering
places and settlements that articulate the route.71
This east-west route in turn links up with the
northsouth axis of agricultural villages that flourished
in late antiquity on the steppe's fringe northeast of
Emesa and stretches from Salaminias / Salamiya toward Apamea and as far north as Chalcis.72 Several

260/754

larger villages, such as Seriane / Isriya, Salaminias,

and, further north, Anasartha / Khunasira would have
served as local markets73 in this L-shaped zone,
which extended in a broad swathe westward to
Emesa and northward to Chalcis. The conflux of
peoples in this region combined with direct access to
the Euphrates-Emesa road to make it a natural choice
for the location of several Roman military installations
positioned to monitor the region. The Notitia dignitatum places the Equites Saraceni Indigenae at Betproclis / Furqlus, 40 kilometers southeast of Emesa;
the Equites Promoti Illyriciani at Occariba /
,
east of Emesa, where
68. One such watering place was reported at
between Anasartha / Khunasira, on the steppe's
fringe, and Seriane / Isriya (Yaqut,
al-buldan
3.137, cited by Musil, Palmyrena, 212 n. 62); Le
Strange, Palestine, 555. In the Roman period, before
the third-century decline of Palmyra, hamlets existed
in the hills northwest of the caravan city thanks to the
collection of rainwater in cisterns (Schlumberger,
Palmyrne, 13132).

261/754

69. Musil, Palmyrena, 149, 154. Musil was told that

the region's grasses also draw gazelle from the inner
desert in the summer.
70. Musil transcribes the toponym as al-Mutayi: see
his map in Northern Arabia.
71. Along this road in late March 1912, Musil met
sheep traders leading large flocks of sheep from the
middle Euphrates, where they were raised, to the
markets in Aleppo and Hama (Palmyrena, 15253).
Dussaud, Topographie, 262, describes this band of
pasture. On the interdependence of pastoralists and
farmers, see Marx in Nomadic Alternative, 4174,
esp. 6871.
72. In late March 1912, Musil met the Hadidiyin tribe,
which cultivated land in the environs of the villages
northeast of Hama, pasturing their flocks at the foot of
Jabal Abu Rujmayn near the western limit of the plain
of Rusafa (Palmyrena, 151).
73. On the geographical setting and water supply of
Seriane, see Gogrfe, DaM 7 (1993): 4648, and
Dussaud, Pntration, 84; on the road between
Salaminias and Seriane, see Mouterde and

262/754

Poidebard, Limes, 14043; Honigmann in RE 4A:

172324; Musil, Palmyrena, 58. On Anasartha, see
Musil, Palmyrena, 204, with n. 58. Ibn Shaddad,
22 (tr. Edd-Terrasse), describes Khunasira's position
as on the edge of the bariya.

Page 74
the
outliers meet the plain; the Equites Scutarii Illyriciani at Seriane at the northern extremity of Abu
Rujmayn, approximately halfway between Emesa and
Rusafa; and the Equites Promoti Indigenae at
Rusafa.74
The regions along the routes linking the triangle of
Rusafa, Emesa, and Damascus were at the focus of
Roman military interest in late antique central Syria. It
is significant that the arms of this triangle span the
mountainous limits of the Palmyrene plain. After
Queen Zenobia's fall, Palmyra continued to function
as a town, although its place as a center for trade was
vastly diminished.75 In the late fifth century, Alexander Acoemetes and his band of monks were refused
entry into Palmyra, on the grounds that such a ravenous hoard would lead the town's inhabitants to starvation.76 Allowing for dramatic exaggeration, the story
reflects the vulnerability of resources in cities set in
the arid steppe when they are not fed by a booming
caravan trade. Thomas Bauzou has convincingly argued that Palmyra and the surrounding plain were
abandoned at some point in the early Islamic period,
since none of the modern toponyms, with two

264/754

exceptions, bear any relation to the ancient names.77

This disjunction between ancient and modern toponyms in the Palmyra region contrasts with the situation in the Hawran, south of Damascus, where a
considerable number of ancient toponyms are recognizable in modern place-names.78 Bauzou ascribes
responsibility for the tenacity of Hawranian toponyms
to the region's Christian population, which has dwelt in
the area continuously since late antiquity.
74. Not. dig. or. 32, for Betproclis / Furqlus, under the
jurisdiction of the dux Phoenicis, see above; Not. dig.
or. 33, for Occariba /
(cf.
, Ptolemy, Geography 5. 14. 13; Occaraba, Tabula Peutingeriana 10;
Orarabon at Raven. Geogr., Cosmogr. 2. 15); and for
Seriane / Isriya, see Musil, Palmyrena, 253. On these
routes in the Arabic sources, see Musil, Palmyrena,
24951, 276; Musil, Middle Euphrates, 318. On the
difficulty of interpreting the terminology that may refer
to Arab members of Roman forces, see Shahd,
BAFIC, 124, 46264, 46667.
75. For evidence of the continuation of urban life after
Zenobia's fall, see building inscriptions dated to 279

265/754

and 280 in
(198687): 168.

and Gawlikowski, AAS 3637

76. Vita Alex. 35. Not much in the generalized description (3235) of Alexander's wanderings in the
Syro-Mesopotamian steppe deserves to be taken at
face value. It does offer an interesting collection of impressions and stereotypes formed in the mind of a
disoriented alien in the steppe. Castra, their inhabitants, their defenders, and their gates stand out in
sharp relief against the otherwise shapeless landscape. The predictable appearance of a caravan with
its cameleers who keep Alexander and his monks
from starving outside Palmyra's walls may be invention. But not necessarily.
77. Bauzou, Syria 70 (1993): 4850. This interpretation is supported by the abandonment in the Middle
Ages of the Damascus-Palmyra segment of the Strata
Diocletiana (Musil, Palmyrena, 249).
78. For definitions of the Hawran, see Macdonald,
Syria 70 (1993): 309 n. 39. I have used the broader
definition, which includes not only ancient Auranitis
but also Ituraea, Gaulanitis, Batanea, and Trachonitis,
that is, from south of Damascus to the Yarmuk, from

266/754

Hermon and the river Jordan to the the basalt desert,

the harra, east of Jabal Hawran.

Page 75
Although Bauzou does not rally evidence from north
of the Palmyra region to support his interpretation of
Palmyra's history, evidence from the area between
Salaminias and Chalcis complements that from the
Hawran. In this region too, many ancient toponyms
have survived the wear of time. And like the Hawran,
the area was home to many Christians until recent
years. Both the Hawran and the area between
Salaminias and Chalcis were areas of convergence
for agriculturalists, semi-pastoralists, pastoralists, and
traders, and in both the Sergius cult flourished, as will
be illustrated later.79
Traces of the Hawranian road system, complemented
by a mosaic from the southwestern Hawran, illustrate
the mixed agricultural and pastoral nature of the region and help to reknit the less-documented and lessstudied area between Salaminias and Chalcis. The
primary crops of the Hawran were cereals, olives, and
the vine. Villages and larger towns were connected by
a dense network of built roads and dirt tracks.80
Damascus was linked to Bostra by two important
paved Roman roads running to the east of the Laja,
which would allow for swift movement of troops and

268/754

supplies. Although much of the course of a third route

to the west of the Laja cannot be followed today, it is
almost certain that in late antiquity a Roman road followed that route too, which is now the main passage
between Damascus and
. Another Roman road
with two bridges is partially preserved between
and Bostra and then carries on to Mushannaf on the
eastern flank of the Jabal Hawran. The visible sections of pavement, even those within the boundaries
of Bostra and Maximianopolis / Shaqqa, show no
signs of grooves from wheeled transport. In general,
there is no evidence that wheeled vehicles were a
common feature of Hawranian transportation, particularly in the case of village to village communication,
which would have been along dirt tracks. Animals
would have been used to transport goods, and a mosaic now in the Bostra museum shows a bearded man
clad in a short, sleeveless tunic and sandals leading
three camels whose saddles are loaded with storage
jars, most likely for wine and oil.81 The mosaic, dated
to 722, decorated the floor of a church dedicated to S.
George at Dayr
, on the main route southwestward from Bostra. The extant panels il-

269/754

79. See Peters, JNES 37 (1978): 315, who identifies

the Chalcis and Beroea region and the Hawran as
scenes of intense encounter between pastoralists and
agricultural settlements. On the linguistic variety of the
Hawran, see Contini, Felix Ravenna 4 (1987), esp.
5166.
80. Bauzou in Hauran, 1: 13958.
81. Donceel-Vote, Pavements 4554. See Bulliet,
The Camel and the Wheel, esp. 1927, for the advantages of camel transport in Syria. Reinvestigation
of Roman roads in northwestern Syria has confirmed
Bulliet's suggestion that the camel replaced the wheel
as preferred means of transport from the sixth century
(French, IstMit 43 [1993]: 451). See also Graf, Electrum 1 (1997): 4349, who argues that camels and
ox-drawn wheeled transport served different purposes: the former were best fit for long hauls, while
the latter excelled locally with large and heavy loads.

Page 76
lustrate a range of activities common to much of the
Mediterranean world, but specifically applicable in this
instance to the landscape of late antique Hawran: the
harvesting of grapes, for example, the capture of
birds, and the hunting of hares. The panel shows
camels, each with a bell around its neck, kept in
single file by a rope attached to their simple harnesses as they pass through a field of flowering
plants. The saddles, loaded with two jars on each side
of the hump, rest on a tasseled blanket and are secured by ropes around the neck and tail. The cameleer is unarmed and carries only a walking
stickclearly, this is a local scene and not a depiction
of a long-distance caravan with armed escorts and
greater numbers of beasts. An inscription names the
cameleer as Mouchasos, a Semitic name well attested in the Hawran. It is likely that Mouchasos was a
pastoralist, full- or part-time, who hired out his services as cameleer to local farmers. The products depicted could well have been his own if he cultivated
the rich soil of the Hawran in addition to herding
beasts. Certainly during the hot, dry season, when the
nomadic segment of the population migrated with their
flocks back to the well-watered Hawran and stayed

271/754

from April until October, the region would have seen

the entire spectrum from settled to fully nomadic
Arabs.
The Roman buildup of Rusafa can be seen as a response to the vacuum created in the region by the demise of the caravan city and oasis of Palmyra in the
late third century. The objects inscribed with
"camelarius of S. Sergius" or "camelarius of the Barbarikon" only hint at Rusafa's place in caravan
trade.82 Rusafa, situated in a plain near the Euphrates, was highly accessible to a Roman force, and
had a long record of service as a crossroads and watering point. To exploit these advantages was part of
Diocletian's larger fortification project in central SyriaMesopotamia. The new network of roads and fortresses, manned by nine legions, would facilitate rapid
concentration of forces in the region, but served
primarily for the maintenance of local order,83 as is
depicted in the Passio of SS. Sergius and Bacchus
when a party from Sura attempted to steal the
martyr's relics.84 While interest in Rusafa as a monitoring station was established by the time of the
tetrarchy, the growth of a martyr cult at Rusafa set the
fort apart as an unusually useful site. In the eyes of

272/754

political leaders, an alliance with a powerful divine defender could offer advantages on many levels. The
convergence at Rusafa of trade, pastoralism, and pilgrimage, presented an opportunity to monitor the
fluidity of movement both within and through the frontier zone. And as the influence of Rusafa's martyr increased, control of the cult and its shrine could offer a
step toward control of the mobile populations who
passed by
82. See above, pp. 3543.
83. Gawlikowski, JRS 84 (1994): 246; Whittaker,
Frontiers, 135.
84. The debatable nature of the early sections of the
fifth-century Passio make it rather weak evidence for
the monitoring duty of soldiers based in frontier zone
castra.

Page 77
the holy site. No doubt with the lesson of Palmyra in
mind, Roman emperors from Anastasius to Maurice
attempted at Rusafa to exert their authority in the
steppe through fortified, non-native urban architecture, and most of all through patronage of the local
martyr Sergius.

THE FORTRESS-SHRINE
Like many Syro-Mesopotamian settlements, Rusafa
was plagued by the disruptive effects of earthquakes
and shifting ground. A complex system of dolines runs
beneath the city itself, and this was probably the
cause of much of the structural damage from which
Rusafa's buildings have suffered. Excavators have
unearthed from the fortress-shrine signs of significant
late antique rebuilding of older, damaged structures
that were dispossessed of their original function and
turned into urban quarries, as in the case of basilica
B. Excavation is still in progress, and new finds reveal
what one would expecta complicated architectural
history. As it is uncovered, material evidence joins the
literary to portray a major frontier pilgrimage center,
which reached its peak of activity and influence in the

274/754

late fifth, sixth, and the early seventh centuries. During this period the relics of S. Sergius were moved to
a new shrine, the second within the walls. The date of
the translation is still uncertain, but both the old shrine
outside the walls and the new shrine within provide
clues to the nature of the Sergius cult at Rusafa.
Much of the walled settlement today lies hidden beneath up to 4 m of rubble, and soil blown in from the
surrounding steppe over the centuries. The visual result was famously captured on film by the scholar-aviator Fr. Antoine Poidebard in 192532 (title page).85
The pocked landscape is punctuated by the ruins of
imposing structureschurches primarilythe object
of comment by travelers and scholars since the late
seventeenth century.86 The massive gypsum walls
that enclose the settlement rise to a height of 14 m
and form an uneven quadrangle measuring 536 m on
the north, 350 m on the east, 549 m on the south and
411 m on the west. The area within the walls measures 210,000 m2.87 Fifteen towers, round or
pentagonal, reinforce the walls, which are pierced by
four main tripartite
85. Poidebard, Trace de Rome, pl. 75.

275/754

86. Clandestine digging is responsible for the lunar

appearance of the area within the walls. Musil,
Palmyrena, 14042, provides a good example of how
an entire cemetery can be destroyed by treasure
hunters in a very short span of time. Trade in antiquities is a long-established business along the
PalmyraArakSukhnaTayyibaal-KawmRusafa
route (Wirth, Syrien, 443). It was well known that the
inhabitants of Sukhna conducted profitable freelance
excavations at Rusafa (Boucheman, Petite Cit caravanire, 8284).
87. Karnapp, Stadtmauer, 89.

Page 78
gates and two minor entrances. The northern gate, facing the Euphrates, is elaborately carved in a style
that has especially close parallels in northern Syria
and Mesopotamia, most notably at Edessa and Dayr
al-Zafaran near Mardin.88 Although the street plan
has been only partially explored by excavation, it is
clear that the eastwest axis linking the east and west
gates intersected two major streetsone leading from
the north gate, whose juncture with the eastwest axis
took the form of an open square, and another further
to the east and proceeding from the southern gate,
which ran alongside the western courtyard of the city's
largest church, basilica A.89 A visitor entering the city
in the mid sixth century through the northern gate
passed along a street lined with columns, archways,
courtyards, and shops, which sprawled out onto the
sidewalks.90 Three churches in particular would also
have caught the eye: basilica A in the southeast
quarter, basilica B south of the city center, and a tetraconch basilica surrounded on all sides by a courtyard, which opened onto the northern artery through
two tetrapyla (fig. 8).91

277/754

As was stated in chapter 1, the brick building located

below the later construction now called basilica B and
dated by finds to no later than 425 has been identified
with the martyrium built by Bishop Alexander of Hierapolis.92 Evidence dating from the Flavian period has
also come to light, although not necessarily military in
nature. This date corresponds with Vespasianic work
on the road between Palmyra and the Euphrates,
which passed through Arak and Rusafa. The finds under basilica B also reveal a gap
88. Kollwitz in Neue Deutsche Ausgrabungen, 50. For
discussion of the style of the architectural ornament,
see Brands in Akten, 59091, with pls. 73 and 74, and
id. in Sptantike und byzantinische Bauskulptur,
passim, esp. 72.
89. See Ulbert, DaM 6 (1992): 4036, on the south
gate street and church courtyard; also the reports in
AA 1991: 67, on the plaza; AA 1990: 613, on a house
excavated at the intersection; Ulbert, Resafa, 2:
14748, on the street leading from the south gate;
Kollwitz, AA 1963: 35055, on the street in front of the
tetraconch church; Kollwitz in Neue Deutsche Ausgrabungen, 5052, for a general overview of Rusafa's

278/754

street plan, and 51 on the north gate street, first excavated in 1956; Kollwitz, AA 1957: 1029, on the
street leading from the north gate and on the khan;
and Karnapp, AA 1978: 13650.
90. Sauvaget, Byzantion 14 (1939): 12627, observes
the high proportion of space within the walls taken up
by nonresidential constructions. In the spring of 1989,
a domestic complex near the southern end of the
north-south street was excavated. Precise dating was
not possible, but it is clear that occupation spanned
from late antiquity to the thirteenth century (Wemhoff,
DaM 8 [1995]: 24768; and see also Logar, DaM 8
[1995]: 26992, for the ceramic evidence).
91. Also excavated is the small basilica C, located
northeast of the basilica A complex, to be published
by Thilo Ulbert. On the tetraconch courtyard, see
Karnapp in Studien, 12532.
92. See the fuller discussion of the evidence above,
pp. 2628.

Page 79

Figure 8. Rusafa: Plan of settlement.

Page 80
in evidence between the Flavian period and the beginning of the late Roman phase.93 The finds help to locate for the first time at least one part of the area that
lay within the tetrarchic castrum, since the early fifthcentury martyrium (the forerunner of basilica B) is said
in the Passio and in the correspondence of Alexander
of Hierapolis to have been built within Rusafa's
walls.94 The location of the original martyrium outside
the walls remains a mystery, although the cemetery
north of the walls may offer a clue.95
Basilicas A and B
Basilica A, Rusafa's largest church, lay in the city's
southwest corner and the monumental three-aisled
basilica was identified by many early travelers as the
martyrium of S. Sergius (fig. 9).96 The basilica is the
focal point of a complex assemblage of buildings that
underwent considerable elaboration at later stages.
On the north, it opens onto a spacious peristyle courtyard, while various annexes, some still not securely
identified, grew up to the south.97 Part of the area to
the east has not yet been excavated. From at least
the eighth century, permanent shops and workshops
lined the perimeter wall of the courtyard west of the

281/754

basilica.98 The shops have been interpreted as part

of the mosque complex that incorporated one-third of
the north peristyle courtyard, but one must wonder
whether they replaced more ephemeral installations
from earlier periods. The interior of the monumental
three-aisled basilica is divided by half piers and wide
arches, lend93. Konrad, DaM 6 (1992): 348. Mackensen, Germania 61 (1983): 56578, esp. 578, discusses a mid
third-century fibula discovered at Rusafa.
94. Konrad, DaM 6 (1992): 350.
95. See Kollwitz in Neue Deutsche Ausgrabungen,
48, on the cemetery north and northeast of the walls;
and see also pp. 14973 below.
96. For the history of scholarship on basilica A, see
Ulbert, Resafa, 2: 15.
97. Ulbert, Resafa, 2: 11827, 14445. Kollwitz in
Neue Deutsche Ausgrabungen, 64, and AA 1959:
88101, and before him Musil, Palmyrena, 157, 268,
had hypothetically located a monastery in this complex. Without argument, Ulbert, AA 1977: 569, n. 23,

282/754

disregarded this view, but the possibility of monastic

quarters in the basilica A complex, at least at a later
stage, has begun to reemerge (AA 1993: 730; Ulbert,
Resafa, 3: 3). Literary evidence of a sixth-century
monastery at Rusafa consists of the following: "The
humble sinner Joseph, bishop of the holy monastery
of Rasiphta" appears in a colophon of a manuscript
containing works of S. Ephrem that belonged to a certain Zooras, son of Paul of Tagrit, who gave it to the
Syrian monastery in Scetis. The final sentence of the
Syriac colophon asks the reader to pray for Bishop
Joseph (Assemani, Bibliotheca orientalis, 1: 117).
Assemani suggests that Zooras might be identified
with the non-Chalcedonian stylite who baptized Empress Theodora in 535 and in whose honor a nonChalcedonian church was dedicated at Amida in 650.
On Zooras, see Frend, Monophysite Movement,
26972. If the Zooras in the colophon is to be thus
identified, Joseph would have headed Rusafa's monastery in the late sixth century at the earliest and may
well have been associated with non-Chalcedonian
circles in the region. The existence of such a monastery at that time would not be at all surprising.
98. Ulbert, DaM 7 (1993): 11516.

Page 81

Figure 9. Rusafa: Plan of basilica A.

ing the space an airy grandeur. An upper order was in
turn supported on the half piers. In a later phase, and
no doubt for structural reasons, the wide span of the
arches was filled in and given further support by

284/754

columns and carved capitals reused from basilica

B.99 At the western end is a narthex divided later into
three rooms. A large bema in the North Syrian style
99. The date of reuse cannot be determined precisely,
but it took place certainly before the eighth century
(see below).

Page 82
crowns the central aisle, and the eastern end is divided into five interconnecting rooms.100
The task of dating and identifying the dedications and
functions of Rusafa's churches has been an arduous
one. And it is by no means complete. Thilo Ulbert's
welcome publication of basilica A in 1986 was the first
systematic presentation of a building since the professional spade first struck the soil of Rusafa.101 Recent
investigation by the architectural historian Gunnar
Brands has radically revised our understanding of the
relationship between basilica A on the one hand, and
three related projects, the tetraconch, basilica B, and
the north gate, on the other.102 The details will be
discussed as necessary, but the crux of the problem
is a reused inscription in basilica A that refers to the
dedication of an unnamed benefaction to the Holy
Cross by Bishop Abraamios in 559. Ulbert understood
this inscription to be the key to the building's chronology, dating basilica A to 559. In 1986, after the publication of basilica A as the Holy Cross basilica, the
discovery of another inscription, which belonged originally to basilica B, led Ulbert to suspect the need to
revise the chronology he had previously established

286/754

for basilica A.103 Subsequently, Brands's study of the

architectural style of basilica A has stimulated him to
date the first phase of the structure to the last quarter
of the fifth century. With minor reservations, Ulbert is
in agreement with Brands's conclusions and no longer
considers basilica A to be the Holy Cross basilica as
argued in Resafa, vol. 2, Die Basilika des Heiligen
Kreuzes in Resafa-Sergiupolis.104 One of the many
important implications of Brands's conclusions is that
instead of being the last church to be built at Rusafa,
basilica A becomes the earliest of those still standing.
This qualification is important and will be discussed in
turn. The present account accepts Brands's chronology and will attempt to serve as a guide through the
somewhat tortuous obstacle course of research at the
site.105
The inscription on a marble plaque that led Ulbert to
identify basilica A
100. On the variety of uses made in Syria of the side
chambers at a basilica's eastern end, see Descoeudres, Pastophorien. Tchalenko, glises syriennes,
20612, reconstructs the developmental phases of
the bema based on his investigation conducted for the

287/754

Service des Antiquits in 1942. His posthumously

published discussion of Rusafa's contribution to the
history of bemata ignores the findings of the German
excavations. See Ulbert, Resafa, 2: 4, 25, 13637, on
the damage subsequently suffered by the bema,
which was left exposed to weather and vandals by
Tchalenko; also the review by Mango, JThS 43
(1992): 26366.
101. Ulbert, Resafa, vol. 2. In 1976 the walls were
published by Karnapp, Stadtmauer.
102. Brands, Resafa, vol. 6 (forthcoming).
103. Ulbert in Gatier and Ulbert, DaM 5 (1991):
16979, esp. 176.
104. Ulbert, letter to the author, 18 October 1996.
105. I would like at this point to express my gratitude
to Thilo Ulbert and Gunnar Brands. Both have responded generously to my questions about the architecture of the site and have commented on an earlier
draft of this chapter. I am particularly indebted to Gunnar Brands for sharing his work, often in unpublished
form.

Page 83
as a church dedicated to the Holy Cross was discovered over twenty years ago. It was found in the
floor of the sanctuary of basilica A, behind the central
opening of the templon, but this was not necessarily
its original location. It had clearly been at least twice
removed during restoration of the floors, broken at
some point, and reset for the last time by someone ignorant of Greek. Groups of letters are turned in improbable directions, like a forced jigsaw puzzle. The
inscription is, however, complete and gives both a
date and the name of a benefactor. Clearly it must
have been positioned so that the benefaction itself
was obvious, since it is not mentioned in the dedicatory inscription. The inscription reads:
Abraham by God's mercy bishop of Sergiopolis built in
honor of the Holy Cross in order to be made worthy of
God's mercy. It was accomplished in the month of
Artemisios in the seventh year of the indiction, in the
year 870.106
The date corresponds to April 559. Abraham was a
signatory at the fifth ecumenical council held at Constantinople in 553, and he later opposed Justin's edict

289/754

espousing Aphthartodocetism.107 Since the architectural decoration of basilica A places its first phase of
construction in the last quarter of the fifth century, the
object in honor of the Holy Cross cannot have been
the entire church. Instead, the dedication inscription
may have referred to the templon, as Brands suggests, or perhaps to the side chapel whose apse is
decorated with a large painted cross, admittedly not
an unusual decoration.108
At this stage, on the basis of architectural style, we
know that basilica A was constructed in the last
quarter of the fifth century and that, according to the
first inscription, in 559 Bishop Abraham adorned the
existing building with some further element, possibly a
templon. In order to assess the function of basilica A
and its relationship to basilica B in particular, we now
need to consider the other inscription already
mentioned.
106.
.
Ulbert, AA 1977: 56369; Gatier in Ulbert, Resafa, 2:
161. In line two,
, as in Ulbert, AA 1977: 56369,
and Gatier in id. and Ulbert DaM 5 (1991): 181, is

290/754

here corrected to read

. For other inscriptions in
basilica A, see Gatier in Ulbert, Resafa, 2: 16569.
107. Devreesse, Patriarcat 289.
108. Brands in Sptantike und byzantinische
Bauskulptur, 6768, suggests that Abraham's dedication was the templon. For a parallel dedication to a
saint of an architectural element within a church, see
IGLS, no. 1970, in this case a door. Also found at basilica A is an inscription on a large marble plaque decorated with a cross, which was part of the templon
and records the dedication in August 559, but without
mention of a donor (see Gatier in Ulbert, Resafa, 2:
166, no. 5). In the remarks following Ulbert in Actes,
457, Falla Castelfranchi raises the possibility of a connection between the painted cross in the side chapel
and the Abraamios dedication.

Page 84
This inscription dates the inception of work on basilica
B to the spring of 518 under Bishop Sergius. It was
discovered in the Umayyad mosque north of basilica
A and first published in 1991.109 Although it was
found reused in the mosque, its original home was in
basilica B, as the comparison of door profiles has confirmed. Basilica B was constructed on the oldest site
at Rusafa to have been used for Christian cult, but the
three-aisled structure visible today belongs to the
building activity dated to 518.110 The inscription
reads:
This holy church, once . . . and made of brick, held the
holy relics of Sergius the victorious martyr until the
other venerable shrine which at present holds the holy
sarcophagus was constructed. It was transformed and
rebuilt from its foundations with great generosity by
the most God-loving bishop Sergius II, the kinsman of
Maronius the chorepiscopus. He began the project in
the month of Dystros, the eleventh year of the indiction, in the year 829 and completed it in the month
of . . .111

292/754

Three churches are mentioned in this inscription: an

old brick church (the former home of Sergius's relics),
the new church on the same spot, and "the other venerable shrine" whither the relics had been removed.
The Passio claims that a church was built within the
city, and that Sergius's relics were translated there
from outside the walls. The inscription supports the
Passio, revealing that the church begun by Bishop
Sergius in 518 replaced an earlier brick-built church
that had formerly held the bones of S. Sergius. Most
importantly, it now becomes clear that the "other venerable shrine" is the late fifth-century basilica A, which
housed the relics of S. Sergius after they were removed from the first brick church within the walls. This
is further confirmed by the plan of basilica A, discussed in detail below.
The east end of basilica A was especially designed to
accommodate martyr cult. The room to the south of
the apse would have served as a pastopho109. Gatier and Ulbert, DaM 5 (1991): 16982.
110. Kollwitz in Neue Deutsche Ausgrabungen,
5360; and id., AAS 89 (195859): 24; DonceelVote, Pavements, 28182.

293/754

111.

vacat. Gatier, DaM 5 (1991): 179; SEG 41

(1991), no. 1537. I would offer a slightly different reading in l. 2. The photograph and drawing show
. The break in the stone allows for only one
letter, which I suggest should be an iota, producing
. Gatier inserts two letters and only brackets
the upsilon. But the photograph and drawing indicate
no omicron.

Page 85
rion, while the plan and decoration of the room immediately to the north marks it as the shrine for relics.
This room was increasingly elaborated over the
course of its four phases of renovation, and the floor,
in particular, underwent several stages of embellishment, first with simple gypsum paving stones, followed by limestone and finally by precious marble in
green, white, and pink.112 Other architectural features once distinguished the shrine as a place of
heightened importance, such as the towerlike structure that covered the space, a vaulted roof with mosaics, and wall revetments.113 On a table fitted into the
platform, at the focus of the room's splendor, would
have rested the silver oblong reliquary sarcophagus of
S. Sergius.114
Access to the shrine north of the apse was possible
from both the north aisle of basilica A and from a
chamber that communicated with the shrine from the
north. A split-leaf capital inscribed with the name "Sergis," written from right to left in Greek, was reused at
the eastern end of the row of columns that divide the
central nave from the northern aisle.115 Of all the
capitals reused from basilica B, this one alone bears

295/754

the saint's name and was no doubt positioned closest

to the shrine in order to further emphasize the holiness of the space. The chamber to the north of the
shrine opened, in turn, onto the monumental peristyle
courtyard on its north side.116 In this way, pilgrims
would have had easy access to the shrine from the
northern courtyard. This architectural arrangement
could also accommodate the ceremonial display of
the relics to a large crowd.117 The walls of the chamber that communicates with the shrine from the north
were carved with pilgrims' graffitinames, prayers to
Sergius and the Trisagion. Fragments of flasks and
the recess in the sarcophagus base found in the
northern chamber suggest that oil, or water, was
poured over the bones, collected, and then distributed
among the pilgrims who were drawn to Rusafa by S.
Sergius's reputation for healing.118 The Coptic recension of the Arab Jacobite Synaxa112. Ulbert, Resafa, 2: 4362, with reference to figs.
28, 30, 31, and 34.
113. A marble plaque with a damaged Greek inscription was found in situ in the final phase of the pavement (Gatier in Ulbert, Resafa, 2: 16566, no. 4).

296/754

114. See Evagr., HE 4.28, on the silver sarcophagus.

115. For other Syrian inscriptions in Greek written
backwards, perhaps under the influence of Semitic
writing, see Prentice, Greek and Latin Inscriptions, 82,
210.
116. Ulbert, Resafa, 2: 143.
117. Ibid., 17177. The courtyard of the healing shrine
of S. Thecla at Seleucia was a central gathering place
that played an important role in the process of pilgrimage and healing (Mir. Theclae 12.2425, 24.22; Dagron, Vie et miracles, 6970).
118. Ulbert, Resafa, 2: 142. We cannot know whether
contact with the sarcophagus was permitted, although
it seems very likely (Kollwitz in Neue Deutsche Ausgrabungen, 57). Sergius's reputation as a healing
martyr had reached Gaul by at least the late sixth century (Greg. Tur., Glor. mart. 96). On the role in martyr
cult of oil blessed by contact with martyrs' bones, see

Page 86
rion mentions the existence at the shrine at Rusafa of
a marble basin into which poured perfumed oil infused
with healing powers.119
The shrine, like all buildings so far excavated at
Rusafa, has been thoroughly robbed of its valuable
objects. Here were found only fragments of pilgrims'
flasks, small change and hundreds of trinketsmostly
earrings and bracelets of cheap metal or bone. These
were the votive offerings of ordinary pilgrims.120 The
greatest number of such objects was found beneath
the floor of the second phase of the martyrium in basilica A. A plausible explanation is that the valuable
votive offerings had been removed at the time of
renovation, while some of those of no monetary value
were left behind.121 These votives can only hint at
the irretrievable variety of the offerings to the miracleworking martyr. The objects found in the shrine of basilica A are precisely what we would expect as offerings of the region's pastoral Arab population, among
whom the cult of Sergius was especially popular.122
The only articles of monetary value that have been recovered by archeological investigation are part of a
five-piece liturgical assemblage dated to the twelfth

298/754

century, which allows us a glimpse of the opulence

that would once have characterized the saint's treasury.123 We also have a story told by Gregory of Tours
that is set at Rusafa. He relates how a poor woman
gave the martyr a few chickens as part of a vow, and
when two unscrupulous men stole one of her offerings, Sergius retaliated by petrifying the dedicated
bird still in the cooking pot, leaving the thieves to disinvite their hun
Lassus, Sanctuaires, 16367; Delehaye, AB 53
(1935): 225. Pieces of cloth rubbed against the holy
relics might also have offered the pilgrims contact with
the martyr (see, e.g., Greg. Tur., Glor. mart. 27).
119. This detail in the undated Coptic recension of the
Arab Jacobite Synaxarion, 7 October, has not been
noticed by students of the Sergius cult or of Rusafa.
The recension gives a boiled-down version of the
Greek Passio, but concludes with this information, added most likely by someone who had visited the
shrine, or had heard about it at second hand. Cf. Mir.
Theclae 7, where oil from the lamp that burned on the

299/754

holy bema was taken away by pilgrims for its curative

powers; also Sophr., Mir. Cyr. et Io. 7.3.
120. Ulbert, Resafa, 2: 5052, 5960, 140, with n.
281. On the coins, see the study of Mackensen in Ulbert, Resafa, 2: 181225. For a photograph of an incense burner, probably seventh-century, discovered
in the 1974 excavations (find-spot unspecified), see
the catalogue in Ruprechtsberger, ed., Syrien, 443,
fig. 87.
121. The Piacenza Pilgrim saw armlets, bracelets,
rings, tiaras, and emperors' crowns among the votives
which adorned the tomb of Christ (Itin. 18). Before the
redating of basilica A to the last quarter of the fifth
century by Brands in Sptantike und byzantinische
Bauskulptur, 5974, Ulbert had suggested that the
votives found in the church's second phase might
have been brought there from the trichoros of basilica
B, considered at that time to have been the martyrion
of S. Sergius (Ulbert, Resafa, 2: 140).
122. On Arab devotion to S. Sergius, see the discussion of the History of Ahudemmeh below, pp. 12128.

300/754

123. Ulbert, Resafa, vol. 3. See also below, pp.

18587.

Page 87
gry guests.124 Pilgrims left birds as ex-votos in the atrium of S. Thecla's healing shrine near Seleucia. This
consecrated petting zoo housed doves, swans,
geese, pheasants and perhaps ibises, and provided
innocent distraction for the pilgrims' children, as well
as for the pilgrims themselves, who fed the birds with
grain, no doubt sold on the spot by enterprising locals.125
The inscription from basilica B advances our understanding of basilica A significantly. Less apparent
from Bishop Sergius's epigraphy is what purpose his
new church served once the relics, formerly in the old
brick church at that site, had been moved to basilica
A. The inscription was carved on a tabula ansata on a
door lintel that originally belonged over the main
southern entrance of basilica B.126 Bishop Sergius
attached his name to the new structure's inception
and left a blank space at the end of the lintel inscription, presumably for the date of its completion.
However, the space was never filled in. We last hear
of Sergius's activities in 524, when he was sent by the
emperor Justin to the conference at Ramla.127 Under
normal circumstances, one would expect the

302/754

construction of such a church to require an average of

ten years. But perhaps by the time the new church
was finished, a bishop other than Sergius occupied
Rusafa's episcopal throne. Bishop Sergius had also
marked his role in the church's progress with another
inscription, which was found reused in the apse floor
of basilica A:
By God's mercy, Bishop Sergius II, kinsman of
Maronius chorepiscopus, built the most holy church
and roofed it and supplied doors and various
marbles[?]. Work was begun in the month of Dystros
in the indiction . . .128
Basilica B was adorned with architectural detail and
ornament to glorify God and awe its visitors. Although
smaller than basilica A, its size, 48.50 m in length and
25.70 m in width, still places it among Syria's largest
churches. The plan reflects the familiar North Syrian
basilical style with the typical
124. Greg. Tur., Glor. mart. 96. Cf. the description of
festal picnicking on the feast of S. Thecla in her shrine
at Seleucia (Mir. Theclae 26, pp. 35657).
125. Mir. Theclae 24.2332.

303/754

126. Ulbert, DaM 5 (1991): 17273, suggests the

southern rather than the western entrance.
127. Justin's choice of the bishop of Rusafa as
Abraham's companion on his diplomatic mission to
the Lakhmid al-Mundhir is another example of
Rusafa's ideally suited situation for mediation between
Rome, Iran, and the Arabs. For the Ramla conference, see Shahd, Byzantium and the Semitic Orient
VI, 11531, esp. 122 n. 29.
128.
.
Gatier, DaM 5 (1991): 181 is a revised reading (in the
light of the lintel inscription) of the inscription published earlier in Ulbert, Resafa, 2: 16264.

Page 88

Figure 10. Rusafa: Plan of basilica B.

porch running along the southern face (fig. 10).129
While the basilica's nave was simple and adorned
only by a Constantinopolitan-style ambo, the plan of
the eastern end was intricate, although it too followed
a pattern common in many Syro-Mesopotamian basilicas. On the south, a square pastophorion communicated with the apse. The pastophorion, in turn,
opened on its south side onto an apsidal chamber that
may have functioned as a funerary chapel.130 North
of the apse a door led into a triconch room, known as

305/754

the trichoros, whose architecture and ornament drew

attention to the importance of the space. The pavement was elaborate opus sectile, the conches were
covered in gold tesserae, and the profuse architectural decoration was painted. In the eastern niche of the
trichoros was discovered a rectangular platform with
four depressions to hold supports, probably for a
table, and
129. Kollwitz in Neue Deutsche Ausgrabungen, 54.
See also Krautheimer with
, Early Christian and
Byzantine Architecture, 151.
130. In the floor of the apsidal chamber a recess was
discovered, which in a parallel installation at Androna
held a sarcophagus. On these grounds, Kollwitz, AAS
89 (195859): 3132, interprets the chamber as a funerary chapel, possibly for the church's founder.
Donceel-Vote, Pavements, 180 n. 1 and 284 n. 20,
argues that the southeastern apsidal room was, instead, the site of martyr cult, since no sarcophagus
has been discovered in the chamber.

Page 89
a recess in the center connected to a channel that led
toward the back of the niche. Most likely the table
would have held a reliquaryperhaps of the sarcophagus type found, for example, at Apameathrough
which oil or water would be poured and distributed
among the pilgrims as an "eulogia."131 A well in the
northeastern corner of the trichoros may also have
played a part in the cult, perhaps by providing water to
be blessed by contact with the saint's relics.132 Small
finds from basilica B have been negligible.133
It has been claimed that the delicate nature of an
opus sectile floor in the triconch shrine in basilica B
excludes the idea that crowds of visitors were allowed
to troop through it.134 This suggestion cannot be
ruled out. It overlooks, though, the possibility that the
floor was covered with carpets, and is further
weakened by the presence of two narrow doorways
into the triconch room that would have been accessible to the laity, one from the north aisle and one from
the northeast room.135 Rather than being evidence
against accessibility, the narrowness of the entrances
can be seen as heightening the drama of encounter
with holy relics.136 Whether or not direct access was

307/754

131. Donceel-Vote, Pavements, 283 n. 15, draws attention to the similar installation for a table at the center of a side chamber in Huwirta's north basilica. For a
discussion of the sarcophagus-type reliquary and the
theology of blessed oil, see Gessel, OC 72 (1988):
183202. Writing in the mid tenth century,
, Muruj
al-dhahab 120, claims to have visited a church in
which he saw stone coffins containing human bones
into which a thick oil was poured so that the Christians
could receive a blessing from it.
132. Kollwitz in Neue Deutsche Ausgrabungen, 56;
id., AA 1954: 12228; and id., AAS 45 (195455):
8082, for reconstruction of the use made of the
triconch shrine and description of the well installed in
the northeastern room, which communicated with the
triconch shrine. Wells were common features of
martyr-cult installations, found also, e.g., at the
shrines of S. Demetrius at Thessalonica and S.
Thecla at Seleucia. See also Donceel-Vote in Akten,
19091. Lassus, Sanctuaires, 173, discusses parallel
martyr-cult installations in Syria. In comparison with
the other examples Lassus cites, the ornament of the
triconch room of basilica B distinguishes it as the
shrine of a particularly revered martyr.

308/754

133. In the triconch of basilica B, the only object noted

was a marble fragment with a letter A from an inscription (Kollwitz, AA 1954: 127).
134. Donceel-Vote, Pavements, 282.
135. On the use of carpets in late antique churches,
see Mango in Byzantium, 52. Perhaps too obvious to
have attracted comment is the long-established practice in the Middle East of removing one's footwear in
holy placesa habit maintained, for instance, in the
shrine of the Burning Bush at the monastery of St.
Catherine, Sinai, where flokata carpets cover the
floor. If public access was forbidden, Huwirta offers a
possible parallel to the martyrium at Rusafa. There
the western entrance to the martyrium north of the
apse seems to have been furnished with a low screen
with an opening in the middle (Canivet, Syria 55
[1978]: 15362, esp. 155). For general comments on
decoration of and circulation around martyr shrines,
see Donceel-Vote, Byzantion 63 (1993): 43738;
see also Lassus, Sanctuaires, 16283, esp. 17781,
on arrangement of and activity at martyria.
136. At the holiest of Christian sites, the tomb of
Christ, access has always been through a very narrow

309/754

and low entrance (Wilkinson, Jerusalem Pilgrims,

17576). Donceel-Vote, Pavements, 282, suggests
that the narrow entrances into the triconch martyrium
would have

Page 90
allowed, a reliquary sarcophagus, if that is what occupied the space, would certainly have been visible from
the two doorways.
On a wide ledge along the eastern wall of the adjoining northeastern annex was discovered another installation for a reliquary sarcophagus, this time with
the sarcophagus in situ.137 In the front of this sarcophagus was a hole from which exuded oil, which was
then collected in a recess below. Whose sarcophagus
this was is not known. South of the apse, near the
passageway between the pastophorion and the socalled funerary chapel, were found two graffiti mentioning the martyr Leontius, a popular soldier-martyr
from Tripolis. Kollwitz, who assumed that the triconch
housed the relics of S. Sergius, offered the suggestion
that it was Bacchus's sarcophagus, since the three
soldier-saints shared a dedication at Bostra's cathedral.138 The evidence that Bacchus's body, or parts of
it, was to be found at Rusafa is inconclusive. The Georgian recension of the lost Syriac Vita Gulanducht
refers only to the reliquary of S. Sergius at Rusafa,
while the later Greek recension mentions relics of
both saints at Rusafa.139 The Piacenza Pilgrim and

311/754

Michael the Syrian report that Bacchus's tomb was at

Barbalissus.140 It would

prohibited public access. She does not offer parallels

to a popular shrine forbidding direct access to its holy
relics, which at Rusafa were protected by a sarcophagus. See, e.g., Joh. Chrys., Hom. in martyres, PG
50.664, where he urges his flock to embrace martyrs'
sarcophagi. The Miracula Cyri et Ioannis, set at their
pilgrimage shrine at Menuthis near Alexandria, depicts numerous instances of the pilgrim's close contact with a sarcophagus: see, e.g., Sophr., Mir. Cyr. et
Io. 1.5, p. 244, where a sick boy's father is pictured
drenching (
) the sarcophagus with tears. Elsewhere, it is clearly stated that a grating enclosed the
sarcophagus and was opened at set hours: Marcos,
"Thaumata," 46. For a vivid description of Western
practices, see Brown, Cult of the Saints, 8788.
137. Kollwitz in Neue Deutsche Ausgrabungen,
5657; Kollwitz, AA 1957: 7076, esp. 74, for a photograph of the sarcophagus installation in the northeastern annex.

312/754

138. In Sptantike und byzantinische Bauskulptur, 64,

Brands also suggests that basilica B may have
housed the relics of S. Bacchus and S. Leontius.
Some later accounts of the martyrdom of SS. Sergius
and Bacchus, such as the mid ninth-century martyrologium of S. Adon of Vienna, include a mysterious
martyr named Julia, who is said to have been buried
with Sergius and Bacchus (cf. Acta. sanct. Oct
3.88385). Cf. also the account in the Coptic recension of the Arab Jacobite Synaxarion 114, wherein, on
their way to Rusafa, Sergius and his persecutors meet
an unnamed young woman who gives them a drink
and then follows the party to Rusafa, where she collects the martyr's blood. No further mention is made of
the girl, whose appearance in the Passio of SS. Sergius and Bacchus may be attributed to the convoluted
hagiographical developments shared by accounts of
the martyrdom of Victor and Corona (see above, pp.
1718n. 37 for bibliography). A philological investigation of these sources might be rewarding.
139. Vita Gulanducht 15.1 (Georgian recension); 20
(Greek recension of Eustathios).

313/754

140. Bacchus is mentioned at Barbalissus in the 570s

by the Piacenza Pilgrim (Itin. 47, p. 191); and by Mich.
Syr., Chron. 9.26 (tr. 2: 220), who reports that in 543
the Iranians stole the bones of S. Bacchus and the
gold covering from the sarcophagus of S. Sergius
from Barbalissus.

Page 91
not be surprising if a small relic or personal possession once associated with Bacchus, such as his
chlamys, was revered at Rusafa. It may also have
been the case that a relic or possession of Sergius
was housed in the trichoros.
Ulbert implies that basilica B, begun in 518, suffered
destruction by an earthquake at an early phase of its
existence. But the precise date when basilica B, irreparably damaged, became a quarry for other building
projects is not known. Ulbert has noted that the use of
different architectural fragments from basilica B in basilica A and in the Umayyad mosque suggests that repairs to basilica A, employing reused capitals and
columns, were undertaken before the mosque construction in the first half of the eighth century, which
exploited much of what materials remained from basilica B, primarily large building blocks from the basilica walls.141 Still undetermined is who or what was
the focus of devotion at basilica B until it was destroyed and its components transferred to basilica A.
Unfortunately, the present state of the evidence, both
material and literary, helps little with these puzzles.

315/754

The Tetraconch
Construction of the tetraconch, dated by finds and
style to the early sixth century, belonged to this buoyant atmosphere in which basilica B, erected over the
site of Alexander's shrine, also shared.142 But the
function of this building, once thought because of its
shape to be the martyrium of Sergius, is still somewhat unclear.143 The difficulties are linked with the
presence in the tetraconch of a synthronon with space
for the bishop's throne, a small baptistry communicating with the apse, and episcopal tombs, most notably
the tomb found north of the apse, which still bears the
name of Abraham.144 Of course, Abraham was the
name of the mid sixth-century bishop who is known
also from the inscription found in basilica A. These
features might suggest that the tetraconch served as
the cathedral, and it has been interpreted by some
scholars as such. However, the earlier basilica A
seems now a more likely candidate for identification
as the cathedral, and it
141. Cf. Ulbert, Resafa, 2: 151. Private houses were
built within the ruins of basilica B by at least the early
ninth century (Kollwitz, AAS 45 [195455]: 79, and
id., AA 1954: 12122).

316/754

142. Brands in Akten, 59097.

143. For the most recent discussion of the tetraconch
in the context of Rusafa's architectural development,
especially in relation to the north gate and basilica B,
see Brands in Akten, 59097, with pls. 73 and 74. For
earlier views, see Kollwitz, AA 1963: 34952;
Donceel-Vote, Pavements, 26870; Ulbert in Actes,
450; and Kleinbauer, DOP 27 (1973): 9598, who
maintains that it was the cathedral. Musil, Palmyrena,
15657, interprets the tetraconch as the martyrium.
Kollwitz in Neue Deutsche Ausgrabungen, 69, emphasizes the absence of evidence for martyr cult in
the tetraconch.
144. See also Kollwitz, AAS 89 (195859): 32;
Donceel-Vote, Pavements, 27172.

Page 92
too boasts tombs, although unidentified, and a lavishly
decorated cross-in-square building in the southeast,
which seems to have served as its baptistry.145 The
simultaneous existence of more than one church with
a baptistry is attested in other settlements too, such
as Zenobia on the Euphrates southeast of Rusafa.
One explanation that has been offered in the case of
Zenobia is that the two churches served two different
rites.146 A suggestive passage that may go some
way toward explaining the opaque situation at Rusafa
survives from the synod of Beth Batin near Mesopotamian Harran in 794.147 It seems that Chalcedonian
clergy had been collecting vows to S. Sergius from
non-Chalcedonians. The complaint voiced at the synod was that the Chalcedonians had not given any of
the collected vows to the non-Chalcedonian church at
Rusafa, which is mentioned specifically. At the synod
it was decided that in the future, non-Chalcedonians
were to give their vows only to their own bishop,
whose jurisdiction covered Rusafa. Although it does
not make clear whether either Chalcedonians or nonChalcedonians had a bishop resident at Rusafa, the
decision implies that both a non-Chalcedonian and a

318/754

Chalcedonian church existed at Rusafa, at least in the

late eighth century.

PILGRIMAGE TO RUSAFA.
At the beginning of the sixth century, several events
occurred that might be linked with the construction of
the tetraconch, basilica B, and the north gate. In the
510s, the emperor Anastasius had Sergius's thumb
translated to Constantinople, a clear sign of the increasing prestige of the saint, whose sacred story was
by then circulating in the form of the Passio known to
us.148 Also at this time, Anastasius elevated Rusafa
to the rank of a metropolitan see,149 and Rusafa
came to be known briefly as Anastasiopolis. More
lastingly, Rusafa was also called Sergiopolisperhaps again the new name should be understood in
connection with the earliest imperial interest in the
shrine and its relics.150 In the mid sixth century, Procopius provided
145. Ulbert, Actes, 44556. The second storey on the
south side of basilica A has been interpreted as episcopal quarters by Ulbert (Resafa, 2: 146).

319/754

146. Lauffray, Halabiyya-Zenobia, 1056; Tchalenko

too had noted this as a possible solution to the presence in basilica A of a North Syrian style bema, and in
basilica B of a Constantinopolitan-style ambo (glises
syriennes, 21112).
147. The decision is discussed by Degen in Ulbert,
Resafa, 3: 7172.
148. Joh. Diakrin., fr. 2.554, p. 156.
149. Ibid.; Honigmann, Evques, 1023; Devreesse,
Patriarcat, 122.
150. George of Cyprus, Descriptio orbis romani, ll.
88283, pp. 15152: "Sergioupolis, or Anastasioupolis, the present day Rattapha" (ninth century). Never
displaced by the name Sergioupolis, variants of
Rusafa are used in the mid fifth-century Passio; in
514, by Severus,

Page 93
a sketch of Rusafa's history, which emphasizes the
shrine's acquisition of wealth and influence:
There is a certain church in Euphratesia, dedicated to
Sergius, a famous saint, whom men of former times
used to worship and revere, so that they named the
place Sergiopolis, and they had surrounded it with a
very humble wall (
), just sufficient to prevent the Saracens of the region from capturing it by
storm. For the Saracens are naturally incapable of
storming a wall and the weakest kind of barricade, put
together with perhaps nothing but mud, is sufficient to
check their assault. At a later time, however, this
church, through its acquisition of treasures, came to
be powerful and celebrated. And the Emperor Justinian, upon considering this situation, at once gave it
careful attention. He surrounded the church with a
most remarkable wall, and he stored up a great quantity of water and thus provided the inhabitants with a
bountiful supply. Furthermore, he added to the place
houses and stoas and the other buildings which are
wont to be the adornments of a city. Besides this he
established there a garrison of soldiers who, in case
of need, defended the circuit wall.151

321/754

We have accumulated sufficient evidence for Sergius

of Rusafa's gradual rise to famefrom the Surans' attempted theft of his bones to Anastasius's successful,
if only partial, translationto treat with some suspicion Procopius's focus on Justinian as the city's
unique provider. The struggle for Rusafa between Alexander of Hierapolis and John of Antioch is clear illustration of the site's attractiveness some fifty years
before Justinian's birth.152 But this does not necessarily convict Procopius of deceit: Alexander's altissimi muri may well have seemed a
to
Procopius a century later.153 And whatever their
state, there is no doubt that the old walls were replaced by the time of Procopius's composition. Basilica A and the complex surrounding it spread out in
the southeast quarter.154 As for who built the walls:
there is a trend toward attributing to Anastasius sites
described as Justinianic by Procopius.155 Early on in
the De aedificiis,

who relies heavily on the Passio in his Hom. cath. 57;

and by John Moschus, Prat. spir. 180 (PG 87.3052b),
in the late sixth century, who calls the pilgrimage
shrine "Safas."

322/754

151. Proc., Aed. 2.9.39 (tr. Dewing).

152. See chapter 1.
153. ACOec. 1.4, p. 163 (Schwartz): "magnum id
ipsum templum fabricatum est et illi altissimi muri et
alia intra eandem munitionem aedificia." Kollwitz, AA
1954: 120, mistakenly interprets muri to indicate a
temenos wall around Alexander's church, rather than
the castrum wall.
154. Ulbert, Resafa, 2: 147.
155. For the history of scholarship on Rusafa's walls
in the context of Procopius's attributions, see
Karnapp, Stadtmauer, 5153; Karnapp, AA 1968:
30811. See Karnapp, AA 1970: 99102, on the north
gate. Dussaud, Topographie, 254, ascribes Rusafa's
walls to Anastasius.

Page 94
Procopius makes it clear that he includes among
Justinian's achievements projects undertaken while
Justin was still nominal ruler.156 The architect Walter
Karnapp, in his thorough study of the walls at Rusafa,
published in 1976, accepts that Justinian may have
been responsible for work at Rusafa already during
the reign of his uncle. Karnapp does not, however,
conclude that the walls are Anastasian: rather he
tends to uphold Procopius's attribution and does not
pretend to have reached firmer conclusions, thus
leaving the issue open to further investigation.157
Subsequently, Ulbert and Brinker in the course of excavation have accepted the Justinianic date with only
slight reservation.158 More recently, however, Gunnar Brands in a thorough study of the architectural
decoration has built up a strong case for the contemporaneity of the walls and gates, the tetraconch, and
basilica B, all of which belong by his calculation to the
first decades of the sixth century, which coincided with
the final years of Anastasius's rule.159 Brands does
not exclude the possibility that they were finished by
Justinian, but notes that no archeological evidence
exists to confirm the dating either way.160 And it
should be noted that even Procopius does not claim

324/754

that Justinian built a church at Rusafa. The "certain

church" known to Procopius would have been the Sergius martyrium, that is, basilica A.
In addition to walls, Procopius attests the existence in
the mid sixth century of cisterns, stoas, houses, and
other buildings characteristic of late Roman urban life.
Urban amenities both impressed visitors and provided
for them, and while the inhabitants of Rusafa reaped
financial profit from the assured influx of visitors, they
also bore the responsibility of furnishing the pilgrims
with food, water, and beds. Although Procopius mentions only houses and "other structures," inns must
have existed, and, in this case, common sense is supported by epigraphy.161 Kollwitz in his excavation of
the sixth-century level of basilica B discovered what
seems to be an enkolpion referring to an ecclesiastical hostel of S. Sergius.162 In addition, a two-storey

Croke and Crow, JRS 73 (1983): 14359, reattribute

Dara to Anastasius, although they are less successful
in redating Rusafa's walls to the earlier emperor. But
see also Whitby in Defence, 719, on the general difficulties of precise dating, and 724, with n. 8, on the

325/754

problems of dating the walls and gates at Rusafa in

particular.
156. Proc., Aed. 1.3.3.
157. See Harrison's review of Karnapp, CR 34 (1984):
106.
158. Brinker, DaM 5 (1991): 144; Ulbert, Resafa, 2:
14748; Kollwitz in Neue Deutsche Ausgrabungen,
46.
159. Brands in Sptantike und
Bauskulptur, passim, esp. 7274.

byzantinische

160. Brands, letter to the author, 7 December 1996.

161. Proc., Aed. 2.9.7.
162. Kollwitz, AA 1963: 35778, fig. 20; Harrauer,
Tyche 7 (1992): 43; Feissel and Gatier, BE 106
(1993): 558, no. 617. Cf. Zacos and Veglery, Byzantine Lead Seals, no. 1090A:
and no.
1318a and b:
.

Page 95
khan whose earliest phase belongs to the sixth century, followed by a series of later additions and reconstructions, has been excavated on the main street
leading from the north gate.163
Naturally, in a steppe settlement, the most fundamental necessity for both pilgrims and residents was water.
The brackish water drawn from wells provided for
general water requirements, while the flask-shaped
cisterns collected the rains that served the drinkingwater needs of Rusafa from at least the Diocletianic
phase. Flask-shaped cisterns have been found beneath basilica B and in the north courtyard of basilica
A, and would have been used in private houses
too.164 In the mid sixth century, a water supply system was constructed, consisting of a dam west of the
city walls, a main canal leading into the city between
two towers and feeding the largest cistern in the
southwestern quarter, and another cistern in the
northwest. At the beginning of the seventh century,
two more cisterns were added to the system, placed
parallel to the great cistern in the southwest.165
Clearly, great effort and expense went into these constructions. Besides the churches, it was the wells and

327/754

cisterns of Rusafa that fascinated later Arab

writers.166 They were exceptional features at a
steppe settlement too far from the Euphrates to exploit that source in any but the most exceptional
circumstances.
The pilgrims who came to pray before Sergius's relics
spread the martyr's fame and increased Rusafa's
prosperity. Theodoret of Cyrrhus, a close friend of Alexander of Hierapolis, noted the rise of the Sergius
cult at Rusafa in the context of more general remarks
about the shift away from the gods of polytheism toward the cult of Christian saints. In the first half of the
fifth century, Theodoret proclaimed that the old festivals, the Pandia,
163. Karnapp, AA 1978: 13650.
164. Brinker, DaM 5 (1991): 12325, 13536.
165. Ibid., 12446, esp. 13543, on the canals and
water collection. See also Kollwitz, AA 1965: 359. According to Brinker, DaM 5 (1991): 14445, the total
volume capacity of the cisterns would have been
about 21,000 m3, and this sixth-century system must
have been in operation by 542, when Rusafa's

328/754

inhabitants successfully withstood a siege by Khusrau

I. The same argument has been used for the existence of the sixth-century walls by 542 (see Karnapp,
AA 1970: 121). According to Proc., BP 2.20.1415,
Khusrau withdrew after a short siege because of lack
of water for his troops, while according to Evagr., HE
4.28, S. Sergius miraculously saved the city. Neither
outcome necessarily depended on a new water supply system within the walls. See Gatier, DaM 10
(1998): 23740, for a dedicatory inscription of a
.
166. See Hamza al-Isfahani in Yaqut,
al-buldan
3.47, where Hamza attributes the repair of Rusafa's
cisterns to
ibn al-Harith. On the identity of this
figure, see Kellner-Heinkele in Sack, Resafa, 4: 146 n.
114. Ibn Butlan in Yaqut,
al-buldan 3.48, describes the cavernous underground cistern with its
vaults and marble columns, like the church above
ground; and
in Yaqut,
al-buldan 3.47, notes
that when, as sometimes happened in the summer,
the cisterns ran dry, the wealthy inhabitants sent
slaves with donkeys overnight to the Euphrates to
fetch water.

Page 96
Diasia, Dionysia, had been superseded by feasts in
honor of Peter, Paul, and a company of Syrian martyrs, Thomas of Edessa, Sergius of Rusafa, Marcellinus, Antoninus, Mauricius of Apamea, Leontius of
Tripoli.167 At the beginning of the sixth century,
Severus of Antioch in his homily on Sergius remarked
how the Arabs from the region around Rusafa were
drawn to Sergius's tomb to "take on the yoke of God,"
in other words, to be baptized.168 Images of Sergius
and his martyrdom, along with their exegesis, must
have played a crucial role in conversion, but also in
the maintenance of the cult.169 This would have been
the case particularly at Rusafa, where a wide variety
of visitors converged, some of whom were surely
drawn to enter the shrine out of curiosity. Unfortunately, no images have survived at Rusafa.170
Gregory of Nyssa's portrait of the martyrium of another popular Eastern saint, Theodore of Euchaita, suggests what we should expect at Rusafa: the ornate
decoration is described as "psychagogic," the mosaic
floor deserves high praise, and the walls are painted
with the martyr's passiones171 The Acta of Anastasius the Persian is one of many late antique accounts
that link conversion with images. Anastasius's story is

330/754

especially illuminating in the context of Rusafa's inevitable linguistic diversity. A defector from the Iranian
army in the early seventh century, Anastasius settled
in Hierapolis, where he, unfamiliar with Greek, apprenticed himself to a fellow Iranian goldsmith. This
man was a Christian and introduced Anastasius to the
martyrs of his religion through the wall paintings of his
parish church.172
Although Sergius received visitors year-round, his annual feast must have provided the opportunity not just
for a communal celebration, but for a market of notable size and diversity, which would have assembled
around the fortified settlement located at a major
crossroads. In the mid sixth century, Choricius of
Gaza described the vibrant feast of S. Sergius celebrated at Gaza in terms that might also characterize
the scene at Rusafa on Octo167. Theod., Graec. aff. cur. 8, p. 335.
168. Sev. Ant., Hom. cath. 57, p. 93; Kugener, OC 7
(1907): 40812. For the use of yoke imagery for baptism, as well as the less probable use to indicate tonsure or marriage, see Lampe, Lexicon, s.v.
.

331/754

169. On the images of S. Sergius, see chapter 1.

170. Tesserae from the destroyed dome mosaic and
traces of wall painting are discernible in the shrine
north of the apse. Cf. also Ulbert in Pietas, 156263,
where he discusses the sigma-shaped mensa plate of
verde antico found in the last phase of the shrine. On
the traces of mosaic decoration found in the basilica B
trichoros, see Mundell in Iconoclasm, 6769, who
suggests on the basis of the colors of the fallen
tesserae that the pattern was more probably floral
than figural.
171. Greg. Nys., De Sancto Theod. 6263.
172. Acta Anast. 8, with Flusin, Saint Anastase, 1: 50
n. 27 and 2: 228 n. 45, on the role of martyrdom depictions in conversions.

Page 97
ber 7, noting that it took place "[d]uring the best time
of the year, when the bodies are neither oppressed by
cold nor enervated by heat, and when day and night
have just made their truce with one another and
agreed to have an equal share; when the weather is
especially pleasant for everyone and gathers for us
many people from everywhere, when there are no
winter showers or the burning rage of the sun which
ravage travellers."173
Already in the mid fourth century, bishops and others
had attempted to exploit the overlapping of spiritual,
social, commercial, and political interests at Christian
feasts. Basil of Caesarea, for example, took the opportunity to extract favors for his clergy from government officials who were drawn to panegyreis along
with the sick, the hopeful, and the ecclesiastical hierarchy.174 He also warned against the excesses that
came in the wake of the week-long festivitiesbawdy
mimes, dancing, and overindulgence of all varieties.
That religious festivals inspire a wide range of activitiesfrom divine visitations to daylight robberyhas
always been the case. Some saw in the Arab conquest God's wrath against the Christians of Syria who

333/754

instead of celebrating the martyrs' feasts with fasts, vigils, and psalms, indulged in drunkenness, dances,
and other forms of debauchery.175 From the many
examples of late antique panegyreis, two in particular
stand as close parallels for Rusafa's early autumn
feast. Still in Sozomen's day, the Oak of Mamre near
Hebron was the scene of annual summer festivities
celebrated together by Christians, Jews, and polytheists: "the panegyris draws . . . locals and others from
further afield, Palestinians, Phoenicians, and Arabs;
many others, both buyers and sellers, gather there on
account of the fair."176 Similarly, Ammianus reported
that merchants flocked to Batnae just west of the Euphrates for the yearly festival (annua solemnitas) held
there in early September.177 The celebration and accompanying market would spread out before and after
the
173. Chor. Gazae, Laud. Marc. 1.14 (tr. Litsas). Litsas, JB 32 (1982): 43031, searches through obscure Sergii and unusual festal dates for Sergius and
Bacchus in order to find a date in late May for the
Gaza feast of S. Sergius. Although the setting depicted could easily be in springtime, there is, in fact, no

334/754

need to look beyond 7 October for a date that fits

Choricius's description of the weather conditions.
174. Basil, Ep. 142, 2.6465; Hom. in divites 8081
(PG 31.281c; 31.1020bd), on the mercantile dimension of martyrs' panegyreis, a characteristic feature
also of pre-Christian festivals. See also Mitchell,
Anatolia, 2: 6970.
175. Mich. Syr., Chron. 11.7 (2.422, tr. Chabot).
176. Soz., HE 2.4.13. On the tree's environment, see
Hepper and Gibson, PEQ 126 (1994): 94105.
177. "[A] great crowd of every condition gathers for
the fair, to traffic in the wares sent from India and China, and in other articles that are regularly brought
there in great abundance by land and sea." (Amm.
Marc. 14.3.3, tr. Rolfe). See also Amm. Marc. 18.9.13,
on the annual fair in Amida's suburbs, which drew the
region's peasants as well as foreign merchants. For
another festival held in honor of the saints at Arsamosata, see Chron. Edessa of 506 26162, p. 193.

Page 98
religious climax, and one can easily imagine the merchants traveling with great profit from feast to feast, as
they still do, for example, in modern Greece.178
The martyrium of S. Sergius at Rusafa had by the
sixth century become one of the major eastern
shrines included on late antique pilgrimage itineraries.
On a grand tour of eastern martyrs, for example, the
monk John, who made his home in a cave in the
Judaean wilderness, visited S. John at Ephesus, S.
Theodore at Euchaita, S. Thecla at Seleucia, and S.
Sergius at Rusafa.179 There is no doubt that Rusafa
was a major shrine. But Rusafa's setting in the steppe
necessarily influenced its everyday clientele, made up
primarily of the mobile population of Syria and Mesopotamia, merchants, pastoralists, even migrant workers whose path brought them to the shrine.180 The
majority of late antique pilgrims' accounts of eastern
holy places were written by Westerners, which helps
to explain their occasional confusion in the face of so
many exotic places and their stories. Rather than itineraries and narrative descriptions, the student of eastern Christian pilgrimage is often obliged to piece together more scattered evidencereferences in saints'

336/754

lives, homilies on martyrs and saints, and inscriptions

left at holy sites. The heterogeneity of this evidence
should cause us to reflect on the variety of pilgrims
and their ways of encountering the shrine at Rusafa.
To begin with, many eastern pilgrims to Rusafa did
not share the western pilgrims' habit of recording their
observations in writing. Many of the eastern pilgrims
were pastoral Arabs.181 Pilgrimage for easterners
was often an accompaniment of journeys taken for
other reasons, such as trade or seasonal migration
with flocks of sheep, goats and camels.
As has been noted already, Rusafa lay on the natural
route for migration dictated by the existence of a major pass through the steep-sided Jabal Bishri and
Jabal Abu Rujmayn. The concluding section of the
Passio states that at the time of the feast on 7 October, wild animals were wont to gather at Rusafa. It
was at precisely this time that the seasonal rains were
expected to begin and that wild animals, especially
gazelle, as well as pastoralists, would be migrating
southward from summer pastures north of the mountains to winter pastures in the Syro-Arabian
desert.182 The addition of this

337/754

178. Tardieu, Paysages, 1314, notes that the major

Syrian panegyreis, including one at Salaminias, occurred during the summer season, which extends into
late September.
179. Joh. Mosch., Prat. spirit. 180 (PG 87.3.3052b).
180. A wealthy man from Sheba learned from merchants of the healing powers of S. Symeon on his pillar (Syriac Vita Sim. 79). Bowersock in Archologie,
68, draws attention to a fleeting reference to harvesters working southeast of Palmyra; for the
Palmyrene inscription, see Teixidor, Syria 40 (1963):
3337, esp. 3435.
181. Theophyl., Hist. 5.1.7: Sergius "whom the nomad
tribes are also wont to revere."
182. For discussion of these migration routes, see G.
Fowden in Roman and Byzantine Near East, 2:
12127.

Page 99
detail by the Passio's author lends credibility to his
local knowledge, at least at the time of composition in
the fifth century. Rusafa's place in this landscape at
the convergence of wadis north of the main pass, suggests that it may also have served as a sanctuary for
animals, who would have been protected here by the
saint from slaughter during their seasonal migrations.183
These physical conditions of Rusafa's setting combined to make it a natural site for a harama hallowed space carefully demarcated and maintained,
usually by a particular tribe for whom the privilege was
hereditary.184 A haram might be considered sacred
by association with a holy tree or spring, but most often it was centered on the tomb of a holy man, a saint.
Usually it occupied a strategic position, most commonly at a source of water near a crossroads.185 It
was a natural magnet for the region and usually the
site of a market. The haram was also understood to
be neutral ground where conflicting parties could meet
and seek resolution of their disagreement. And it was
often an asylum for animals. The tomb might be openair or covered with a simple building. The authority of

339/754

the saint was the backdrop for all activity that transpired at the site: judgments, business deals, social
agreements. The haram was not simply a saint's
tomb; it was a sacrosanct site, distinguished by the
saint's presence, where worldly leaders exercised
their authority under the saint's protection.
It is clear that the pilgrims who congregated at Rusafa
to stand in the saint's presence cannot easily be
typecast. The faint traces that they have left behind
convey at least an outline of the people who sought
the martyr's helpbut also of the saintly company
Sergius kept. Among the graffiti that anticipant pilgrims carved on the walls of the northern annex, while
waiting just outside Sergius's martyrium, Syriac and
Arabic messages seem to have been mixed in with
Greek. Unfortunately, the non-Greek graffiti are illegible today, with the exception of two late Arabic inscriptions.186 The
183. In RE, 2A: 168485, Honigmann notes the emphasis given in the Passio to Sergius's power over animals, an attribute the Christian saint shared with the
Arab god . For other similarities between these two,
see above, pp. 4042. It would not have been unusual

340/754

if animals were also brought to the saint for healing,

as at the shrine of S. Thecla (Mir. Theclae 36). Cf.
also Evagr., HE. 1.14, where it is reported that pilgrims would circumambulate the column of S. Symeon with their beasts of burden. In the medieval
Greek epic Digenis Akritis (Grotta-ferrata 3.141),
Mecca is evoked as a sanctuary for animals, as well
as pilgrims.
184. Donner, Early Islamic Conquests, 3437; Serjeant, Studies in Arabian History III. It is a matter of
ongoing debate just how far back the haram tradition
reaches, although few would doubt the existence of
such a pattern in the pre-Islamic steppe and desert.
185. In the modern period, beduin gathered in the
steppe year after year at the same site. A well or
cistern, caves or ancient buildings usually marked the
frequented spot, which would be used while they
waited for the next seasonal signs for migration: see,
e.g., Lewis, Nomads and Settlers, 12729.
186. Khoury in Resafa, 2: 17981, publishes the Arabic inscriptions. Syriac graffiti noted in AAS 33.2
(1983): 70, were not published in Resafa, vol. 2. On
the important differ-

Page 100
name Sergius is recorded as Sergios and Sergis.187
Some of the names carved in Greek on the wall are
clearly of Semitic origin: Abbibas, Salmis,
Marousa.188 The frequency of the name Symeon
may reflect the popularity of the famous Syrian saint,
who, according to the accounts of his life, was especially beloved by Arab pastoralists.189 Not all of the
names belonged to the pilgrims themselves. Some
seem to evoke saintssuch as John and Thecla,
whose shrines, along with Sergius's, had been
numbered by John Moschus among the East's
greatest. The martyr Sergius was anything but isolated in his Barbarian Plain, where Arab raiders
threatened caravan trade and faint-hearted pilgrims.
His shrine had grown over the course of two and a
half centuries into a stable center within a complex
network of routes, which joined often dissimilar regions both within and without Syria and Mesopotamia.

ences in the uses to which various scripts found in

Syria were put, see the excellent article by Macdonald, Syria 70 (1993): 30388, esp. 38288.

342/754

187. Rmer in Resafa, 2: 173, no. 17; 175, no. 41. On

Arabic and Syriac forms of Greek Sergios, see Gero
in Syrie, 51 n. 24.
188. Rmer in Resafa, 2: 17177. See Gatier, Inscriptions, 2: 69, 97; Sartre, Bostra, 215, 236; Wuthnow,
Semitischen
Menschennamen,
7,
103;
with
Macdonald's cautionary discussion of the difficulties
involved in distinguishing different Semitic names
from a consonantal skeleton and identifying that skeleton with a Greek version (Syria 70 [1993]: 37879).
189. Rmer in Resafa, 2: 172; Vita Sim. Styl. (Syriac)
67, 77; Theod., HR 13,16,18, with Canivet's
comments.

Page 101

FOUR
The Spread of the Sergius Cult in Syria
and Mesopotamia
Abba Sergius, a sixth-century anchorite in Sinai, was
traveling with a party of mules and muleteers when
they came face to face with a lion. Before the lion had
an opportunity to seize its prey, Abba Sergius stepped
forward, held out a piece of bread, an eulogion, and
ordered the beast to move off the track. Sated by this
substitute, the lion allowed the party to carry on its
way.1 Lions, raids, heat, and cold were the obstacles
that travelers encountered along waterless tracks, not
only in the Sinai wilderness, but across Syria and
Mesopotamia too.2 Shelters from these threats, such
as the monastery of Saint Catherine in Sinai, were
bolstered by saints' relics. At shrines, travelers would
pray for the saints' guidance and protection. Many
Christian travelers carried with them from their baptism the name of a saint: "and to their children they
take care to give the names of these [martyrs], engineering for them in this way security and protection."3

344/754

The dramatic rise of Sergius's popularity, apparent in

personal names and church dedications, reflects the
sixth-century growth of belief in Sergius's power as
protector.
From the early sixth century, the name Sergius was
commonly borne by men of religious vocation. A deacon Sergius from Amida oversaw Anastasius's fortification of Dara.4 A presbyter named Sergius co-built a
church dedicated to S. George at
, west of Maximianopolis/Shaqqa in the
1. Joh. Mosch., Prat. spir. 125.
2. On the threat of raids in the Sinai peninsula, see
Gatier in L'Arabie prislamique, 499523. Lions still
posed a threat to inhabitants of and travelers in the
middle Euphrates region in the last century. Blunt, Bedouins of the Euphrates, 1: 8993, describes an attack in 1878.
3. Theod., Graec. aff. cur. 8.334.
4. Chronicle of Amida of 569 7.6.

Page 102
Hawran.5 East of Emesa, the area of Salaminias was
served by a periodeute and also by a chorepiscopus,
both named Sergius.6 Further east and a century
later, Rabban Sargis, "destroyer of the mighty," recorded the history of the Nestorian holy men of Beth
Garmai, the central Mesopotamian region north and
east of Takrit.7 Sergii also rose to sit on patriarchal
thrones: in Antioch from 557 to 560 and in Constantinople under Heraclius (Sergius I, 61038); while
Sergius of Rome, who reigned between 687 and 701,
was one of five Syrians who took their place on
Peter's throne between 686 and 730. It was this Pope
Sergius who introduced to Rome the cult of the Holy
Cross.8
By the early sixth century, bearers of the name Sergius had risen to high positions outside ecclesiastical
circles too. The Prosopography of the Later Roman
Empire reveals a dramatic increase in the name's attestation among higher-ranking officials between the
late fifth and the early seventh centuries: from two
Sergii between 260 and 395 to nine between 395 and
527, and to fifty-five attested between 527 and 641.
Since the name Sergius was already attached to a

346/754

famous Roman family in the Republican period, not

every Sergius necessarily owed his appellation to the
saint. This seems to be the case with the earliest attestations in the Prosopography.9 But the great popularity of the name in the East, attained over a short
period of time, strongly associates the increase with
the growth of the Sergius cult. A debt to the SyroMesopotamian saint is certain in some cases, such as
Sergius, son of Bacchus, who from his birthplace near
Dara became a governor of Tripolitania and by 559
had been made patricius.10 The family of another
prominent Sergius came from Rusafathe father of
this Sergius had probably become a vir illustris at
Edessa, where in 590 he vied with Edessa's elite in
banqueting the fugitive Khusrau II. The son Sergius
was later taken hostage by the same Iranian despot,
but spent his time among Khusrau's inner circle until
he returned to Edessa, most likely after the truce with
Heraclius in 62829.11 In the early to mid sixth century, Sergius of Rhesaina, also in northern SyriaMesopotamia, rose to fame as a medical doctor and
theologian.12 Another Sergius, probably also a Syrian, was prized by Khusrau I as

347/754

5. Waddington, Inscriptions, no. 2092. Sartre, IGLS

13.197 n. 3, mistakenly includes this church in a list of
dedications to Sergius in the Hawran.
6. IGLS, no. 2517, for the periodeute; Lucas, in von
Oppenheim and Lucas, BZ 14 (1905): 29, for the
chorepiscopus, whose inscription is dated to 564.
7. Thomas of Marga, Book of Governors 1.33.
8. Bishop, Liturgica historica, 14446.
9. E.g., PLRE, 1: 826. Like any prosopography, the
Prosopography of the Later Roman Empire must be
used with discretion: see Barnish, JRS 84 (1994):
17177, esp. 175.
10. PLRE, 3: 112428.
11. Ibid., 681, for the father, Ioannes 102, and 1133,
for Sergius 38.
12. Ibid., 112334, Sergius 1.

Page 103
an interpreter and relied on by Agathias as a translator of Iranian royal archives.13 Less glamorous jobs
were held by other Syrian Sergii one was a baker
at Constantia /
east of Edessa; another was a
stone cutter at Maximianopolis / Shaqqa in the
Hawran.14 On a church floor at Dionysias / Suwayda,
south of Maximianopolis, a full-length mosaic portrait
commemorates a Sergius, probably an important local
donor.15 In various oriental Christian versions of the
Bahira legend, the name Sergius, or Sergius-Bahira,
is given to the famous monk who discerns in the
young Muhammad his as yet unrevealed identity as
the Prophet.16 Sergius, or rather, Sarjis, was a popular name among Arab Christians: "I was a Christian
called Sarjis, the surest and best guide in the sandy
desert," explains an early follower of Muhammad in
Ibn Ishaq's Life of the Prophet. "During the pagan
period [jahiliya] I used to bury water which I had put in
ostrich shells in various places in the desert and then
raid men's camels. When I had got them into the
sand, I was safely in possession of them and none
dared follow me thither. Then I would go to the places
where I had concealed the water and drink it."17

349/754

A saint's name could be grafted onto a family tree and

reappear generation after generation, as once purely
secular names had. The name's original association
might become dissipated through force of habit. But
name-giving in late antique Syria and Mesopotamia
must be seen in the context of the saints' annual festivals. No evidence could prove that each Sergius
thought of the martyr Sergius as his personal patron
simply because his parents had chosen to give him
the name. Nonetheless, the calendar of religious
feasts that punctuated late antique conceptions of
time forced on the bearers of such an important
saint's name a consciousness of their name's history.
Names, especially saints' names, carried powerful associations, and popular names in particular should be
read as indicators of cultural trends. It was not coincidence that joined the widespread adoption of the
name Sergius from the late fifth through the early seventh centuries with other
13. Ibid., 1129, Sergius 9. Cameron, DOP 2324
(196970): 162, discusses the Syrian origin of Sergius; see Agathias, Hist. 4.30.284300; also PLRE, 3:
1124, Sergius 3, an Edessene envoy to Iran in 543,
54445.

350/754

14. Lucas, in von Oppenheim and Lucas, BZ 14

(1905): 61, no. 96, for the
; Waddington, Inscriptions, no. 2162, for the stone carver.
15. The figure has been variously interpreted: see
Donceel-Vote, Pavements, 31013, with fig. 305.
This selection attempts to represent the variety, but by
no means the quantity, of Syrians and Mesopotamians named Sergius in late antiquity. Clerics named
Sergius appear frequently in the Synodicon orientale.
See also the examples cited by Poidebard and Mouterde, AB 67 (1949): 11314. The female form Sergia
is also recorded. A Sergia wrote De translatione
Sanctae Olympiadis, and another is mentioned in a
dedication in the so-called Riha treasure (Mundell
Mango, Silver, no. 35).
16. For the oriental sources, see Gero in Syrie,
4758.
17. Ibn Ishaq, Sirat 985, p. 668 (tr. Guillaume).

Page 104
signs of the growth of the martyr Sergius's reputation,
such as building projects at Rusafa and the spread of
sanctuary dedications beyond the cult center. That a
name could reflect a political and religious climate is
clearly illustrated by the frequent adoption of names
such as
, Yazid, and Walid in Umayyad Syria.
The choice of these names expressed engagement
with the current political and religious situation. The
Umayyad caliphate offered a new set of names that
stood in deliberate contrast to names such as
,
Hasan, or Husayn, which were seldom given in Syria,
but common in Egypt.18
Sergius's name, whether simply chosen at baptism, or
carried on the lips of pilgrims, was the most mobile
accessory in the martyr's cult. Its use in dedications
must serve as tracing dye in any attempt to recapture
the cult's spread. A bird's-eye view of Syro-Mesopotamian dedications to Sergius at the time of the erection of basilica B at Rusafa in 518 describes a great
arc, from the Hawran southwest of Rusafa through the
villages east of the Anti-Lebanon and continuing along
the southern foothills of the Taurus mountains. At that
time, starting in the southwest and traveling clockwise

352/754

among dated inscriptions only,19 Umm al-Surab had

a church of Sergius and Bacchus from 489, while
Bostra's cathedral honored Sergius, Bacchus, and Leontius in 512 / 13. In northeastern Hawran, a well-preserved inscription records a martyrium of Sergius in
517 at Raham,20 and at the same date, on the southern edge of the rocky basalt Laja, a temple to Sergius
is recorded at Busr al-Hariri. Moving further northward, Severus of Antioch preached at the Sergius
church of Chalcis in 514, while Zabad in the basalt
country southeast of Chalcis honored Sergius with a
martyrium in 512. West of Edessa, near modern Nizip,
a Sergius church was consecrated in the village of
as early as 431. No evidence of dedications east
of Edessa before 518 has been discovered, but the
eastern arm of this arc of dedications would begin to
take form by the middle of the century.
What follows is an attempt to evoke from a selection
of dedications in late antique Syria and Mesopotamia
a picture of where the cult of Sergius was spread, by
whom, and for what reasons. A striking characteristic
of the Sergius cult that will emerge from this overview
is the cult's adaptability to the distinctive Syro-Mesopotamian landscape and the various populations

353/754

inhabiting it. Springing from Rusafa, the cult was

woven into pastoral migration and trade patterns, defense organization, and the fabric of monastic and village life. The densest cluster of surviving dedications
from the late fifth and early sixth centuries is found in
the Hawran and the zone
18. Gilliot, Arabica 40 (1993): 35758.
19. Bibliography appears with the discussion of individual sites, with the exception of Raham, which is
only mentioned here.
20. On Raham, see Dunand in Mlanges syriens,
571, no. 290.

Page 105
from Salaminias to Chalcis. In these areas, as we saw
in chapter 3, agricultural and pastoral societies overlapped, and both regions were touched by muchtraveled trade routes.21 One of the great nodes along
the routes in Syria was the city of Damascus. On the
decumanus, the famous "Street called Straight," a
shrine ( ) dedicated to Rusafa's martyr had already
been established before the fifth- or sixth-century
widening of the road that absorbed part of the
shrine.22

THE HAWRAN
South of Damascus and its oasis, the Ghuta, the
Hawran plain provided an environment suitable for
both pastoralism and agriculture.23 At the village of
Eitha / Hit, northwest of Maximianopolis, the deacon
and oikonomos Sabinianos built a church dedicated to
Sergius and dated it by reference to a trio of local clergymen: an archimandrite who was also a priest, another priest, and a deacon.24 The presence of an
archimandrite suggests that the church belonged to a
monastery.25 What is most intriguing about the inscription is its date, which, according to Waddington,

355/754

reads
(March 249). Waddington assumed that the date is to be reckoned according to
the calendar of Provincia Arabia and arrived at a date
of 2231 March 354 or 121 March 355 of the Christian era, making the Sergius sanctuary at Eitha one of
the earliest-known examples of a building dedicated
to the memory of any saint at a site other than that of
his or her martyrdom. In this case, it would have been
only four decades after Sergius's death under the
tetrarchy, according to the Passio tradition. The date
354 has been almost universally accepted, usually
without any refer21. Donner in Tradition and Innovation, 7385, and
Conquests, 1620, esp. 16 n. 5, provides a succinct
discussion of social, economic and political interrelations of settled, seminomadic, and nomadic
populations.
22. The shrine is mentioned in the Vita of S. Stephen
of the Saba monastery, the eighth-century hymnographer and nephew of S. John of Damascus, Acta
Sanct. Jul. 3.555 61. See Sack, Damaskus, 16, with
a plan of late Roman Damascus in fig. 5. Will, Syria
71 (1994): 16 with n. 38, considers the shrine, known

356/754

as the Maqsallat al-Baris, in the context of the city

plan of Damascus.
23. Villeneuve, Hauran, 1: 12125; Bauzou in Hauran,
1: 158. For a general account of Hawranian architecture, see Farioli Campanati, Felix Ravenna 14144
(199192): 177232.
24. Waddington, Inscriptions, no. 2124. I am indebted
to Kalliope Kritikakou for discussing the problems of
Syrian calendars with me, with special reference to
Eitha/Hit. I would also like to thank Maurice Sartre for
his comments on my interpretation of the Eitha inscriptions and for sending his most recent view, which
is in agreement with my own.
25. Sartre, IGLS 13.197 n. 3, identifies the church as
monastic, presumably because of the inclusion of an
archimandrite in the inscription. The inscription does
not allude to a monastery or the type of sanctuary
dedicated, but employs the general term
.

Page 106
ence to its uncertainty.26 In 1982, Maurice Sartre
consulted the stone again and read the date not as
CM, as had Waddington, but as CME, with the result
that the date, if reckoned according to the era of Provincia Arabia, would be March 35051, instead of
Waddington's 35455.27 Still, either of these dates in
the 350s would be the second-earliest attestation of
the use of Roman months with the chronological system of Provincia Arabia. The earliest was found southeast of Bostra at Imtan, the site of a Roman fort and
explicable by that association.28 No other inscription
from Eitha uses the era of Provincia Arabia,29 and
also unusual for the mid fourth century is the accumulation of clerical names. The earliest-known evidence
of a bishop at either Maximianopolis or Philippopolis /
Shahba is their attendance at the Council of
Chalcedon in 451but lack of earlier evidence does
not necessarily mean lack of a bishop. Nevertheless,
a date according to the era of Provincia Arabia would
place Eitha at the forefront of various trends, not only
reception of the Sergius cult, but also the honoring of
any martyr with a church, and the use of Roman
months and ecclesiastical titles in inscriptions. This

358/754

accumulation of irregularities certainly seems to demand some explanation.

It is not out of the question that the cult of Rusafa's
martyr had begun to spread by the mid fourth century
and that the earliest surviving material evidence
comes from Eitha, located between the Laja and the
Jabal Hawran near the two main roads linking Damascus and Bostra. But as Honigmann pointed out in a
review of Robert Devreesse's Le Patriarcat d'Antioch
in 1947, a calculation of the inscription's date according to the local era of Maximianopolis would place the
dedication in the mid sixth century, where it would correspond to the known development of martyr cult in
general, and that of the veneration of Sergius in particular.30 The era of Maximianopolis begins either in
287 or 302, which would mean that the Sergius
church at
26. To their credit, Poidebard and Mouterde, AB 67
(1949): 112 n. 8, note that the date is suspect. Meimaris with Kritikakou and Bougia, Chronological Systems, 154, acknowledge the unusual features of the
early date, stating that the alternative to the era of
Provincia Arabia would be local era; but they conclude

359/754

in favor of 354. Before revising his reading of the inscription, Sartre, Bostra, 121 n. 184, took the early
date as certain : "Pour ne parler que de celles o une
inscription permet d'assurer la date, mentionnons
l'glise Saint-Serge de Hit, vers 354 au plus tard."
Trombley, Hellenic Religion, 2: 34344, 37374, who
also accepts 354, is silent about the problems surrounding the date.
27. Sartre, letter to the author, 30 January 1997.
28. Waddington, Inscriptions, no. 2037; Meimaris with
Kritikakou and Bougia, Chronological Systems, 154,
190.
29. Sartre underlines this point, noting that other inscriptions at Hit employ royal or imperial dating systems (ibid.).
30. Honigmann, Traditio 5 (1947): 15960.

Page 107
Eitha was dedicated in either 536 / 37 or 551 / 52,
dates coinciding with the period of greatest building
activity in honor of Sergius in the Hawran, according
to known inscriptions.31 It was a time when Roman
months were commonly in use in inscriptions, as well
as the clerical titles that appear in the Eitha dedication. In the light of other inscriptions from Maximianopolis, we might expect an inscription dated by the local era to be marked by
, which is lacking
from the Eitha inscription. But the use of this phrase is
by no means systematic.32 No other inscriptions have
been discovered outside the city of Maximianopolis
that are dated by that city's era. However, Eitha's
proximity to Maximianopolis, only a few kilometers to
the southeast, makes the possibility of attraction to
that city's era very likely. The historical context is
weighted heavily against the early date. The argument
for a sixth-century date according to the Maximianopolis calendar would be further strengthened by new
evidence, such as whether or not Eitha belonged to
the territorium or ecclesiastical jurisdiction of Maximianopolis. Nonetheless, it should be clear that no arguments for the precocious spread of the Sergius cult in
the Hawran can be based on the evidence from Eitha.

361/754

Situated on the road between Mushannaf and Bostra,

the village of Kafr boasted several churches. In 1939,
Maurice Dunand published an inscription whose damaged second line begins with space for two or three
letters followed by PI, and could allude to the popular martyr George just as well as Sergius. Dunand did
not note the possibility of reading George, and his
has been accepted without comment in subsequent publications.33 However, already at the time
of Dunand's publication another
31. Meimaris with Kritikakou and Bougia, Chronological Systems, 32123. Koder and Restle, JB 42
(1992): 7981, are unaware of the solid arguments for
287 or 302 for the beginning of the era of Shaqqa put
forward by Jones, Cities, 465 n. 82, and discussed at
length by Robert, Hellenica 1112 (1960): 30615.
The problem has been discussed more recently in
Meimaris with Kritikakou and Bougia, Chronological
Systems. Koder and Restle, who date the beginning
of the era of Shaqqa to 286, do not in their brief discussion even raise the vital issue of which calendar
was used in the inscriptions dated by the era of
Shaqqa. The crux of the problem for the Eitha inscription (if we accept that it employs the era of Shaqqa,

362/754

as I believe we should), as for all inscriptions dated by

the era of Shaqqa, is our ignorance of which calendar
was usedeither the Graeco-Arabic, which began on
22 March, or the Julian, which began on 1 January.
For this reason, the question of whether the era of
Shaqqa began in 286 or 287 must still remain open.
32. I thank both Glen Bowersock and Maurice Sartre
for insisting on this point. Sartre observes that in
neighboring villages, such as
and Hayat, the use
of the provincial era is made explicit, in order, it
seems, to draw attention to the exceptional use of that
era. The implication of this practice is that usually a
local era would be expected.
33. Dunand in Mlanges syriens, 559, no. 245 with
drawing; followed by Poidebard and Mouterde, AB 67
(1949): 112; Devreesse, Patriarcat, 231; Sartre, IGLS
13.197 n. 3.

Page 108
inscription was known that attests the dedication of a
church to S. George at Kafr in 652. Unless the village
possessed two churches dedicated to S. George,
Dunand's interpretation was justified.34 The inscription published by Dunand attracted the attention of
Robert and Aigrain for its unusual use of
for the
building dedicated to Sergius.35 Aigrain explains the
rare usage as another instance of Christian absorption of a pre-Christian name of a holy place, such as
and
all of which are likewise attested
in the Hawran in the context of the Sergius cult.36
Other names used to mark Sergius dedications in
Syria-Mesopotamia are
and
.
In the foothills of Jabal Hawran, two monasteries were
dedicated to Sergius. The first inscription was found in
situ over a door in the ruined buildings of Dayr alQadi, west of Suwayda.37 The undated metrical dedication commemorates in archaizing style renovations
made
to
the
sanctuary
of
Sergius:
. East of Salkhad, at Dayr alNasrani, another lintel, this one found on the ground
near the eastern entrance of the monastery courtyard,
preserves in a partially illegible inscription a prayer to

364/754

the "God of Sergius and Bacchus."38 Seen in conjunction with the three crosses etched across an inscription that
(that is
, servant of Baal)
carved over the door of a tower later incorporated into
the monastery, the Christian prayer proclaims forthrightly that the God of Sergius and Bacchus has supplanted the region's older god, Baal.39
South of Dayr al-Nasrani, another dedication was
made to the pair Sergius and Bacchus. Umm alSurab, on the road leading from Bostra to Gerasa,
twelve kilometers northwest of Umm al-Jimal, preserves the basalt remnants of a prosperous late antique village. The dedicatory inscription, carved on a
tabula ansata flanked by two damaged crosses, still
stands over
34. Brnnow and Domaszewski, Provincia Arabia, 3:
360.
35. Robert, Bull. (1940): 232, no. 189; Aigrain, OCP
13 (1947): 2224. Both Robert and Aigrain accept
Dunand's reading.
36. In general, Christian use of
is only rarely attested in the Hawran, although it is common in Syria

365/754

and Palestine for Christian and Jewish tombs. More

common in the Hawran are
and
, the latter
attested twenty-eight times (Sartre, Bostra, 3839).
37. Waddington, Inscriptions, no. 2412.
38. PAES, 3A5: 33334, no. 722, with drawing, where
Littmann reads the dedication to the "God of Sergius
and Bacchus" in the first line and the first half of the
second line. Only the first two letters of the first line
were legible to Dussaud and Macler, Nouvelles
archives 10 (1902), no. 48, with drawing. Littmann's
reading also diverges significantly from that of Dussaud and Macler. For one of many examples of this
phrase, see Mundell Mango, Silver, 256, no. 85: "The
God of S. George"; and for a general discussion, see
Peterson,
, 21012.
39. PAES, 3A5: 334, no. 723; PAES, 2A5: 33435, on
the tower's incorporation and the possibility that the
site was formerly a polytheist high place.

Page 109
the western entrance of Umm al-Surab's largest
church.40 The benefactors were Ameras and Kyros,
sons of Ulpianos, although Littmann's restoration of
is conjectural.41 Littmann's suggested reading
of the date as 489 is accepted in the recent publications of Sartre and King.42 The ruins at Umm alSurab include two undated churches in addition to the
Sergius and Bacchus church, and many examples of
domestic architecture, much of which has been modified over centuries of habitation, which continues
today.43 Howard Crosby Butler described the basalt
church of Sergius and Bacchus, which he thought it
one of the most interesting in the southern Hawran, as
"medium-size."44 In form, it is a three-aisled basilica
with two rows of five columns, which supported a gallery. Three doors in the western faade opened onto
the aisles, the faade was sheltered by a portico, and
the main entrance was flanked on either side by a
kolymbion, a basin for blessed water. Above the door
was the inscription. The dreary effect of so much
basalt was relieved by painting and mosaic, traces of
which have been found in the church.45 To the north
and south of the church extend courtyards surrounded
by small buildings. Butler is silent about the southern

367/754

courtyard, but interpreted the northern complex as

"the complete domestic establishment of an ecclesiastical community." However, King has recently called
into question the architectural and chronological unity
of the northern structures and their relation to the
church.46 Both Butler and King agree that the church
of Sergius and Bacchus was perhaps already converted into a mosque and fitted with a minaret under the
Umayyads.47
Moving west of Bostra, a cluster of villages allied
themselves with Rusafa's martyr in the last decade of
the sixth century. Ghasm, on the road between Bostra
and
, commemorated Sergius and Bacchus together in a
40. PAES, 3A2: 5758, no. 51, with drawing. The
building's dedication to SS. Sergius and Bacchus is
clear on the inscription, but the beginning of 1.4,
where the type of building was named, is damaged
and only an initial can be read. Littmann restored
the word as
. See above for the frequency of this
term in the Hawran. For a discussion of the architecture with reconstructions, see PAES, 2A2: 9499.

368/754

41. Sartre, Bostra, 17475, endorses Littmann's reading without mentioning that it is a conjecture, and considers it a unique variant of the Semitic name Amerus,
especially popular in the Hawran. Trombley, Hellenic
Religion, 2: 32122, also generalizes from the inscription without qualification.
42. Sartre, Bostra, 197 n. 3; King, DaM 1 (1983): 112,
116; King, DaM 3 (1988): 47, 50.
43. On the two churches, see King, DaM 3 (1988):
4051. Butler, PAES, 2A2: 95, describes Nabataean
and early Roman epigraphical and architectural fragments from Umm al-Surab, whose remains are
primarily late antique.
44. PAES, 2A2: 96.
45. King, DaM 1 (1983): 117, 12526.
46. PAES, 2A2: 99; King, DaM 1 (1983): 124.
47. King, DaM 1 (1983): 13336.

Page 110
, where relics of the saints were perhaps
revered.48 The inscription, found reused as part of a
manger, dates the original building to 593. Straddling
the Wadi al-Zaydi, the village of Jiza has produced
two dedicatory inscriptions: one for a church of
Theodore, and one for Sergiusthe word used in
both cases is
.49 The Sergius consecration is
dated to 590. At Tayyiba, where a bridge dated to the
reign of Marcus Aurelius carried the
road
across the Wadi al-Zaydi, a church ( ) was built in
589 to 590, possibly to Sergius, but damage to the
inscription's first line makes that reading conjectural.50 And a fourth, but undated, dedication was recorded in this small areathis one called a martyrium of
Sergius, at the non-Chalcedonian monastery of
.51
A remarkable lintel inscription found in Zorava /
,
at the southern limit of the Laja, was published with
commentary by Claude Mondsert in 1960.52 Discovered in situ, having been missed by earlier scholars, its fourteen lines tell with striking poetic style the
story of the dedicant's devotion to Sergius and that
martyr's passion and victory:

370/754

And now, seeing the power of the savior, master, God,

glorify the holy king, who has destroyed the works of idols.
For this house was once adorned with images of demons
and bound by rough stones, which the logos of Christ
has freed and has reestablished, with finely polished stone,
the house of his servant, the well-mounted Sergius,
through the zeal and efforts of the children of noble Theodore,
who desired to have Sergius himself as their divine defender,
he who spurned worldly authority,
and accepted bitter tortures from head to foot.
Although his feet were pierced with nails, he did not spare h
but having given his spirit unto his master and savior, he delivered it up
to death,
and in exchange for a worldly life, received his portion of celestial life.

371/754

The inscription falls into three interrelated parts. The

first six lines allude to the erection of a church on the
site of a polytheist temple. In this category it joins the
better known church of S. George also at Zorava,
where Christians still worship today, as well as another church of S. Sergius, also on
48. PAES, 3A5: 27879, no. 619.
49. Savignac and Abel, RB 2 (1905): 59798.
50. Dussaud and Macler, Nouvelles archives 10
(1902): 693, no. 155, for the possible Sergius inscription, no. 154 for the bridge. The drawing of the church
dedication shows CEB, whereas Dussaud and Macler
read
.
51. In the so-called "Letter of the Archimandrites," in
Documenta 218 (tr. Chabot, 15152). For other Sergius monasteries mentioned in the letter, as well as
general comments on the Documenta, see below, pp.
143n.45, 144.
52. Mondsert, Syria 37 (1960): 12530.

Page 111
the edge of the Laja but to the east, at Busr al-Hariri.53 The verbal image of Sergius as a horseman is a
welcome partner to the visual representation of the
rider saint found on the metal medallions, the seal,
and the stone mold discussed earlier. This image is
immediately followed in lines seven and eight by the
church's benefactorsthe children of Theodorewho
chose the cavalier saint expressly for his power as a
divine defender.54
What follows in the last six lines is a distilled version
of Sergius's martyrdom along the lines of the Passio,
punctuated by carefully weighted contrasts, as elsewhere in the inscription: Sergius renounces earthly
power; he undergoes ruthless punishment with nails
in his feet; he surrenders his body to be killed, but receives his heavenly reward.55 This inscription bears
witness to the circulation in the Hawran of the story as
told in the written Passio. Since the Passio has been
shown in chapter 1 to date from after the 430s, the
erection of the otherwise undated church must belong
to the late fifth or sixth century, the most intense
phase of the cult's dissemination.

373/754

Between September 512 and March 513, Archbishop

Julian of Bostra consecrated a church (
) near
the Arabian capital's centera tetraconch in form with
echoes of Mesopotamian architectural style.56 The
church bore a unique dedicationto SS. Sergius,
Bacchus, and Leontius. The soldier who joined Sergius and Bacchus was martyred at Phoenician Tripolis
(modern Tripoli in Lebanon), possibly in the reign of
Trajan.57 Two churches in the Hawran were dedicated to Leontius: one in 483 at Dur, southeast of Busr
al-Hariri, and the other in 565 at Sur in Auranitis.58
Also, Justinian built a church (
) dedicated to Leontius in Damascus.59 Known dedications to Leontius
were concentrated in southern Syria and the coastal
city of Tripolis. The period of the Bostra church's construction and consecration was fraught with ecclesiastical controversy in this city.
53. For the
church of George, see Waddington, Inscriptions, no. 2498, where the dedication describes
the substitution of choirs of angels for the sacrifices to
idols at the old shrine. For Busr al-Hariri, see ibid., no.
2477, where it is persuasively suggested that
refers to a polytheist shrine.

374/754

54. Mondsert, Syria 37 (1960): 129, suggests that

Theodore may have been a philosopher or rhetor and
the children his disciples, although he admits that no
external evidence supports this interpretation.
55. See Mondsert's comments on the rhythm and
rhetoric of the inscription in ibid., 12728.
56. IGLS, no. 9125. See Sartre, Bostra, 12426, with
nn. 215, 216, for studies of the church, which has yet
to be published systematically. On its architectural
style, see Lassus, Sanctuaires, 9596, and Farioli
Campanati, Felix Ravenna 14144 (199192), esp.
22224.
57. After studying law at Beirut, Severus of Antioch received the sacrament of baptism in the church of Leontius at Tripolis and later composed a hymn to the
martyr (Evagr., HE 3.33; James of Edessa, Hymns of
Severus of Antioch, 138).
58. For Dur, see Waddington, Inscriptions, no. 2412p;
for Sur, see PAES, 3A7: 426, no. 7973.
59. Proc., Aed. 5.9.26.

Page 112
Bishop Julian opposed Severus's election as patriarch
of Antioch in November 512, which eventually cost
him his see.60 According to John Moschus, who heroizes the opponent of Severus, before Julian's exile in
513, several Bostran "enemies of Christ," that is, nonChalcedonians, bribed an attendant to poison the
archbishop. But, informed by divine revelation that he
would remain unharmed, Julian humiliated his wouldbe assassins by drinking the tainted cup before their
eyes.61 By the early sixth century, Sergius was acquiring a history of mixed associations, with devotees
among both Chalcedonians and non-Chalcedonians.
Severus of Antioch wrote three hymns in honor of
Sergius, one with Bacchus, and in 514, the year after
Julian consecrated Bostra's cathedral, Severus
preached a sermon on Sergius at Chalcis.62 In the
same climate, Jacob of Sarug (ca. 451521) composed his metrical homily on Sergius, and the emperor Anastasius openly declared his support for the nonChalcedonian cause and for the martyr Sergius of
Rusafa.

376/754

WEST OF THE STEPPE

Leaving the Hawran, we travel northward to Salaminias at the southern base of another basalt landscape.
Situated northeast of Emesa, on the main route to the
important settlement of Seriane / Isriya, Salaminias is
set in the fertile zone of hills and depressions that
reaches as far north as Chalcis, but to the east and
south is bordered by an arid landscape. The town was
integrated into the network of routes that linked the
many villages between Emesa and Chalcis, as well as
the east-west route in the direction of Rusafa.63 An
inscription found near the city gate commemorates
local patronage of Salaminias's fortification wall:
This is the gate of the Lord, the righteous shall enter through it.
The noble namesake of the glorious martyr Sergius,
made it secure by restoring the entire citadel,
this stronghold that was the salvation of many mortals,
and has won him eternal memory.64
60. On Julian's opposition to Severus, see Evagr., HE
3.33. See Sartre, Bostra, 10913, for an account of

377/754

Bostra's ecclesiastical history in the early sixth

century.
61. Joh. Mosch., Prat.spir. 94.
62. James of Edessa, Hymns of Severus of Antioch,
14345.
63. For a description of Salaminias's setting, see
Waddington, Inscriptions, in his discussion of no.
2633. On Salaminias's place in this network of communications, especially with regard to Rusafa, see
Mouterde and Poidebard, Limes, 13947; for the
Emesa-Salaminias-Anasartha-Barbalissus route, see
17778.
64. IGLS, no. 2524. The first line is Psalm 117.20.
Trombley's location of Salaminias as "Salamis [sic] on
the riverine plain of Orontes," needs correction
(BMGS 21 [1997]: 173).

Page 113
Unfortunately, this inscription is undated. One of the
few dated inscriptions from Salaminias records the
consecration of a church to the Theotokos in 604.65
Despite its regional importance, little evidence remains to help reknit the town's history. Bishop Julian
of Salaminias was present at the consecration of
Severus as patriarch of Antioch.66 Traveling from Jerusalem to Emesa, the ailing English pilgrim Willibald
spent Lent of 726 at Salaminias.67 Besides the walls
and the church of the Theotokos, Willibald may have
visited the church of "the holy and victorious martyr
Sergius," in honor of whom an inscription attests the
construction of an oratory (
) "from the foundations."68 No date accompanies the dedication, although suggestions include an early date of 43031,
which is not impossible, but arrived at, it seems, by
the association of the Sergius inscription with a dated,
but unrelated, colonnette also discovered at Salaminias.69 A safer date between 550 and 620, when the
cult was especially vigorous, has also been advanced.70
Sergius was also honored at
, a small village
between Epiphania / Hama and Qasr ibn Wardan.71

379/754

The damaged inscription, probably from the lintel of a

main entrance, mentions the door of Sergius and is
dated to 890 or 806 in the Seleucid era, that is, A.D
578 / 79 (Prentice), or A.D. 494 / 95 (Jalabert and
Mouterde).72 To the east at Nawa, one of the largest
late antique settlements in this basalt region, the exterior of an impressive church was decorated with excerpts from the Psalms and the Song of Songs, the
letters rendered in fine, high relief.73 A lintel inscription from another church at Nawa is dated to A.D.
46877.74
After
the
date,
it
reads
. Prentice supposed
to stand
for
and supplied
, so that a priest
named Daniel built a church to the martyr Sergius. But
Jalabert and Mouterde propose an interpretation that
is preferable in the light of nearby parallels. They read
as
and
for , so that an unspecified
church at Nawa was dedicated by Daniel, captain of
the guard of the fortified post of Sergius. Just north of
Nawa, at Tal al-Dahab, an inscription dated 489
65. IGLS, no. 2512; Prentice, Greek and Latin Inscriptions, 237, no. 287, including a photograph.

380/754

66. Devreesse, Patriarcat, 207; Honigmann, Evques,

31.
67. Hugeburc, Life of Willibald, ch. 26, p. 100.213.
68. IGLS, no. 2530.
69. IGLS, no. 2525, with commentary.
70. Hartmann, ZDPV 23 (1901): 110.
71. Dussaud, Topographie, map 8 at 2C. See also
Butler's map of the
opposite PAES, 2B: 1.
72. IGLS, no. 1970; PAES, 3B1: 9, no. 823, with
drawing.
73. IGLS, nos. 194549; Dussaud, Topographie, map
8 at 2D.
74. IGLS, no. 1952; PAES, 3B1: 14, no. 834.

Page 114
records some unnamed construction
Prentice this time interpreted as
.75

, which

Most towns in the region had towers, although the settlements on the whole were unwalled. Where dated,
the towers, like the churches, belong to the sixth century.76 To the southwest of the basalt plateau,
Raphanaea / Rafniya guarded the pass through the
Anti-Lebanon that connected Emesa with Tripolis.
Along with traders and soldiers, the cult of Sergius
took this route and was established at Orthosias, an
important caravan stop just north of Tripolis, where
Peter the Iberian saw a church dedicated to Sergius
and Bacchus.77 At Raphanaea, an undated inscription
on
a
limestone
lintel
identifies
the
.78 In the Nawa and
Raphanaea inscriptions, fortified posts are defended
by the same divine patron. At Ghur, southeast of
Raphanaea, another fortified post (
) was protected by a trio of military saints: Longinus, Theodore,
and George.79 In the same region, at Burj, the tower
that gave the village its present name was marked in
526 as the
of the archangel Michael and Saint
Longinus.80

382/754

In 540, after Khusrau I had come besieging and plundering along the northern route to Antioch, he took a
southward detour on his homeward journey to empty
Apamea of its treasures. No doubt smaller settlements along the way were also exposed to Khusrau's
raiding army. Among these may have been a village
known from Greek inscriptions as Kaper, or Kapro,
Koraon.81 The name was made famous in this century thanks to an impres75. IGLS, no. 1924; PAES, 3B1: 20, no. 850.
76. Butler, in PAES, 2B1: 15, gives a general account of the
and its context.
77. Vita Petri Iberi 109; on Orthosias, also known as
Artousathe name is preserved in the modern Ard
Artousasee Dussaud, Topographie, 7880 and map
14 at 3A; Tabula Peutingeriana 10.3 (Ortosias); Miller,
Itineraria romana, 823.
78. IGLS, no. 1397; Mouterde, MUSJ 28 (194950):
3738, with a barely legible photograph, pl. XV.2.
Raphanaea appears on the Tabula Peutingeriana
10.4 (Raphanis). See Mouterde and Poidebard,
Limes, 3031, on Raphanis.

383/754

79. IGLS, no. 155. Mouterde, MUSJ 12 (1927): 274,

identified the site of Ghur with Carion, placed twelve
Roman miles southeast of Raphanaea on the Tabula
Peutingeriana. This identification was accepted by
Honigmann in RE, 4A: 1674, but has more recently
been reconsidered by Rey-Coquais, who suggests
that a location southwest of Raphanaea is preferable
since the site of Carion was on the route between
Raphanaea and Orthosias: Arados, 8588. See Dussaud, Topographie, map 8, for Rafniy (3A) and
Ghour (3B).
80. IGLS, no. 1610; PAES, 2B: 103; Dussaud, Topographie, map 8 at 3A.
81. Mundell Mango proposes an identification with the
modern village of Kurinsuggesting that Kafr (Arabic
for village) has fallen away to leave only Kurin. For
her hypothetical reconstruction of the history of Kaper
Koraon, see Mundell Mango, Silver, 3334; and on
the Hama region's political and cultural links in late
antiquity, see ibid., passim, but esp. 67. More recently, Fiaccadori has argued in favor of an identification of Kaper Koran with Burj Haydar, a village with

384/754

conspicuous late antique architecture in the central region of the Jabal

(PP 48 [1993] 28791).

Page 115
sive hoard of over fifty silver objects discovered in the
Hama region and now attributed to the village of
"Kaper Koraon"in particular to its church of S. Sergius. The identification of precisely which village was
known in late antiquity as Kaper / Kapro Koraon has
not been made for certain, although it seems most
likely that it was located in the Hama area. The mainly
undated objects have been assigned dates on the
basis of stamps and style to between 540 and 640,
suggesting pious activity in the wake of Khusrau's visit. Many of the liturgical objects declare that their
donors dedicated them to the treasure of S. Sergius,
sometimes also adding "of Kaper / Kapro Koraon."82
A particularly elaborate paten showing Christ celebrating the first Eucharist with his disciples has attracted
attention because it was dedicated to Sergius by Megas, an honorary consul, patricius and curator domus
divinae known from the reign of Maurice.83 Another
impressive gift, consisting of two lamp stands, was
offered by four brothers to Sergius and Bacchus.84
They are described by Marlia Mundell Mango as "the
finest Byzantine lampstands in silver . . . produced by
the most complicated manufacturing methods of any
object in the Kaper Koraon treasure."85 Mundell

386/754

Mango ascribes the joint attribution to Sergius and

Bacchus to coincidence or "simple misunderstanding."
What sort of misunderstanding can, in fact, be identified. The words of the dedicatory inscription are
identical on both objects, although the word order has
been muddled on no. 11 (IGLS 2034 II), where the
opening
has been transposed to the end of the
dedication.86 On both objects, the inscription runs
first along the base of the lampstand and is continued
on a second line directly above the first:
.87
Having vowed, they fulfilled their vow to (the church)
of S. Sergios and Bacchos Sergis (and) Symeon
(and) Daniel (and) Thomas, sons of Maximinos, of the
village of Kapro Korao(n).
82. E.g., Mundell Mango, Silver, no. 3, a chalice from
the early seventh century; no. 4, a paten from the mid
sixth century; no. 7, a cross from the sixth to seventh
centuries; no. 14, a ewer from the mid sixth century;
no. 28, a chalice from the sixth to seventh centuries.

387/754

83. Mundell Mango, Silver, no. 35, and 810 on Megas; also PLRE, 3: 87071, Megas 2; Feissel,
T&MByz 9 (1985): 46576. Megas also dedicated two
ewers to Sergius at Kaper Koraon.
84. Mundell Mango, Silver, nos. 11 and 12.
85. Ibid., 99.
86. Jalabert and Mouterde publish the lampstands as
IGLS, no. 2034.I ( = Mundell Mango no. 12) and no.
2034.II ( = Mundell Mango no. 11). They include the
text of only no. 2034.I and neglect to mention the different word order in their no. 2034.II.
87. IGLS, no. 2034.I.

Page 116
Another item, a ewer, was dedicated to Sergius alone
by five men: Daniel, Sergius, Symeon, Bacchus, and
Thomas, but without additional identification as sons
of Maximinus.88 Comparing the names in the dedications of the lampstands and the ewer, it becomes
clear that we are dealing with the same namesonly
Bacchus is a saint in the lampstand dedication and a
brother on the ewer inscription. What transpired to
produce a dedication to Sergius and Bacchus must
have been the following: the craftsman was given a
written copy of the dedication he was to inscribe, and
after
followed the list of brothers, which
happened to begin with the brother named Bacchus.
The craftsman, who was not from the area, was familiar with Sergius and Bacchus as a pair, made a natural
slip and sanctified the first brother, Bacchus. The result was a unique dedication at Kaper Koraon to Sergius and Bacchus.
Two more building dedications to Sergius appear in
northwestern Syria. At Dar Qita, the ruins of several
churches have been studied, including one consecrated to Sergius in 537, adjoined on the southeast by
a baptistry. Although no date is recorded for the

389/754

baptistry's dedication, it was in need of repair in

56667, when the door was renovated by five priests
named John, Sergius, Danos, Bacchus, and
Ramlys.89 Nineteen years earlier at Babisqa, the next
village to the east, a Syriac inscription records the
purchase of gardens by Sargon, Theodore, and Bacchus, and most likely the construction of a stoa by
John.90 Given the proximity of place and date of the
two inscriptions, it is probable that the same John,
Sergius and Bacchus took part in the construction of
Babisqa's stoa and Dar Qita's church. Babisqa also
had a church, dedicated in 60910, whose lintel inscription
reads:
.
Prentice translates the dedication as "Holy Sergius,
help (us)! Receive the offering of Solomonidas, of the
(family?, tribe?) of Zoryn (or Zaryl)," without commenting on the ethnic background of the name Zoryn or
Zaryl, but assuming, probably rightly, that it refers to a
family group, rather than a region or town.91 This is
the last-known dated Greek inscrip88. Mundell Mango, Silver no. 14 = IGLS, no. 2033.

390/754

89. Prentice, Greek and Latin Inscriptions, 8081, no.

61 (church); no. 62 (baptistry), with Prentice's commentary. PAES, 3B4: 12526, no. 1086 (church), and
no. 1088 (baptistry) with same commentary as in
Prentice, Greek and Latin Inscriptions; map of Dar
Qita opposite PAES, 3B: 119. Cf. IGLS 545 (church),
546 (baptistry). See also Tchalenko, Villages antiques, 2: pl. 134, no. 24; pl. 9, no. 3 (church).
90. See Littman, PAES 4.B: 6465, for the Syriac inscription. For discussion of the Syriac name Babisqa
(*BYT BZQ'), in the light of other toponyms in the regions, see Feissel in
, 29697.
91. Prentice, Greek and Latin Inscriptions, 88 no. 71
(tribe?); Prentice in PAES, 3B4: 13233 (family?). Although the designation
with a proper name is not
commonly found in sixth-or early seventh-century epigraphy, it has parallels in, e.g., a late seventh-century
document

Page 117
tion in northern Syria, and the spelling of the martyr's
name offers an additional variant:
.92

NORTHERN SYRIA AND MESOPOTAMIA

At Sujin, southeast of Beroea/Aleppo, a dedication inscription was discovered worn by reuse in an irrigation
canal. The highly abbreviated text states simply that a
martyrium of Sergius was built in the year 839, that is,
A.D. 527/28.93 Further southeast of Beroea, the famous trilingual inscription of Zabad records in Greek
and Syriac the laying of the first stone of a church
dedicated to Sergius on 24 September 512, the feast
of Thecla, another popular eastern saint. The Greek
and Syriac dedicatory inscriptions flank a disk with a
chi-rho monogram.94 The texts differ slightly, reflecting the liberties taken by the two carvers.95 One
Greek and one Arabic inscription consisting mainly of
names were added along the lintel's lower edge below
the dedications and the central medallion. Adding to
Littmann's reading of the Arabic inscription, Kugener
has shown that the Arabic names correspond to those
in Greek, and represent other, possibly later benefactors of the church, although this point is still unclear.

392/754

Semitic names appear in all four of the inscriptions

and two (perhaps three) of the dedicants recorded in
Arabic bear the saint's name. One is called Sergius,
son of
al-Qays, a familiar name in the history of
Roman-Arab relations in the eastern provinces.96
Arab enthusiasm for Sergius at Zabad in the early
sixth century corresponds to Severus's contemporary
witness to the popularity of the shrine at Rusafa
among the region's Arab population. Between Zabad

from the Nessana papyri, where it clearly refers to a

tribal affiliation (Kraemer, Excavations at Nessana, 3,
no. 92.4045). Although it is never stated, it is perhaps against this background that Trombley, BMGS
21 (1997): 182, 195, has interpreted the dedication to
refer to an Arab devotee of S. Sergius. It is unnecessary to create a female name
, which Trombley
Arabizes to Sulaymana, suggesting that she was
"probably the wife of a sheikh." The inscription clearly
reads
, the genitive of the masculine name
.

393/754

92. IGLS, no. 563; PAES, 3B4: 13233, no. 1100,

with a drawing of the inscription on the lintel of the
western entrance and a plan of Babisqa opposite 128.
93. IGLS 258, with drawing; Robert, Bull. (1940): 228,
no. 172, accepts the IGLS reading; Dussaud, Topographie map 10.
94. IGLS, no. 310, with drawing. Kugener in JA 9
(1907): 50924; RSO 1 (1907): 57786 offers important interpretations of the inscription and corrections of
IGLS. Littmann responds to Kugener in RSO 4 (1911):
19398. See also Kugener, RSO 1 (1907): 57779,
on the tradition of lintel disks in Syria and
Mesopotamia.
95. Kugener, JA 9 (1907): 51920, argues persuasively that two separate hands carved the Syriac and
the Greek.
96. The inscription at Zabad is one of the oldest examples of Arabic script. See Shahd, BASIC, 1:
699702.

Page 118
and Rusafa lay the monotonous, arid plain of alMatyaha, literally, the place without landmarks where
one loses one's way. But having traversed this exposed plain east of Zabad and descended from the
watershed to the Euphrates, one traveled easily to
Rusafa along the caravan route that ran west of the
river. It was also possible to cross directly to Rusafa
through the steppe along the tracks used by pastoralists. Located in a mixed zone of agricultural and pastoral land, at the northern edge of the basalt Jabal Shbayt, the Sergius church with its inscriptions in Greek,
Syriac, and Arabic illustrates with particular clarity the
power of this cult to draw together the region's diverse
cultural strands.
North of Beroea a somewhat unusual inscription was
carved onto the lintel of a small building preceded by
a vestibule at Kafr Antin.97 In the lintel's center is a
square plate in reliefperhaps it held an image of the
saints mentioned in the inscription. Around the square
runs a prayer for the builder and a date, August A.D.
523. Left of the blank plate is a separate inscription invoking S. Theodore and S. Sergius, spelled
.
Farther to the northeast of Beroea, at the modern

395/754

village of
, just west of the Euphrates, an important Sergius dedication was discovered in 1974.98 The
mosaic inscription is part of a geometrically patterned
mosaic floor. Column bases and fragments of capitals
scattered among the adjacent houses must originally
have been part of the same church. The inscription is
clear
and
dated:
. The Seleucid
date of 743 places the martyrium's consecration in
A.D. 431, precisely when Alexander, bishop of Hierapolis, was constructing the first martyrium of Sergius
within Rusafa's walls. Southwest of
, another mosaic inscription at
also evoked the name Sergius, presumably as the church's dedicatee.99 East of
, on the west bank of the Euphrates just north of
ancient Zeugma, the village of Ehnesh (Turkish
) preserves a chapel of Sergius in a grove
alongside a stream, known still at the beginning of this
century as the Serkis-su.100 Although the
97. PAES, 3B6: 2034, no. 1202, with drawing; Dussaud, Topographie, map 8 at 3B.
98. Candemir and Werner in Studien zur Religion, 1:
23031.

396/754

99. Ibid., 22627. Robert, Bull. (1979), no. 603, accepts the view of Candemir and Werner that the isolated name Sergis refers to the church's dedication.
Robert also repeats the authors' innaccurate comment
that attestations of the cult of Sergius (not George, a
common confusion made in antiquity and also here by
Robert) appear in Syria from the early sixth century,
overlooking the mid fifth-century church of Sergius
nearby at Edessa.
100. Cumont, tudes syriennes, 15253, records the
stream's name and the primarily Armenian composition of the village in 1907. Palmer in Polyphonia byzantina, 4584, has thrown much light on Ehnesh's
rich, but until now blurred, epigraphical record. See
also Palmer, Seventh Century, 7174. Chabot first recorded the inscription (although he could not recall
whether it belonged to a church of Sergius or George), without a map or careful commentary in JA 9
(1900): 28588. For sketch maps of the vicinity, see
Palmer in Polyphonia byzantina, 81, and Cumont,
tudes syriennes, 289, map 7; for a broader geographical context,

Page 119
church itself is not dated, its antiquity and original association with a non-Chalcedonian community is confirmed by the unusual Syriac inscription carved in estrangela script on the church's exterior.
Continuing around the northern edge of the SyroMesopotamian plain, the city of Edessa could boast
two churches dedicated to Sergius, one within the
walls and another outside the eastern city gate. The
church outside the walls was built by Bishop Hiba, the
successor to Rabbula, who died in 436.101 S. Symeon was added to the church's dedication sometime
between the Stylite's repose in 459 and 498, when it
is first identified as the church of SS. Sergius and Symeon. In May of that year, when Anastasius abolished
the chrysargyron tax paid by artisans, a jubilant celebration broke out and spread to the church outside
Edessa's walls: "The whole city rejoiced and they all,
both small and great, put on white garments, and carried lighted tapers and censers full of burning incense
and went forth with psalms and hymns, giving thanks
to God and praising the emperor, to the shrine of S.
Sergius and S. Symeon, where they celebrated the
eucharist."102

398/754

In 503, Kawad besieged Edessa, against the good advice of a Christian from Hira, who knew of Christ's
prophecy that the city would never fall into enemy
hands. Before the shah's arrival, the Edessenes
gathered the relics into the city from the surrounding
countryside, both to save the holy objects from desecration and to bolster the city's defenses. Areobindus,
the Roman commander at Edessa, accepted to
emerge only a short distance from the city walls, in order to treat with the Sasanian general Bawi at the
church of S. Sergius.103 The church was later burned
by the Iranian forces after their failure to take the city.
The presence of a church dedicated to SS. Sergius
and Symeon just outside Edessa reflects the popularity of both saints among the pastoral and semi-pastoral Arabs, whose control over the region could at times
reach as far as the city walls.104 But its location outside the walls did not inhibit city dwellers from participating in festivities there.
A sanctuary dedicated to Sergius in the sensitive military zone east of Edessa, between Dara and Nisibis,
served as funerary chapel for the Iranian

399/754

although Ehnesh is not marked, see Honigmann, Ostgrenze, map 3, between Zeugma/Balqis and the
Merziman-ay.
101. Chron. 1234 1. 180.
102. Chronicle of Edessa of 506, 25758, tr. Chabot,
190. See also Luther's commentary, Syrische
Chronik, 166.
103. Chronicle of Edessa of 506, 28487, tr. Chabot,
20911. Although the Chronicle refers to the church
as that of S. Sergius only, it must be identified with the
church (referred to earlier) of SS. Sergius and
Symeon.
104. See ibid. 3089, tr. Chabot, 207, on Iranian and
Roman attempts to control Arab raids against the villages of northern Syria-Mesopotamia. Dilleman,
Haute Msopotamie, 7677, describes the threat of
raids that characterized this region, especially
between Edessa and Nisibis and at times of drought.

Page 120
martyr Gulanducht, who died there in 591. The ruins
of a fortress named Sargahan correspond to the location given in the Vita, and the name preserves the
dedication to Sergius.105 The Vita also records that
Gulanducht, on pilgrimage to the Holy City from Nisibis, visited the shrine of Sergius in the "Barbarian
Plain."106

THE IRANIAN EMPIRE

The Sergius cult spread beyond the western part of
Mesopotamia that was under Roman control into the
heavily Christianized eastern region in Iranian territory. Theophylact, writing in the 630s, calls Sergius
"the most efficacious saint in Persia."107 In the Acta
of Mar Qardagh, S. Sergius makes three encouraging
visits to Qardagh, a probably fictional marzban of Arbela during the reign of Shapur II (31079).108 The
appearance of the military martyr Sergius to another
in the making is a vivid illustration of the spread of the
Sergius cult eastward.109 Although the date of the
Acta is not secure, the wide dissemination of S.
Sergius's influence in the sixth century points to the
period between 550 and 650 for the Acta's

401/754

composition.110 S. Sergius's role in the Acta also

demonstrates that the military saint had not been
monopolized by the Romans. At a critical turning point
in the story, Qardagh, the protg of Rusafa's martyr,
successfully engages and routs the "Romans and
Arabs" who have been raiding his territory.111
The known dedications to Sergius within the Iranian
empire are primarily found west of the Zagros, where
most of the Christians lived. Qaraqosh, a village near
Mawsil, had a church of Sergius and Bacchus, probably dating back to the growth of anti-Chalcedonian
influence in the area during the second half of the
sixth century.112 Not far from Hira and
, a Nesto105. Vita Gulanducht 24 and 26. The classical or late
Roman fort preserved at Sargahan, in the borderland
between Roman and Iranian territory, has elicited discussion from, e.g., Peeters, AB 62 (1944): 121; de'
Maffei in Seventeenth International Byzantine Congress, 24042, with photographs 15; Sinclair,
Eastern Turkey, 3: 34142, 367.
106. See above, pp. 6566.
107. Theoph. Sim., Hist. 5.14.3.

402/754

108. Acta Mar Kardaghi martyris 7, 30, 34, 53.

109. See chapter 1 on the image of the saint in
Qardagh's visions.
110. In his excellent article in Festgabe, ed. Eilers,
150 with n. 40, Wiessner argues persuasively for a
sixth-century date. Cf. also Peeters, AB 43 (1925):
29899, who preferred a seventh-century date. For a
recent discussion of the date and historical context of
the Acta, see Walker, "Your heroic deeds," esp.
2732.
111. Acta Mar Kardaghi martyris 41. See the translation by Walker, "Your heroic deeds," 23536.
112. Fiey, Assyrie chrtienne, 2: 45859. This church
preserves traces of an ambo (Fiey, Mossoul chrtien,
9899).

Page 121
rian monastery dedicated to S. Sergius flourished
from before the reign of Khusrau II until at least the
tenth century.113 As a Zoroastrian, the future Mar
observed the catechumens gathering at the
church of S. Sergius to receive the sacrament of baptism during the night of the Easter service. By God's
grace, he saw the angels crown the newly baptized as
they emerged from the font, their garments radiant
with the true light. The Zoroastrian too was moved to
shed his old skin and receive the seal of baptism.
After spending time in Hira as a disciple of Mar Babai,
later took up residence in a cave, where he once
applied healing oil to a hunter who had been savaged
by a lion. George of Izla, an Iranian aristocrat who
converted to Nestorian Christianity and was martyred
in 615, was buried in a monastery dedicated to Sergius at Mabrakta near Seleucia-Ctesiphon.114 The
Acta of Anastasius the Persian (d. 628) mention a
monastery dedicated to the holy martyr Sergius six
miles outside Dastagird, at the village known in Greek
as Bethsaloe (
) where the king had a residence.115

404/754

These scattered signs of the Sergius cult in the Iranian empire do not on their own justify Theophylact's
grand claim. Fortunately, two exceptional sources survive to shed further light on the cult's eastern dissemination. They are both Histories of non-Chalcedonian
bishops who encouraged devotion to the saint as they
evangelized the countryside. In 559, Ahudemmeh was
consecrated bishop of the non-Chalcedonian diocese
of Beth
and metropolitan of the East.116 He had
grown up in Balad near the Tigris, the eastern limit of
Beth
. The city of Takrit formed the region's
southern extremity, Nisibis marked the north, and the
Khabur the west. Through a shrewdly articulated mission, Ahudemmeh promoted non-Chalcedonian Christianity in Mesopotamia, particularly among the
"barbaric" tent-dwellers of the Jazira, until his martyrdom in 575.117 Christ's exhortation "feed my lambs"
rang with immediate clarity for the newly consecrated
bishop.118 His missionary strategy was founded on
the establishment of a clerical hierarchy, cultivation of
pious habits, and construction of
113. Chronicle of Seert, pp. 54950. See Fiey, AB 79
(1961): 110.

405/754

114. Syn. or. 625. See also Fiey, AB 79 (1961): 110.

115. Acta Anast. Pers. 32, 39, 43. On the Christian
community of Bethsaloe, see Flusin, Saint Anastase,
24346. The village Bethsaloe and its Sergius monastery, located one mile outside the village, is otherwise
unknown. Hoffmann, Auszge, 120, located Bethsaloe on the Diyala river (Acta Anast. Pers. 38), which
flows six miles to the north of Dastagird.
116. According to Bar Hebraeus, Chron. eccl.
2.100102, Christophorus, catholicus of the Armenians, consecrated Ahudemmeh as bishop, while the
tireless
non-Chalcedonian
missionary
Jacob
Baradaeus made him metropolitan. Nau convincingly
argues, in his edition of the History of Mar Ahudemmeh, 20 n. 3, that Jacob Baradaeus was the hierarch
who, according to the Hist. Ahud. 20, consecrated
Ahudemmeh to both dignities at the same time.
117. See also Bar Hebraeus, Chron. eccl. 2.99100.
118. Hist.Ahud. 4, p. 21.

Page 122
sacred buildings among the tribes in order to focus
Arab devotion within the Church's control.
Of course, conversion was the prerequisite, but both
the language and manner of the nomads presented
daunting obstacles.119 Traveling from campsite to
campsite, Ahudemmeh worked a spell of conversion
over the tribes by a twofold plan of negation and revelation. Old forms of worship were shattered: the
Arabs saw their holy places destroyed and stone idols
broken by the power of his prayer.120 Like flies,
demons were chased from their accustomed seats of
worship, the sick cured, lepers purified.121 With a
flood of miracles and signs, Ahudemmeh affirmed the
true religion. Yet as he followed the tribes' movements, some encampments still met him with a hail of
stones.122 Only after the dramatic exorcism of a
shaykh's daughter were Ahudemmeh and his God finally embraced by one camp. Others were to follow.
After breaking through the first barrier, he traveled
among the encampments, sharing the hardships of
extreme cold, heat, difficult tracks, and scarce supplies of water, while at the same time teaching, baptizing, and consecrating a priest and a deacon for each

407/754

community. He also made certain to name churches

after the heads of each clan, in order to encourage
their leadership in ecclesiastical affairs.123
Ahudemmeh was credited with the conversion to nonChalcedonian Christianity of Arab pastoralists in the
region of Beth
between Nisibis, Jabal Sinjar, and
Balad, but also in the area of Takrit, as far south as
Hira.124 Christianity had to adapt to the demands of
the landscape and the way of life it imposed. Impromptu or portable altars served the needs of Arab
Christians, who would seize the opportunity, whenever it presented itself, to participate in the divine mysteries.125 Stories about Arab conversion
119. Ibid., p. 22.
120. Ibid., p. 23.
121. Ibid., pp. 2324.
122. Ibid., p. 24.
123. Ibid., p. 26 for baptism and p. 27 for clergy and
the naming of churches. Cf. Sozomen's description of
Arab Christians: "[Some of the Saracens] began to

408/754

convert to Christianity not long before the present

reign [Theodosius II]. They became familiar with faith
in Christ through their encounters with the priests and
monks who lived close to them, virtuous men and
wonder-workers, who were living a contemplative life
in the neighboring desert" (Soz., HE 6.38.14).
124. Morony, Iraq, 374. On Beth
, see Dillemann,
Haute Msopotamie, 7578; and see also Donner,
Early Islamic Conquests, 15773.
125. E.g., Leontius of Jerusalem, Contra monophysitas, PG 86.19001901 (I thank Garth Fowden for this
reference). Use of portable altars by the Christian
Taghlib tribe is attested as late as the eleventh century in the Murshid of Abu Nasr al-din Yahya ibn Jarir,
cited by Fiey, L'Orient Syrien 8 (1963): 32021. In this
account, a bishop attired in ecclesiastical dress
traveled with the tribes and celebrated the liturgy in
Arabic. When presbyters and deacons journeyed in a
desert land, taking with them a chalice and paten, but
lacking an altar, it was per-

Page 123
often include renunciation of camel meat, a feature of
desert life particularly repugnant to many settled
people and no doubt woven into conversion stories for
that reason.126 An eleventh-century account of the
conversion of a Turkish king illustrates the creative
flexibility demanded of the spiritual shepherds of pastoral communities. The unnamed ruler had converted
to Christianity, thanks to a rescue mission by S. Sergius, who saved him when he lost his way on a hunting
expedition. His subjects were nomadic peoples, and
in order to nourish their newly adopted Christianity,
"the king set up a pavilion to take the place of an altar,
in which was a cross and a Gospel, and named it after
Mar Sergius, and he tethered a mare there, and he
took her milk and laid it on the Gospel and the cross
and recited over it the prayers which he had learned,
and made the sign of the Cross over it, and he and his
people after him took a draught from it."127 In the
case of the Lenten fast, the Nestorian patriarch allowed that, since their diet consisted of only milk and
meat, the nomad Christians should abstain from meat
and consume only sour milk, a sacrifice, since they
usually took it sweet.

410/754

The tactics Ahudemmeh employed to encourage the

Arabs' affection for the Christian religion were not designed to disrupt the established rhythms of nomadism. For example, the History nowhere hints that
the bishop encouraged sedentarization among the
Arab groups. Instead, he stressed care for the indigent, for strangers, and especially for the monasteries
of central Syria-Mesopotamiain both Roman and
Iranian territory.128 Care for these communities in
practical terms meant subsidy, and this was possible
only if the Arab pastoralists maintained the traditional
migrations with their flocks, their primary source of income. The tribes "made generous gifts that would sell
at high prices" to sustain the monks' bodily needs.129

mitted that the deacon hold the two sacred implements and himself become the altar (see West Syrian
Synodicon 237, ed. and tr. Vbus). Portable plaques
carved with crosses have been discovered in the ruins
of a Christian monastery from the early Islamic period
at Ain
in the southwestern Iraqi desert (Okada,
11 [1990] 10312). The excavators interpret the
plaques as personal icons. I would like to thank

411/754

Yasuyoshi Okada for providing me with offprints from

.
126. E.g., Theod., HR 26.13. Some established forms
of behavior were especially resistant to change. John
Moschus relates an incident involving a group of
Ghassanid-allied Arabs whose booty from a successful raid included an adolescent boy. After some argument in Greek, a priest who happened to witness the
event managed to rescue the boy from becoming a
first-fruits offering, made in honor of the priest back in
the Arabs' camp (Joh. Mosch., Prat. spir. 155). The
identification of the Arabs as allied with the
Ghassanids, well-known non-Chalcedonians, should
be taken with caution in the light of Moschus's commitment to Chalcedon.
127. The account is from the Nestorian patriarch Mari,
Kitab al-mijdal, cited in Mingana, BRL 9 (1925):
31011. (I owe this reference to Peter Brown.)
128. Hist. Ahud. 4, pp. 2728.
129. Ibid., p. 28.

Page 124
Once Christian customs had been planted among the
tribes, Ahudemmeh initiated the next stage of mission,
the construction of a shrine to Sergius.
He built a great and beautiful house of dressed stone
in the middle of Beth
in a place called
. He
placed in it an altar and some holy relics and called
the house by the name of Mar Sergis, the famous
martyr, because these Arab peoples bore great devotion to his name and had recourse to him more than to
all other men. The saint (Ahudemmeh) attempted by
means of this house, which he had built in the name
of Mar Sergis, to keep them (the Arabs) away from
the shrine of Mar Sergis of Beth Rsapha, since it was
far distant from them. He made it, as far as he was
able, resemble the other, so that its beauty might hold
them back from going to the other. Near this shrine he
had built, he further constructed the great and famous
monastery of
.130
The project was, in other words, a deliberate attempt
to interrupt the flow of local Arab pilgrims across the
Euphrates to the Sergius shrine at Rusafa. Ahudemmeh carefully channeled their enthusiasm for local

413/754

benefit. Not only its name, but the very plan of the
new shrine imitated its rival at Rusafa.
In 1956, J. M. Fiey identified Ahudemmeh's complex
as the ruins of Qasr Sarij, east of Jabal Sinjar near
Balad, and he was followed by David Oates, who
studied the complex and dated it to 565.131 The
church, like that described by Ahudemmeh's biographer, is of dressed limestone. Its plan is unique in
Iraq, but immediately identifiable as the fifth- and
sixth-century North Syrian model also adopted at
Rusafa. The central nave with semi-circular apse at
the eastern end was flanked by two aisles. The northern aisle terminated in a diaconicon, which communicated with the apse. To the south of the apse was a
martyrium,132 which opened onto the southern aisle,
not through a small door, as in the case of the diaconicon, but through a wide arch nearly spanning the
aisle's full width. Access was not allowed from the
apse to the north, as considerations of symmetry
might have dictated. Instead, a door led into the martyrium from the porch on the south side. In this way,
the martyrium south of the apse was accessible from
the south porch, as the martyrium in basilica A was
from its north courtyard. This porch, a feature

414/754

especially suited to the region's heat,133 seems to

have ex130. Ibid., p. 29. Cf. Symeon of Beth Arsham's activities at Hira also in the mid sixth century. After converting Arabs of the Lakhmid tribe, Symeon induced the
leaders to build a church (Joh. Eph., Vita sanct. 140).
131. Fiey, Sumer 14 (1958): 12527, with figs. 15;
Oates, Ancient Iraq, 10617. See Okada,
12
(1991): 7183, where he notes that of all late antique
churches in Iraq, only the two in the north, Qasr Sarij
and the basilica at Tal Musayfna, are built of dressed
stone and have semi-circular apses.
132. See Hist. Ahud. 4, p. 29, for the installation of
relics.
133. Amm. Marc. 20.6.9 describes the area surrounding Sinjar as especially desiccated.

Page 125
tended around the southern, western, and northern
sides of the church, and on the west, it evidently acted
as a narthex, since there are no traces of an internal
narthex.134
Near the shrine, Ahudemmeh constructed a monastery, now largely destroyed, where he brought together a substantial community.135 Poor and stranger received hospitality in this "garden filled with goods for
the entire country where it was situated."136 Nearly
half a century later, Maruta, the non-Chalcedonian
metropolitan of Takrit (62949), erected a monastery
dedicated to Sergius beside a spring called
in
the middle of the desert between Takrit and Hit on the
Euphrates.137 Situated on the main route to
, the
new Sergius monastery became, like Ahudemmeh's
foundation, a harbor in the wilderness carefully
chosen for its well-watered location. The arid region
south of Takrit between the two great rivers was controlled by Arabs who monitored the steppe route
passing between the rivers and provided an alternative to the riverine routes, where merchants paid heavy
tolls. Maruta's monastery attracted the Arabs of the
region as well as merchants and travelers. Not only

416/754

the physical setting, but also the splendor and holiness of the establishment drew visitors. The History of
Maruta describes the elaborate decoration and equipment that embellished the monastery: elegant architecture, costly fabrics, and precious sacred vessels.
The monks who moved into the foundation and
served as missionaries for the neighboring region
were described as vigorous ascetics.
By them [the monks] and by this monastery all Mesopotamia was pacified because [the monastery] was
situated at the center of it. God, by the hands of our
father [Maruta], made of it a refuge, a harbor [
],
and a place of repose for all who travel and dwell in
this desert, and at the same time a joy, a refuge, a
protection against danger, against hunger and thirst,
for all who pass the place. Those who cross the
desert to reach
find rest there, it is their harbor.
Those who travel from the Euphrates to the Tigris or
the Tigris to the Euphrates stop there. One must see
the [multitudes who] camp there and pass on, and
others who dwell there; feeding their hunger they are
sated, drinking, they are refreshed. The indigent, the
afflicted, the sick, and the feeble are brought there,
above all by the people who live in Mesopotamia, and

417/754

are cured and leave strengthened, in good health and

succored as much in body as in spirit. [This monastery] saved many men, it protected them and guarded
them from lions, from chill, from heat, and from other
dangers and rescued them. The monks who live there
led many souls that were far from God and
134. Oates, Ancient Iraq, 107 n. 4.
135. Hist. Ahud. 4, p. 30. Traces of the monastery remain, but the structures were built, not of dressed
limestone like the church, but of rubble and mortar,
making reconstruction of the plan impossible (Oates,
Ancient Iraq, 106).
136. Hist. Ahud. 4, p. 30.
137. Denha, Hist. Marut. 6, pp. 8586; see also
Morony, Iraq, 37578.

Page 126
from knowledge of Him to the orthodox faith, and they
were a cause of good to them. This was the case not
only for travelers in the desert, but also for those who
dwell in the fortresses of the Euphrates.138
This flow of visitors across Syria and Mesopotamia
was essential to the function of the holy sites created
by Ahudemmeh and Maruta. Their services were
carefully tied to the terrainshelter for travelers, food,
drink, and security for their financial resources.139
The cool and green of a monastery could easily conjure up images of paradise among desert-dwellers.
Monastic settings, including those within the urban
surroundings of Rusafa and Hira, were known to offer
refuge in times of plague, but on a more regular basis
played a vital role in social life.140 When Hind, the
eleven-year-old daughter of the Lakhmid king
of Hira went to church on Great Thursday to communicate, she was espied for her beauty by
ibn Zayd,
the Christian poet and diplomat. Her maids noticed
and indulged his attentions, while their mistress remained for some time unaware that she was being
watched. Once she realized her maids' complicity with
the man's illicit gazes, Hind fell into a rage, even

419/754

striking one of the girls. A year after the incident in

Hira, one of Hind's maids described to her mistress
the splendid church at Tuma, also in Hira, lingering on
the monks, the beautiful lamps and building, and the
girls who visited it. Intrigued by the description, Hind
received permission from her mother to visit the
church herself. The maid then tipped off
, who arrived at the church adorned in his richest Iranian
finery, surrounded by a great entourage of young
men. Hind's maid drew her mistress's attention to
,
praising his beauty as even greater than that of the
lamps and all the other decorations in the church.141
After the coming of Islam, the Umayyads in particular
were impressed by these built oases, by their soothing atmosphere and architecture.142 Especially during great feasts, when wine and song filled the
sheltered spaces, monasteries could also provide an
inviting backdrop for courtly love. The
poet
sang the praises of the garden around
a church dedicated to S. Sergius, where he met with
his beloved at the
138. Denha, Hist. Marut. 6, p. 86 (tr. adapted from
Nau); cf. pp. 8889.

420/754

139. Theod., Graec. aff. cur. 8.63, describes how travelers prayed for guidance and protection at martyrs'
shrines before setting off on a journey and, when possible, returned to give thanks.
140. Caliph Hisham retired to Rusafa in order to escape the plague and to be near the monastery (Yaqut,
al-buldan 2.510, 3.47, tr. Le Strange, Palestine,
432, 522). On the commonplace of retiring to the
desert in time of plague, see Conrad in Quest for
Understanding, ed. Seikaly et al., 26971.
141. al-Isfahani, Kitab al-aghani 2.12223.
142. See, e.g., Hamilton, Walid and His Friends, 86.

Page 127
saint's feast.143 A similar scene is described in the
Miracula of S. Thecla. At her shrine near Seleucia, a
group of pilgrims gathered for a meal after the saint's
feast, and entertained themselves by describing, each
in turn, what they had found most delightful: the brilliance and the splendor of the church, the magnitude
of the crowd, the noble assembly of bishops, the inspired teaching, the melodious chanting and the wellordered vigils, prayers, and ceremonies. One man
confessed that he was so transfixed by the dazzling
beauty of a young woman also attending the service,
and so consumed by his desire for her, that he could
not tear himself away to pray to the martyr. After a visitation from the indignant Thecla, the lecherous man
died three days later.144
Clearly, monasteries, especially when set near roads,
could exercise extraordinary magnetism over a
region's settled and mobile population. Ahudemmeh's
justification for his new Sergius shrine, namely, the
great distance that separated Beth
from Rusafa,
was in fact an attempt to conceal his effort to control
the pastoral Arabs, whose migration patterns
stretched across the Syro-Mesopotamian steppe.145

422/754

The monasteries established by the leaders of nonChalcedonian Christianity served to anchor mobile
groups into a program of Christian behavior focused
on monastic communities. And the monasteries depended in turn on a reciprocal relationship with visitors. Our accounts of the activities of Ahudemmeh and
Maruta reveal the centrality of economic factors to religious and political life in Syria-Mesopotamia. Here
Procopius too is useful. In his description of Rusafa
he explains how "this church, through its acquisition of
treasures, came to be powerful and celebrated."146
The History of Ahudemmeh relates how even before
the construction of the shrine of Sergius and the adjacent monastery at Qasr Sarij, the Arabs supported
monasteries by rich donations that the monks could in
turn sell to provide their daily bread. Sanctuaries and
monasteries became repositories of material as well
as of spiritual wealth; a pious donation might augment
the dedicant's local prestige and power, since investment at these holy places took the form of display before peers. But the sphere of influence within which
the holy places operated
143. al-Isfahani, Kitab al-aghani 19.251. The tenthcentury poet
al-Basri records his fascination

423/754

with a statue in a convent of Sergius (cited by Fars,

Vision chrtienne, 7). The Sergius church is not identified more specifically in either poem. The poet Abu
Nuwas similarly evokes the pleasures of pilgrimage to
a Sergius monastery, possibly that near
, attested
in the mid tenth century, even comparing it to the
great Meccan hajj (cited in Wagner,
111,
19799).
144. Mir. Theclae 33; cf. 34.
145. Until the imposition of the modern borders, the
Jazira tribes traveled regularly between the Upper Tigris and the Middle Euphrates (Oates, Ancient Iraq,
116).
146. Proc., Aed., 2.9.56 (tr. Dewing).

Page 128
was not limited to their immediate locality. Maruta's
Sergius monastery was said to attract visitors from as
far afield as the "fortresses of the Euphrates," which
must refer to Circesium, Zenobia, Callinicum, Sura,
and even Rusafa.
For the political leaders of Rome and Iran, the consolidation of Arab religious allegiance at shrines and
monasteries was not important only for political and
economic reasons. Through permanent holy places,
settled powers could attempt to exercise political control over transhumant as well as settled tribes. Roman
and Iranian negotiators in the peace treaty of 561 took
pains to monitor Arab involvement in cross-border
trade and in tribal conflicts across the political frontier.147 Arab disputes over pasture had been known to
escalate into Roman-Iranian confrontation.148 Competition for control inspired not only bishops in the Iranian empire, but also the king of kings. When Nestorians burned Ahudemmeh's complex, it was Khusrau I
who immediately had it reconstructed, sparing nothing
on its ornament. Later, in 575, near the end of
Khusrau's long reign (53179), Ahudemmeh died in
the prison to which the Iranian king had committed

425/754

him after the bishop converted an Iranian prince and

aided his flight to Roman territory. Despite the beneficial effects of Ahudemmeh's activities among the Arab
tribes, Khusrau did not forget that tolerance of Christianity was a policy in the service of political expediency. Khusrau's ostentatious rebuilding of the Sergius
monastery at
was just one response to the increasing involvement of Roman emperors from
Anastasius to Maurice at the popular frontier shrine of
Rusafa.
In late antique Syria and Mesopotamia, the name Sergius became a byword for power and success. Ahudemmeh claimed in the mid sixth century that love for
the name of Sergius would keep the Arabs bound to
his new shrine at Sinjar. Jacob of Sarug saw the world
through the same lens when he wrote "beautiful name
Sergius, which is beloved to all ears" and "a region full
of danger fell to this energetic soldier, and his fame
and his name made peace in it and he is its pride."149
In his ex-voto dedication to the saint, Khusrau II
ascribed his wife's conception of their child to
Sergius's "most holy name."150 Adopting the martyr's
name and evoking the name Sergius were considered
tantamount to making the martyr present instantly.

426/754

The name was believed to possess the power to

break through the deceptive constructions of time and
space that were thought to separate the living from
147. Men. Prot., fr. 6.1, ll. 314407; see also Shahd,
Byzantium and the Semitic Orient VII, 181213.
148. For the dispute near Nisibis in 484 caused by a
two-year drought, see Nau, Arabes chrtiens, 1315.
See the discussion of the political dimension of the
grazing disputes of 536 and 539 (the "Strata dispute")
above, pp. 6667.
149. Jacob of Sarug, "Mimra on the Victorious Sergius," tr. Palmer, 118 (see epigram) and 137 (see pp.
2526 above).
150. Theoph. Sim., Hist. 5.14.8.

Page 129
the dead. In a story set in the reign of Leo I, a Jew
who was burning at the stake called out, "God of Sergius, help me. Saint Sergius, you know!" and immediately was rescued by the saint on horseback, accompanied by Bacchus. When the Jew was later baptized,
he adopted the name Sergius as a constant protection.151 That the story may be a literary invention is of
little interest. It is the vivid link between name and
presence that we must retain.
The power and influence of the name Sergius were
expressed, not only in the spiritual life of the individual, but also in the political arena. One example of this,
discussed in chapter 5, is the reinstallation of Khusrau
II on the Sasanian throne. By the late sixth century,
Sergius's fame had reached the shores of the Atlantic,
where a Syrian merchant had installed the martyr's index finger in a house-chapel in Bordeaux. Bishop Bertramn of Bordeaux had heard rumor of an eastern
king who charged victoriously into battle with S.
Sergius's thumb attached to his own. Craving such
miraculous triumphs, the bishop attempted to wrest
the relic from the Syrian. But in the struggle that ensued, Sergius's finger was splintered into tiny

428/754

fragments.152 Considering the potential for garbling

reports between Mesopotamia and Bordeaux, it is not
improbable that the eastern king of the rumor was originally Khusrau II.153 Sergius had become a preeminent patron of victory, and of military victory especially. But the distant king in the East could easily
have been Anastasius, Justinian, Khusrau, or the
Ghassanid phylarch al-Mundhir, all of whom showed
their faith in the saint's power.
151. The story, whose source is unknown, is told in
the fourteenth-century ecclesiastical history of
Xanthopoulus, Eccles. hist. 25.24.
152. Greg. Tur., HF 7.31.
153. Tillemont, Memoires, 5: 496, makes this
suggestion.

Page 130

FIVE
Frontier Shrine and Frontier Saint.
JUSTINIAN, THEODORA, AND S. SERGIUS
At the heart of Constantinople rose the most elegant
church ever erected in honor of Rusafa's martyr. Its
plan is an octagon set in a square. The eastern wall of
the square has an apse and the western wall is
pierced by five doorways linking the narthex with the
main body of the church. Through the narthex, the
church was connected on the south to the now destroyed basilical church of SS. Peter and Paul, with
which it shared a propylaeum and courtyard.1
Between the central octagon and the outer square is a
sheltered ambulatory on the ground floor, with a gallery on the second storey. Eight piers are joined by
exedrae, either rectangular or deep semi-circles, arranged in an alternating order to create a dynamic,
undulating movement around the central space. Two
columns of colored marble with elegant folded or Ionic
capitals punctuate the space within each exedra.

430/754

Finely wrought columns support the architrave on the

ground floor, while on the gallery level arches spring
from the piers to crown each niche. From these eight
piers and arches ascends a pumpkin dome divided into sixteen segments with alternately straight and
curved surfaces. According to Procopius, the walls
were richly adorned with marble revetment and gold
mosaic.
The monogram carved on the ground floor capitals
and the dedicatory inscription running along the entablature leave no doubt that the church was dedicated by Justinian and Theodora to the holy martyr
Sergius:
Other sovereigns have honored dead men whose
labor was unprofitable, but our sceptered Justinian,
fostering piety, honors with a splendid abode Sergius
1. Proc., Aed. 1.4.19. Janin, Gographie ecclsiastique, pt. 1, 3: 470, notes another, later church dedicated to Sergius and Bacchus in Constantinople,
known only from synaxaria.

Page 131
the Servant of Christ, Begetter of all things; whom not
the burning breath of fire, nor the sword, nor any other
constraint of torments disturbed; but who endured to
be slain for the sake of Christ God, gaining by his
blood heaven as his home. May he in all things guard
the rule of the sleepless sovereign and increase the
power of the God-crowned Theodora whose mind is
adorned with piety, whose constant toil lies in unsparing efforts to nourish the destitute.2
The epigram praises the martyr's spiritual and physical endurance in a generic manner, without specific reference to the tortures described in the Passio. Likewise, Sergius's fellow martyr Bacchus, although
present in the Passio, is absent from the epigram.
Justinian beseeches the soldier-martyr Sergius to
guard his rule. Theodora seeks the martyr's aid in the
accomplishment of good works, an interest of the
empress admired by Procopius and often shared by
the emperor.3 Despite the original dedication to S.
Sergius alone, which was maintained by some later
writers, others, as early as 536, included S. Bacchus.4

432/754

The church has attracted attention mainly from scholars interested in the debate over centrally planned
palace churches, and little has been said about the
church's broader significance for the cult of S. Sergius.5 This
2. CIG 4.8639. See von Hammer [-Purgstall], Constantinopolis und der Bosporos, 1: xiixiii; van Millingen, Byzantine Churches, 74; Ebersolt and Thiers,
glises, 24; Mercati, Rendiconti 3 (192425):
197205, for the history of the publication of the inscription and the text. Tr. Mango, JB 21 (1972): 190,
with slight alteration.
3. For example, after the fire started by the Nika Riot
in 532 had destroyed a hospice that nestled between
the old churches of Hagia Eirene and Hagia Sophia,
Justinian and Theodora rebuilt and expanded the hospice, absorbing two other hospices nearby in the process (Proc., Aed. 1.2.1418). An embroidered cloth
adorning the altar in the church of the Holy Wisdom
depicted the couple's ministry to the sick at hospices
and their visits to churches; another portrayed them
with hands entwined with those of the Mother of God

433/754

and Christ (Paul. Silent., Descriptio Sanctae Sophiae

796805).
4. The dedication to S. Sergius alone is maintained in
ACOec. 3.174; Joh Mal., Chron. 18.485; the ninthcentury Georgian Life of John, Abbot of S. Sergius by
John the Sceuophylax, 159, 163, but at 165 the dedication is to both; Const. Porph., De cer. 1.11; Anth. gr.
1.8. On the other hand, a dedication to S. Sergius and
S. Bacchus appears in Proc., Aed. 1.4.3; in the tenthcentury Patria Konstantinoupoleos, in Script. orig.
Const. 2.23132; and in the twelfth-century historian
Cedrenus, Hist. 1.642 (Bekker). Pierre Gilles, who visited the church in 1544, discerned in the pilaster
carvings of vines intermixed with clustered grapes
hints at the church's association with Bacchus (De topographia Constantinopoleos 9597). Three hundred
years later, Joseph von Hammer [-Purgstall] arrived at
the same conclusion after viewing the elaborate
carving of the entablature (Constantinopolis und der
Bosporos, 1: 37677); van Millingen, Byzantine
Churches, 72, corrects this misperception, noting that
"though the flower points upward it has been mistaken
for clusters of grapes," and adding that if the carving

434/754

did portray grapes, the symbol would refer to the vine

in the Gospel of John, and not to the god Bacchus.
5. Mango, JB 21 (1972): 18992; Krautheimer, JB
23 (1974): 25153; Mathews, Revue de l'art 24
(1974): 2229; Mango, BZ 68 (1975): 38592. I would
like to thank Irfan Shahd for sending the abstract of a
paper in which he touches on some issues of general
importance to the subject of Justinian and Sergius.

Page 132
brief discussion does not aspire to join in the architectural controversy, but rather to shed some light on the
importance of this Constantinopolitan church in the
context of frontier religion during Justinian's reign.
To begin with, it is worth recalling that Procopius describes the church as an adornment for the entire city.
Regardless of whether or not public access was allowed originally, as it certainly was by the twelfth century, the monumental church of S. Sergius could not
have escaped the notice of the city's inhabitants and
visitors. As Mango has underlined, it was built at a
smart address. The decision made by Justinian and
Theodora to dedicate the church to S. Sergius deserves to be seen within the broader context of the
benefactors' activities, and not just Theodora's charities, which were sometimes enjoyed by non-Chalcedonian monks. Justinian's interest in the Syro-Mesopotamian martyr has been given short shrift. Mango justifiably disregards the dramatic early seventeenth-century account in which Sergius and Bacchus appear to
the emperor Anastasius to warn him against executing the pair of officers, Justin and Justinian, who stood
accused of treason.6 In gratitude for the martyrs'

436/754

intervention, the legend continues, Justinian erected a

church in their honor in Illyricum, where the story was
set. But this late fantasy is hardly Justinian's only association with Rusafa's martyr. Justinian followed the
lead of Anastasius, whose defense of Rome's eastern
frontier zone employed frontier martyrs as well as fortifications. Procopius mentions a well or cistern built
by Justinian at a monastery of S. Sergius on Mount
Cisseron in Palestine (possibly Galilean Kisra), and a
chapel (
) of S. Sergius in Phoenician Ptolemais.7
It is not out of the question that the church at Constantinople would have housed the thumb of Sergius
translated to Constantinople by Anastasius and never
heard of again. Anastasius may have planned a
church for the martyr but simply run out of time, since
the translation took place at the end of his life,
between 514 and 518. Justinian, in turn, took an interest in Rusafa and at some point between 527 and
540 sent to the shrine a gold, gem-encrusted cross
dedicated to S. Sergius by both himself and
Theodora.8 Perhaps also, on the model of the cross
sent to Jerusalem by Theodosius and Pulcheria, the
royal pair had expected to receive a relic of the martyr
Sergius in return for their gift. Through building projects and this impressive joint gift, Justinian and

437/754

6. The legend and the "Life of Justinian" in which it is

found are discussed by Bryce, EHR 2 (1887): 65786,
a good piece of detective work. Cf. also Mango, JB
21 (1972): 191, with n. 7. One goal of this "Life of
Justinian" is to provide an etiology for the dedication
to SS. Sergius and Bacchus of the monastery that
stood on the banks of the river Bojana, near Shkodra
in northern Albania. The adjacent fortress, which commands the surrounding plain, is known as Rozafat: on
the toponomy, see
, Sitzungsberichte der kaiserlichen Akademie der Wissenschaften 136 (1897):
4954; Norris, Islam in the Balkans, 6162, with n. 38.
7. Proc., Aed. 5.9.20 and 25.
8. Evagr., HE 4.28, reports that the gift was joint; HE
6.21, that it was from Theodora alone.

Page 133
Theodora allied themselves with Rusafa's popular
saint: they imposed their presence at Sergius's shrine
in the frontier zone where defense, the fidelity of Arab
allies and opposition to Chalcedon were live issues.
The alliance was advertised at the capital too. While
reinforcing their authority at the martyr's frontier
shrine, they at the same time claimed the soldier-saint
as their own by establishing S. Sergius in a position of
honor in Constantinople.
The gift was a potent gesture at Rusafa, particularly if
seen in the light of Sergius's reputation among the
Arabs of Syria, whose political and religious allegiance to Constantinople was often suspect. In 542 or
543, Justinian bestowed on the Ghassanid leader alHarith the title of patricius.9 But as head of the Ghassanid confederation from 529 to 569, al-Harith zealously supported the non-Chalcedonian cause. At
more or less the same time as Justinian was conferring honors on him, al-Harith turned to Theodora to
approve the consecration of two non-Chalcedonian
bishops, Jacob Baradaeus and Theodore.10 Seen in
the context of two of Justinian's most considerable
preoccupations during his reignreconciliation

439/754

between Chalcedonians and their opponents, and

maintenance of the eastern frontierit was a wise decision to ally himself and his empress with the most
influential martyr in the frontier zone, a military saint
who enjoyed great popularity among the region's Arab
and mainly, but not exclusively, non-Chalcedonian
inhabitants.

IRANIAN INTEREST IN S. SERGIUS

Making his way up the Euphrates with his army in the
spring of 540, Khusrau I bypassed the naturally fortified site of Circesium and made a half-hearted attempt to force Zenobia's surrender before moving on
to Sura.11 The soldiers at Sura put up a valiant resistance until their commander was killed. On the following day, the Surans, in despair, sent their bishop to
Khusrau's camp to negotiate a surrender. But once
the city gates were opened, Sura was taken by trickery, its houses looted, and many inhabitants slain.
The survivors were taken hostage and their fate was
put in the hands of Bishop Candidus of Rusafa by
Khusrau, who presented the bishop with the opportunity to ransom 12,000 captives for two hundred pounds
of gold. Candidus reluctantly promised to raise the

440/754

money and the hostages were set free. Two years

later when Khusrau returned to collect his due, Candidus emerged from Rusafa empty-handed to plead
with the king.12 Under tor9. Shahd, BZ 52 (1959): 32143, with additions in id.,
Byzantium and the Semitic Orient XI.
10. Joh. Eph., Vita sanct. 499504; Bar. Hebr., Chron.
eccles. 1.46; Mich. Syr., Chron. 9.29 (tr. Chabot, 2:
24546).
11. Proc., BP 2.5.133.
12. Ibid., 20.116.

Page 134
ture, Candidus bade Khusrau take the sacred vessels
from Rusafa's shrine, and in due course an armed escort accompanied some of Candidus's men into
Rusafa, where they emptied the treasury, save the
martyr's silver reliquary. In particular, they took with
them the gem-encrusted gold cross dedicated to the
saint by Justinian and Theodora.13 But these holy
goods were not enough for Khusrau, who had been
informed that the rest had been hidden by the inhabitants. Six thousand men were sent to besiege the city,
but the plan to take it by stealth was foiled by an informant, an Arab Christian named Ambrus who was
fighting in Khusrau's army. Only two hundred Roman
soldiers were present to defend Rusafa, and the population was on the verge of surrender. At this point,
Ambrus detached himself again from the Iranian camp
to inform the inhabitants that the Iranian army had
enough water for only two days longer, and would
then be forced to retreat. According to this version,
told by Procopius, faith in S. Sergius among the
region's Arab population saved Rusafa. In Evagrius's
account, the martyr's intervention was more directalthough the Iranians had heard that only women, children, and invalids remained within the walls,

442/754

an immense army miraculously appeared on the battlements to defend the fortress-shrine.14 According to
both reports, Khusrau's army retreated the next day.
Half a century later, Khusrau II inherited a complicated history of invasion and dtente between his empire and Rome's. Both the conflict and its resolution
revolved not around differences but around shared interests, making trade, Arab allies, and Christians central issues in the empires' confrontations. The virtually
self-governing community of Christians in the Iranian
empire provided Khusrau II with a bridge into Roman
Syria and Mesopotamia. Association, not simply with
the Christian faith, but with the cult of S. Sergius in
particular became a means of making his influence
felt in the sensitive frontier zone between the two empires. During his long reign from 590 to 628, Khusrau
II learned from his grandfather's political shrewdness
in dealing with Christians. The later king's involvement
with Christianity was, however, more subtle and more
complex.15 Evagrius and Theophylact provide evidence for Khusrau's involvement with S. Sergius within
Roman territory, while for his patronage of the cult in
the Iranian empire material must be drawn from more
scattered and often later sources. The difference in

443/754

sources does not necessarily impede a broad view of

Khusrau's interest in Rusafa's saint. That interest
makes most sense when understood, not according to
territorial divisions, but in the context of frontier
politics.
Rusafa played a pivotal role in Khusrau's relationship
with Maurice. It
13. See Theoph., Chron. 5.13.12.
14. Evagr., HE 4.28.
15. Flusin, Saint Anastase, 2: 95127, traces the fluctuations in Khusrau's policies toward Christians.

Page 135
was the Roman military commander at Rusafa whom
Khusrau chose as the mediator who would convey to
the emperor his plea for protection.16 In the west
Syrian Chronicon ad annum Christi 1234, that commander is identified as "the Arab general who dwelt at
Rusafa as a subject of the Romans, a zealous Christian man named Abu Jafna
b. al-Mundhir."17 This
Jafna may, as his name suggests, have been the son
of the Ghassanid phylarch al-Mundhir. Maurice embraced Khusrau's cause with enthusiasm and established him at Edessa with John of Rusafa, in whose
household the exiled king could easily have heard
stories of the miracles performed by Rusafa's saint.18
Khusrau's flight from his empire and eventual restoration had been prophesied by the Iranian apostate Gulanducht, who was born into a Zoroastrian family in
Seleucia-Ctesiphon and married to a well-placed husband, but spent her Christian life in the frontier zone.
Inspired by a vision, she was baptized and subsequently imprisoned in the Castle of Oblivion. After
countless tortures, culminating in decapitation (from
which she recovered, with the help of an angel, who
reunited her head and body), Gulanducht journeyed

445/754

into Roman territory, and on her way to Jerusalem as

a pilgrim, she visited Rusafa. Later, she returned to
live in Hierapolis, where she dreamt about Khusrau's
future struggles. Her association with the frontier saint
Sergius extended beyond her repose in 591, since
she was buried at a chapel dedicated to Sergius
between Nisibis and Dara, in order to intercede on behalf of all the region.19 Evagrius gives a highly condensed account of Gulanducht's life in which he mentions the tortures she underwent at the hands of the
Zoroastrians and the miracles she wrought, and then
refers the reader to the Vita written by Bishop Stephen of Hierapolis.
Evagrius uses the story of Gulanducht to make a
transition from the events following Bahram's usurpation to the thank offerings dedicated to S. Sergius at
Rusafa by Khusrau after his restoration. Evagrius describes the gifts and includes their dedications, which
are nearly identical to Theophylact's transcriptions, all
of which will be discussed in due course. But first it is
worth noting the last three items that follow Evagrius's
extended account of Khusrau's gifts to S. Sergius and
close his Historia ecclesiastica. In

446/754

16. Mich. Syr., Chron. 10.23 (tr. Chabot, 2: 37172).

17. Chron. 1234 90 (Lat. tr. Chabot, 1: 169; Eng. tr.
Palmer, 115). On this chronicle, which preserves parts
of the secular history of Dionysius of Tall-Mahre (d.
845), see Palmer, Seventh Century, 85104.
18. For John of Rusafa, see above, p. 102.
19. Vita Gulanducht (Georgian redaction) 20.2. The
Georgian Vita is considered to be a faithful redaction
of the original Vita (probably Syriac), composed by
Bishop Stephen of Hierapolis, who knew Gulanducht
and recorded her life story between the saint's death
in 591 and Evagrius's mention of the work in 59394
(Garitte, AB 74 [1956]: 41925). For an abbreviated
Greek account written by the priest Eustratius before
602, see Peeters, AB 62 (1944): 74125.

Page 136
these items, the Antiochene historian focuses on important participants in the political life of the frontier
zone: Arab allies, Christian leaders, both Chalcedonian and oppositional, and miracle-working saints. The
Lakhmid leader
, who had formerly delighted in
human sacrifice, is converted to Christianity and commits to the flames his gold statue of Aphrodite;
Gregory, the patriarch of Antioch, after solemnly depositing Khusrau's gifts at S. Sergius's shrine, makes
a whirlwind tour of the frontier zone, where, Evagrius
reports, he reclaims from the Severan heresy many
fortresses, villages, monasteries, and entire tribes;
and Symeon the Younger, who, like the elder stylite,
had by his miracles attracted both Romans and "barbarians," finishes his earthly days.20
The order and substance of what Theophylact chose
to relate about Khusrau's reinstallation on the throne
also conveys a clear sense of the primary political and
religious influences in the frontier zone. As was noted
in chapter 2, the account of Khusrau's gifts to Sergius
at Rusafa opens Theophylact's fifth chapter, directly
following Bishop Domitianus's triumphalist hymn in
honor of the Iranian martyrs of Mayperqat at the end

448/754

of book 4. Theophylact gives a fuller account of

Gulanducht's life story than Evagrius, but both authors
clearly assume the reader's familiarity with the famous
saint's deeds. The history of John of Epiphania,
Evagrius's contemporary, provided the organizational
framework within which both Theophylact and Evagrius wrote. Differences arose from their diverging interestsTheophylact's in Khusrau's relationship to
Rome and Evagrius's in church history.21
The chronology of Khusrau's fall and rise can be
briefly outlined as follows:
15 FebruaryKhusrau II is crowned
590
9 March 590 Bahram II is crowned
spring
/Khusrau appeals to Maurice
summer 590
7
JanuaryKhusrau prays to S. Sergius
591

449/754

9 Februarythe head of Zatsparham, one of

591
Bahram's main commanders, is
brought to Khusrau
late summerBahram is defeated and Khusrau
591
is restored
autumn
Khusrau sends his first offering
591winter to S. Sergius
592
592

Khusrau marries Shirin,

Aramaean Christian

an

AprilAugustKhusrau sends his second offer593

ing to S. Sergius22
20. Evagr., HE 6.2223.
21. Khusrau's dedications to Sergius at Rusafa are
described by Evagr., HE 6.21, and Theoph. Sim., Hist.
5.13.114.12. See also Higgins, BZ 48 (1955):
98102.
22. For a political narrative, see Whitby, Emperor
Maurice, 23436.

Page 137
We should not underestimate the influence of the
widespread faith in S. Sergius's supernatural power
on Khusrau's decision to turn for assistance to the
Christian rider saint. Neither should we overlook that
the choice also allowed the Sasanian king to establish
a presence in the Barbarian Plain and create a rapport with the audience of S. Sergius's major shrine at
Rusafa. Khusrau's first offering to S. Sergius was a
gold cross, accompanied by the gem-studded gold
cross dedicated by Justinian and Theodora to the
saint and plundered from the treasury by Khusrau I.
Khusrau II sent to Antioch a request for the manufacture of a cross, along with funds and a letter stating
his motivation for the offering. After slight editing, the
letter (or, for practical reasons, more likely an abbreviated version) was inscribed on the cross made for
Khusrau and then sent to Rusafa.23 The inscription
reads:
This cross I, Chosroes [Khusrau], King of Kings, son
of Chosroes [dedicate]. When we departed for Romania on account of the devilish operations and villainy of the most ill-starred Baram son of Bargusnas
and of his associate cavaliers, and because the ill-

451/754

starred Zadesprates left the army and made for Nisibis in order to seduce the cavaliers of the district of
Nisibis into rebellion and complicity in revolution, we
also sent cavaliers with an officer to Charcha and,
through the fortune of the most holy and renowned
saint Sergius, when we heard that he was the granter
of petitions, in the first year of our reign, on the seventh of January, we petitioned that if our cavaliers
should kill or defeat Zadesprates, we would send a
bejeweled gold cross to his shrine for the sake of his
most holy name. And on the ninth of February they
brought before us the head of Zadesprates. So, since
we were successful in our petition, because each part
was unambiguous, we have sent to the shrine of the
most holy Sergius the cross made by us in honor of
his most holy name, together with the cross sent to
his shrine by Justinian, king of the Romans, which
was brought here in time of estrangement between
the two states by Chosroes our father, King of Kings,
son of Koades, and which was discovered in our
treasuries.24
Once reinstalled on his throne, Khusrau took a Christian named Shirin as a wife. Ten days after Khusrau
prayed to S. Sergius that Shirin conceive, the saint

452/754

visited him in a dream.25 In thanksgiving for Sergius's

intervention, in the spring or summer of 593, Khusrau
sent to Rusafa a golden paten and chalice, along with
a gold cross and a silk curtain embroidered in gold to
be
23. For Higgins's reconstruction, see BZ 48 (1955):
9298. The main difference between Evagrius and
Theophylact here is that the former understands the
text as an inscription on the gift, while the latter takes
it to be an accompanying letter. The same is true of
the second offering.
24. Theoph. Sim., Hist. 5.13.46 (tr. Whitby and
Whitby, with alterations).
25. Evagr., HE 6.21; cf. Theoph. Sim., Hist.
5.13.114.12. For Khusrau's involvement with Christians in Iran, see Nau, Arabes chrtiens, 4649, and
Morony, Iraq, 33283.

Page 138
deposited at Sergius's shrine.26 The gold paten's inscription elaborately denies possible misconstructions
of Khusrau's gift:
To the great martyr Sergius, Chosroes, King of Kings.
I, Chosroes, King of Kings, son of Chosroes, have dispatched the gifts accompanying the paten not for the
sight of men, nor so that the greatness of your most
holy name be known from my words, but so that the
truth about the events be recognized as well as the
many favors and benefactions which I had from you:
for it is my good fortune that my name should be carried on your holy vessels. During the time when I was
in Berthamas, I petitioned of you, holy one, to come
to my aid and that Seirem (Shirin) conceive in her
womb. And since Seirem is a Christian and I a Hellene, our law does not grant us freedom to have a
Christian wife. So on account of my gratitude to you,
for this reason I disregarded the law, and I held and
hold from day to day this one among my wives as legitimate, and thus I resolved now to beseech your
goodness that she conceive in her womb. And I petitioned and ordained that if Seirem should conceive in
her womb, I would send to your most holy shrine the

454/754

cross that she wears. And with regard to this both I

and Seirem have this purpose, that we should have
possession of this cross in remembrance of your
name, holy one. And we have resolved that for its
value, although this does not extend beyond four
thousand three hundred standard miliaresia, five thousand standard coins should be dispatched in its place.
And from the time when I had this petition in my mind
and made these calculations, until the time we came
to Rhesonchosron, ten more days did not elapse and
you, holy one, not because I am worthy but because
of your goodness, appeared to me in a dream at night
and thrice declared to me that Seirem had conceived
in her womb. And in the dream itself, I thrice
answered you in return and said, "Wonderful, wonderful." And because of your holiness and charity, and
because of your most holy name, and because you
are the granter of petitions, from that day Seirem did
not know what is customary for women. I was in no
doubt of this, but trusted in your words, even because
you are holy and a true granter of petitions. After she
did not experience womanly ways, from this I recognized the power of the vision and the truth of what you
had spoken. So straight away I sent the cross and its
value to your most holy shrine, giving orders that from

455/754

its value one paten and one chalice should be made

for the sake of the divine mysteries, but also that a
cross, which is owed, be fixed on the honored altar,
and a solid gold censer and a Hunnic curtain adorned
with gold; and that the remaining miliaresia are for
your holy shrine, so that through your fortune, holy
one, in all things, but especially in this petition, you
may come to the assistance of myself and Seirem.
And what has come to us through your intercession
by the mercy of your goodness, may it also advance
to completion at the
26. Evagrius describes the fabrics as Hunnish; on this
term, see Kislinger and Diethart, Tyche 2 (1987): 8.
According to the Piacenza Pilgrim, who visited the
Holy Land in the 570s, the entrance to the tomb of
Christ was shrouded with gold-embroidered hangings
(Itin. 25.7).

Page 139
wish of myself and Seirem, so that both I and Seirem
and everyone in the world may have hope in your
power and still trust in you.27
According to Evagrius, who was primarily interested in
ecclesiastical affairs, Khusrau did not send the first offerings directly to the shrine, but to Gregory, the
Chalcedonian patriarch of Antioch, who personally,
after receiving Maurice's blessing, dedicated them to
the saint at Rusafa. The passage in Evagrius makes
clear that this process guaranteed that the Iranian
king's gifts would be delivered to the saint by
Chalcedonian hands. By sending his offering first to
Antioch, as diplomacy must have dictated, Khusrau
acknowledged Roman control of the shrine, for the
crosses were more than private ex-votos: they were
symbols of the Iranian king's alliance with the widely
invoked frontier saint. Khusrau's offerings joined those
of countless other devotees, many of whom were responsible for the maintenance of Roman control of the
region. Sergius's reputation among the Arab tribes
who inhabited the territory joining the two empires had
helped to create a Syro-Mesopotamian world that on
one level ignored political allegiances. But the same

457/754

Arab tribes that were joined through the cult of Sergius were often drawn into the political conflicts
between the two empires.
Khusrau's dedications to S. Sergius need also to be
seen in the context of political tensions in Iran, where
Iranian Christianity was evolving an important political
as well as cultural identity. Reverence for the rider
saint Sergius was particularly fervid in Iran, where dissemination of the cult was aided by the way in which
Sergius fitted so comfortably into the traditional heroic
ideal that permeated Iranian culture.28 Through his
elaborate donations, Khusrau was not only associating himself with the Arab and Iranian devotees of Sergius. In their thank offering of a gold cross for Shirin's
conception of a son, Khusrau and his wife also
stepped deliberately into the place of Justinian and
Theodora, whose gold cross Khusrau had returned
with his first dedication.29 The parallelism between
the royal couples extended also to building projects,
since after his return to power Khusrau constructed
three churches in his empire: one for the Apostles,
one for the Theotokos, and one for the martyr Sergius.30 Although no source states

458/754

27. Theoph. Sim., Hist. 5.14.211 (tr. Whitby and

Whitby, with alterations); cf. Evagr., HE 6.21. The intrusion of the Greek editor is especially evident when
Khusrau refers to himself as a "Hellene," i.e., a nonChristian.
28. See Wiessner in Festgabe, 14155, esp. 15055.
29. In the context of the interest that Khusrau and
Shirin showed in S. Sergius and the Holy Cross, it
should be recalled that in 559, Bishop Abraamios
made a dedication to the Holy Cross, probably in the
basilica A complex (see chapter 3).
30. Flusin, Saint Anastase, 2: 1016. Accounts of the
churches vary. Mich. Syr., Chron. 10.23 (tr. Chabot, 2:
372), lists the Mother of God, the Apostles, and the
martyr Sergius; Agapius,

Page 140
explicitly for which of his two Christian wives, Maria or
Shirin, the Sergius church was built, in the context of
the miraculous bond linking Khusrau and Shirin to S.
Sergius, and manifested in their votive offerings to S.
Sergius at Rusafa, it was most probably for Shirin.31
The Nestorian Chronicle of Seert is somewhat more
specific, agreeing that Khusrau built three churches,
but adding that two were for Maria and one large one
(together with a palace) in the region of Beth Lashpar
for Shirin the Aramaean.32 The palace can be identified with the extensive ruins of Khusrau's summer
palace, which give their name to the modern Kasr-i
Shirin, Shirin's palace. Thirty-three kilometers northwest of the ancient city and important Christian center
of Hulwan, the palace was sited beside the river Hulwanrud and on the important caravan route between
Ctesiphon and the Zagros plateau.33 Babai the
Great's History of the early seventh-century martyr Giwargis mentions a famous monastery dedicated to the
martyr Sergius, where the queen would celebrate the
saint's festival. This monastery was located in the Hulwan region and should no doubt be identified with the
church of S. Sergius built for Shirin by Khusrau II.34

460/754

Like Theodora, Shirin involved herself in court politics

and, in particular, in disputes between the nonChalcedonians, whose patroness she was, and rival
Nestorians.35 Khusrau, however, had his own quarrel
with the Nestorian catholicos Ishoyahb of Arzun
(58295), who had made the unfortunate decision to
remain in his see rather than follow Khusrau when the
deposed king was forced into exile in the Roman empire.36 In a move demonstrating to Maurice Khusrau's
respect for the Chalcedonians, and by extension for
the Roman emperor's religious stanceand at the
same time putting Khusrau's own treacherous catholicos in his placeKhusrau invited Anastasius, the patriarch of Antioch, to consecrate the new churches
between 593 and 595. The choice illustrates beyond a
doubt Khusrau's mastery of the intricate relationship
between Christianity and political allegiance in Roman
and Iranian relations. When Ishoyahb died in 595, he
was succeeded at the helm of his Church by the elderly Sabrisho, whose reign was distinguished by good
relations with Khusrau, who even took part in

Kitab
, pp. 18687, speaks of one at Ctesiphon to
S. Mary and the other to S. Sergius the martyr; while

461/754

the Chron. 1234 171, mentions S. Sergius and the

Mother of God, both for his wife.
31. Flusin, Saint Anastase, 2: 1024, with n. 42, also
arrives at this conclusion.
32. Chronicle of Seert 58, pp. 14647 (Scher).
33. Streck and Lassner in EI2, 4: 73031.
34. Babai the Great, History of Mar Giwargis 518 (tr.
Braun, 258). See Hoffmann, Auszge, 120, for the
identification.
35. For a discussion of Shirin's role in Iranian religious
politics, see Flusin, Saint Anastase, 2: 10618.
36. Labourt, Christianisme, 2017.

Page 141
Sabrisho's consecration. The choice also pleased the
emperor Maurice, who sent an artist to Ctesiphon to
paint the pontiff's portrait. Maurice kept the portrait in
his treasury, along with holy relics and a hat belonging
to Sabrisho. As a sign of great favor, the emperor sent
the catholicos a piece of the True Cross contained in
a reliquary covered in jewels. In a foreshadowing of
his later translation of the True Cross to his capital
from Jerusalem, Khusrau intercepted the first piece of
the Cross that had been sent to Sabrisho, making a
gift of it to Shirin.37
After the fall of Maurice in November 602, Khusrau
again exploited his familiarity with tensions among
Christian groups, this time outside Iranian territory.
When he invaded and occupied all Syria and Mesopotamia, he decided, in order to undermine the
region's links with Constantinople, to refrain from requiring apostasy from Christianity. Instead, he offered
a choice between allegiance to the non-Chalcedonian
or the Nestorian Church, since the decision of
Chalcedon had come by the early seventh century to
stand for Orthodox Roman Christianity.38 NonChalcedonian Christianity was widespread throughout

463/754

Syria and Mesopotamia, in the villages, cities, monasteries, and among the tribesthat is, the same places
Patriarch Gregory had visited on his eastern tour. A
large part of the region's pastoral Arab population, in
particular those allied with the Banu Ghassan, was
non-Chalcedonian. Khusrau's aim in cultivating
Rusafa's saint and non-Chalcedonian Christianity was
to shift the political balance of Syria and Mesopotamia
in the Iranian empire's favor. Almost two centuries
earlier, Yazdgard had attempted to associate himself
with Marutha and the cult of martyrs at the strategic
site of Mayperqat. Khusrau I had supported
Ahudemmeh's cultivation of devotion to Sergius as a
means of monitoring Arab pastoralists who crossed
the frontier zone. The Arab tribes who inhabited and
defended the frontier zone were crucial for the maintenance of both empires' territorial claims. They were
allied with either Rome or Iran, but had been known in
the past to change sides, so by making his presence
felt at an important center such as Rusafa, Khusrau II
was also courting their favor, just as Justinian had
tried to do in the 530s and 540s.

464/754

THE GHASSANID CONFEDERATION

Any investigation of Ghassanid involvement with the
Sergius cult in Syria, and at Rusafa in particular, runs
up against serious difficulties with the
37. Chronicle of Seert 67, pp. 17277. On the Cross
in Ctesiphon, see Flusin, Saint Anastase, 2: 17072.
38. Agapius, Kitab
, pp. 198200, referring to the
terms offered at Edessa. For the broader historical
context of the association of the Chalcedonian position with the Roman Empire, see Fowden, Empire to
Commonwealth, 123.

Page 142
sources. Irfan Shahd has labored to fill in the gaps of
evidence and reconstitute a living picture of Ghassanid control in Syria, Palestine, and Arabia.39 Because
of his particular interestwhich he shares with the
sources of the periodin the political and religious relations between Rome and its Ghassanid allies,
Shahd has focused his attention on the Ghassanid,
specifically Jafnid, ruling elite. As Shahd has often
emphasized, the leaders of the Ghassanid confederation, those whom Nldeke called the Princes of the
House of Jafna, were settled Arabs interested in the
horse and the hunt, occupations closely linked to their
role also as a mobile fighting force. There is no reason to claim, in contradiction to Shahd, that the Jafnid
elite followed their own flocks in the manner of Arab
pastoralists. But on the other hand, it cannot be overlooked that they were leaders of a confederation that
spanned a region including rich agricultural land, villages on the fringe of the steppe, and arid steppeland,
swathes of which were never cultivable. The present
study of Ghassanid interest in the Sergius cult will endeavor to bring into the picture the pastoral dimension
of the Ghassanid confederation, which our literary
sources, on the whole, leave to one side.

466/754

Shahd has underscored the sophistication with which

the Ghassanid elite conducted their relations with other political and religious leaders.40 What is commonly
called the Ghassanid confederationproperly speaking, the body of Arab allies united under the leadership of the Jafnid dynasty of the Banu Ghassan, originally from South Arabia41performed a twofold
service for the Roman empire. On the one hand, it
guarded the frontier zone against external penetration, in particular by Iran's Lakhmid allies. At least
equally importantbut much less well documentedwas the Ghassanid federation's role in the
internal policing of Roman territory, in particular of nomadic groups in relation to each other and to the
settled populations. Unfortunately, little is known in
detail about the political and military organization of
this confederation, although it seems that members of
the Jafnid dynasty may have been responsible for
particular regions and the tribes based there.42
In addition to the culturally diverse political world in
which the Ghassanid leaders exercised their admirable diplomatic skill, it is also necessary to see them
in the Arab context, which embraced Arab pastoralists

467/754

as well as farmers, merchants and the ruling elite.

Ghassanid power in the eyes of
39. Shahd, BAFIC and BASIC, 1. For a rich but utterly neglected account of the Ghassanids, see Musil
et al.,
, 13051.
40. For Ghassanid involvement with the antiChalcedonian cause in Syria, see, with discretion,
Shahd, BASIC, 1: 691995; on the Sergius cult in
particular, see 94962.
41. Robin, CRAI 1996: 698 n. 118, stresses that it is
by no means certain that the entire tribe of Ghassan
migrated north to Palestine and Syria.
42. Donner, Early Islamic Conquests, 4344; Shahd,
Byzantium and the Semitic Orient I, 6365.

Page 143
Rome was founded both on the speed of their cavalry
and on their knowledge of the land. Such intimate
knowledge is rooted in the world of steppe pastoralism. It is to the further credit of al-Harith's diplomacy
that he understood when to play on Roman cultural
stereotypes of Arab life, which were certainly familiar
to him. Perhaps the best example of this is an incident
preserved by Michael the Syrian, which probably took
place in 536 / 37 at a meeting between al-Harith and
the notoriously zealous Chalcedonian patriarch of Antioch, Ephrem.43 The latter was pressing al-Harith to
enter into communion with the Chalcedonian Church,
an invitation the Ghassanid patron of the nonChalcedonian cause repeatedly refused. Finally, in order to drive his point home, al-Harith invited the patriarch to a banquet at which camel meat was the only
fare offered. When, as expected, Ephrem was
repelled, al-Harith seized the opportunity to compare
the hierarch's disgust with his own at the thought of
taking communion from Chalcedonian hands. alHarith's coup worked because it tapped into deep-running prejudices against the nomadic Arab way of life,
in which the consumption of camel meat was a mainstay.44 Even though there is no evidence to suggest

469/754

that al-Harith was himself a pastoralist, to assume that

he was somehow remote from this nomadic way of life
is to sever an important bond that linked the Ghassanid elite to the varied constituency the confederation
embraced. Because of the biases of the written
sources, it is only with great difficulty that a picture of
the diversity of the Syro-Mesopotamian and, specifically, the Ghassanid sphere can be recaptured. But
that diversity contributed to the important role played
by Rusafa as a fixed center in the steppe, and every
attempt must be made to reach a fuller understanding
of the ways in which settled and nomadic groups interacted in this region.
Literary evidence records a number of Ghassanid-associated churches south of Damascus that were dedicated to S. Sergius. In the sixth century, there was a
non-Chalcedonian monastery dedicated to S. Sergius
at
near Jilliq, some 20 kilometers south of
Damascus and the site of a Ghassanid residence.45
At the same time, another non-Chalcedonian monastery dedicated to S. Sergius flourished at Jabiya (Gabbitha in Syriac) in the Jaw-

470/754

43. Mich. Syr., Chron. 9.29 (tr. Chabot, 2: 24648).

See Shahd, BASIC, 1: 74655, for a discussion of
the passage.
44. Cf. Theoph., Chron., A.M. 6122, p. 333: Jewish
converts to Islam realized that Muhammad was not
the Messiah when they saw him eating camel meat.
On the unflattering and common identification of
Arabs as camel eaters in Greek literature, see Cook,
13 (1992): 10, with n. 55. For the provision of
meatmutton, goat, and camelto the settled population by Arab pastoralists in the modern period, see
Jabbur, Bedouins of the Desert, 35.
45. "Letter of the Archimandrites," in Documenta 221
(tr. Chabot, 154). The precise vocalization of the site
is not known. Lamy published an earlier edition
and translation in Actes du onzime congrs international des orientalistes (1898): 11737. See also
Nldeke, ZDMG 29 (1875): 427. For discussion of the
Documenta, a redaction of non-Chalcedonian
documents

Page 144
lan, the celebrated hirta and principle residence of the
Jafnids.46 This monastery was perhaps built by alHarith.47 In 587, the phylarch Jafna offered this
church of S. Sergius as the site for a meeting between
the quarreling non-Chalcedonian patriarch Damian of
Alexander and the dethroned patriarch of Antioch,
Peter of Callinicum.48 Jabiya, which spread out over
several hills, was the most famous of many sites associated with the Ghassanids in the Hawran. The area
was renowned for its mild climate, plentiful grasses,
and perennial waters, as the name suggests: jabiya in
Arabic indicates the shallow, man-made depressions
from which animals would drink water supplied from
wells or springs.49 The situation of the Ghassanid-allied Arabs of the Hawran bears close resemblance to
that created by Ahudemmeh and Maruta of Takrit, in
which pastoral Arabs were linked through their migratory patterns to the life of monastic communities.50
The pastoral as well as settled Arabs would have
provided gifts and protection to the monasteries,
which, in turn, served as permanent centers of spiritual sustenance and venues where the Arabs could exhibit their influence but also as bases in the Hawranian agricultural zone where many pastoralists would

472/754

spend the summer months. Active Ghassanid patronage of non-Chalcedonian monasteries in the Hawran
is explicitly attested in the "Letter of the Archimandrites," signed in 570 by 137 non-Chalcedonian abbots, priests, and deacons (fifteen of whom were
named Sergius) in condemnation of the "pseudopastors" of the tritheist heresy. This document provided
the raw material for Nldeke's groundbreaking study
of Hawranian topography in 1875, in which he
mapped the monasteries and sites with known Ghassanid associations. The past century of scholarship
has fine-tuned Nldeke's work, and Irfan Shahd's recent study of the document has further clarified our
picture of how the association between the Ghassanid
confederation and the non-Chalcedonian monasteries
worked, a potentially important parallel for Rusafa.51
We can glimpse this overlapping of settled and pastoral life in the Hawran

dating between 535 and 580, with the majority

between ca. 560 and 568, see van Roey and Allen,
Monophysite Texts, 267303, esp. 290 on the socalled Letter of the Archimandrites.

473/754

46. "Letter of the Archimandrites," in Documenta 215

(tr. Chabot, 149); Mich. Syr., Chron. 10.22 (tr. Chabot,
2: 367); Nldeke, ZDMG 29 (1875): 430; Aigrain in
DHGE, 3: 121819.
47. Shahd, BASIC, 1: 958.
48. Mich. Syr., Chron. 10.22 (tr. Chabot, 2: 36470).
See also Shahd, BASIC, 1: 932, and id., ARAM 5
(1995): 491503.
49. Conrad, BMGS 18 (1994): 46 n. 131; EI2, 2: 360.
50. See chapter 4.
51. See Shahd, BASIC, 1: 82443, esp. n. 118, on
the problems of transliteration. See also Sartre in
Trois tudes, 17788, a study of the Ghassanids and
non-Chalcedonian monasteries primarily from a topographical point of view.

Page 145
in poetry composed by Hassan ibn Thabit (ca.
570ca. 659), who spent time in the company of the
Ghassanid princes of the house of Jafna before offering his talent to sing the Prophet's praises shortly after
622.52 "Its banners descended on Busra, and in
Rumah like whirlwinds it left the smoke of its blazing
passage": the image of this opening verse of the
shorter of two poems composed by Hassan between
590 and 610 on the theme of the plague, comes
straight from the open steppe, conjuring up the dust
devils that have struck the imagination of countless
travelers. Hassan animates the plagueit is the jinn,
the demons who overturn both daily routine and annual celebration, bringing calamity upon laborer and
prince alike. His lament takes on the texture of the
landscape with its variety of inhabitants: he follows
after the bereaved, naming their villages, recalling
their tents nearby, their herds of horses, thereby tracing the epidemic's path, but also the Ghassanid presence in the Hawran:
To whom belongs the abode rendered desolate in
,
From the heights of al-Yarmuk unto al-Khamman,

475/754

Then al-Qurayyat down from Balas, then Darayya,

Then
, then the compounds [qusur] close by,
Then the hinterland of Jasim, then the wadis
Of al-Suffar, where herds of horses and fine white thoroughbreds feed?
That was the abode of him whose tent is raised on lofty p
The one dear unto me beyond the measure of his friendship and favour.
Their mother was bereaved, and was bereaved of them
On the day they stopped to camp at Harith al-Jawlan.
Easter approached, so the young girls
Sat to arrange garlands of coral,
Gathering saffron in finely perforated cloths,
Wearing saffron-dyed gowns of linen,
And not busying themselves with resin, nor with gum,
Nor with the seeds of the colocynth.
That was a home of the alJafna when calamity struck,
As the vicissitudes of the ages claimed their due.
I did indeed consider that there I behaved as a resolute man should,

476/754

When the place where I sat and stayed was in the pres
ence of the wearer
of the crown.53
All the toponyms are identifiable and all represent
known haunts of the Ghassanids. They form a triangle
that closely follows the major, paved Roman roads
from Damascus in the north to Dayr al-Kahf southeast
of Bostra,
52. Conrad. BMGS 18 (1994): 1258.
53. Hassan ibn Thabit, Diwan 1.255, no. 123 (tr. Conrad, BMGS 18 [1994]: 30).

Page 146
continuing along the
road as far west as
Gadara / Umm Qays, moving northward again
through the Jawlan past Jabiya and Jasim, and back
up to the Ghuta plain of Damascus.54 The sites that
Hassan and an anonymous reviser of his poem mention share certain important characteristics besides an
association with the Ghassanids and a position on
main routes. Most command a defensive advantage,
are near water and fertile fields and include a monastery, thus confirming the association suggested in the
"Letter of the Archimandrites." Sufferers include those
living in tents and in qusur, which Conrad translates
here as "compounds" rather than the more traditional
"castle," "palace," or "fortress." Spreading south from
Damascus, the plague devastated the fertile Ghuta,
described as a patchwork of villages and qusur. While
Hassan evokes the "wadis of al-Suffar, where herds of
horses and fine white thoroughbreds feed," a contemporary reviser of his poem substitutes the "buildings of
Suffar."55 This alternative is not, however, incompatible with the pastoral image. Encampments included a
wide variety of structures, from the heavy basalt
houses and churches that have outlived centuries of

478/754

inhabitants in the Hawran, to more ephemeral structures that usually elude archeologists.56
To the number of Ghassanid-associated sites known
from the "Letter" and from Hassan ibn Thabit's poems,
archeological survey may have recently added another. Only the site's modern name is known, al-Ramthaniya,57 located on well-watered ground surrounding
a volcanic summit in the east54. See the map in Conrad, BMGS 18 (1994): 22.
55. The variant reading is recorded in
1.257.

edition,

56. Rather than associating permanent buildings more

or less exclusively with the sedentarization of pastoral
nomads, it is helpful to look at the use made by beduin in the modern period of permanent structures. For
example, a shaykh of the Banu Sakhr in the 1920s
built what was called a qasr in the Syrian steppe. This
residential compound is described as follows in Lewis,
Nomads and Settlers, 231 n. 30: "[P]art storehouse,
part family and tribal headquarters, they [the qusur]
were used by the shaykhs particularly to entertain
guests and as the venue of tribal gatherings." See

479/754

also 139. Usually, the skaykh lived with his family in a

tent pitched nearby.
57. Note the older transcription al-Rumsaniyya in
Schumacher, Jaulan, 23135, 292; Oliphant, PEQ
(1886): 81. I would like to thank Claudine Dauphin for
providing me with offprints. A full appreciation of the
interesting work of Dauphin and her team is frustrated
by the absence of a detailed plan of the ecclesiastical
complex on the summit. Only a schematic plan of the
entire settlement and field systems, as well as photographs of limited use, are available in print. For a general plan, see Dauphin in Early Christianity, ed.
Manns and Alliata, 70, fig. 2, and id. in Akten, 668, fig.
1, with pl. 83a, a photograph of the summit showing
the monastery and martyrium from afar. See also
Dauphin, Archologia 297 (1994): 5264; 294 (1993):
5057; Gibson and Dauphin, BA-IAS 12 (199293):
731; Dauphin and Gibson, IEJ 41 (1991): 179; Gibson and Dauphin in Archologie et espaces, 3545;
Dauphin, IEJ 36 (1986): 27475; 34 (1984): 26869;
33 (1983): 11213. For publication of the Greek inscriptions, see Gregg in Dauphin et al., POChr 46
(1996): 32428.

Page 147
ern Jawlan, approximately twenty kilometers due west
of Jasim.58 A spring rises in the northwest sector of
the settlement, and at the eastern foot of the volcanic
summit is a built pool. The dense assortment of secular and ecclesiastical buildings at the site includes facilities for storing water and corralling animals. Enclosures have been discovered on the northwest,
north, and northeast of the settlement, and traces of
fields marked out by boulders stretch out toward the
west and southwest.59
At the highest point of the settlement is a large rectangular structure with ancient foundations, but rebuilt at
various periods, in part with reused cut stones, some
decorated in relief with vines, palm trees, crosses,
and geometrical designs. A Greek inscription carved
on the far upper right-hand corner of a basalt lintel
gives a date in the Seleucid era of 685, that is, A.D.
37374.60 This date would seem to mark the erection
of the building, while another Greek inscription from
the same building identifies the structure as a martyrium of S. John the Baptist and is dated June 377.61
This longer inscription of eleven lines records that the
illustrius ordinarius Flavius
, who had been

481/754

responsible for a military division, built the martyrium

after completing his military service. The inscription's
date, 377, may record the consecration, since it
postdates the first inscription by at least three
years.62 The presence of relics of S. John the Baptist
at al-Ramthaniya has been plausibly explained in the
context of the destruction of the shrine dedicated to
the prophet and saint at Sebaste under the emperor
Julian. Once the relics were dislodged from their original cult site, it is not difficult to imagine that some of
them ended up in the Jawlan. Another inscription, discovered on the ground outside the western entrance
by a late nineteenth-century traveler who published a
drawing, records the restoration of the martyrium by a
clarissimus Balbion.63 The rebuilding included wall
carvings in high relief and a mosaic floor. The restoration is undated, but on the basis of the letter forms,
the survey team has assigned it to the
58. For al-Ramthaniya's proximity to Jabiya and
Jasim, which is not mentioned in any discussion of the
site, see Dussaud, Topographie, map 1.2D for
"Roumsaniy" and 2.2A for "Djabiya" and "Djasim."
59. Dauphin, IEJ 36 (1986): 274.

482/754

60. Gregg in Dauphin et al., POChr 46 (1996): 326,

no. 26, with drawing in VI, fig. 20.
61. Ibid., 32526, no. 25, with photograph in XV, pl.
Va.
62. Dauphin in Dauphin et al., POChr 46 (1996):
33536.
63. Gregg in Dauphin et al., POChr 46 (1996): 327,
no. 28, with drawing 7, fig. 22 and photograph at XVI,
pl. VIa. See also the brief discussion by Dauphin in
Dauphin et al., POChr 46 (1996): 33637. Schumacher, who published his work in 1888, discusses the site
briefly (Jaulan, 23135); fig. 125 is a drawing of the
inscription without translation or commentary; see
also fig. 117, for a plan of the building now on the site
of the martyrium. Oliphant, PEQ (1886): 81, also includes a drawing of the inscription. On use in the
eastern provinces of the title clarissimus, see Shahd,
Byzantium and the Semitic Orient III, 32322.

Page 148
sixth century. In addition to the inscriptions that clearly
identify the original building and its rebuilding by Balbion as a martyrium, the Christian nature of the building is also indicated by the many stone blocks carved
with Christian symbols, some of which are scattered
near the site, while others were incorporated in later
reconstructions of the building.64 That the present
rectangular construction is a product of the Islamic
period is suggested by the pointed internal arches and
the plan, which is unlike that of any known Christian
martyrium, with no eastern apse, entrances on the
east and west sides, and a long, narrow annex attached on the north side.65 A reused block from the
vicinity seems to have been fitted out to hold an icon
of the saint, since over an oval recess in the stone the
name of John was carved and adorned with a vine
trellis.66
John the Baptist was commemorated on several occasions throughout the year, the two most popular
falling in the summer (his birth on 24 June and his beheading on 29 August), when the Hawran, and related
Jawlan, would have been host to nomadic groups
pasturing their flocks in the dry season. The remains

484/754

of al-Ramthaniya suggest that the martyrium of John

was part of a cluster of related buildings, perhaps including a funerary chapel to the west, a ring of natural
caves to the southeast, and a monastery arranged
around a courtyard to the south.67 This group of ecclesiastical buildings dominated the settlement. It was
surrounded by an enclosure wall and linked to the enclosures southwest of the settlement by a processional way, which can still be made out. Evidence of ancient tent enclosures mixed in with a few permanent
structures suggests that pastoralists would have
pitched camp just beyond the settled area in order to
exploit the pasture and water supply around alRamthaniya.68 Additionally, they would have been
drawn to the feast of the saint, perhaps also for baptism, as at the shrine of Symeon at
and of
Sergius at Rusafa, where groups of Arabs gathered
for that purpose.69 Discovered in the area northeast
of
64. Schumacher, Jaulan, figs. 11825.
65. Dauphin, IEJ 34 (1984): 268, states that the rectangular building represents the western end of the

485/754

martyrium. The existence of a late antique martyrium

on the site is clear from the inscriptions.
66. Gregg in Dauphin et al., POChr 46 (1996):
32627, no. 27 with VI, fig. 21; XV, pl. Vb. See also
Dauphin, IEJ 34 (1984): 268. Cf. the similar arrangement at Kafr Antin in northern Syria (see above, p.
118).
67. Dauphin in Dauphin et al., POChr 46 (1996): 337;
Dauphin, IEJ 34 (1984): 268.
68. Dauphin in Akten 67173; Dauphin, BA-IAS 8
(198889): 834; Dauphin, IEJ 36 (1986): 274. For a
discussion of the archeology of nomadic enclosures,
see Rosen, J. Field Archeology 20 (1993): 44151,
with useful drawings and a consideration of a camp in
the Negev within the wider context of nomadism in the
surrounding region.
69. Sev. Ant., Hom. cath. 57, p. 93. Dauphin does not
draw attention to parallel evidence for nomad baptism
at great pilgrimage sites, although she does conjecture that a pool of undisclosed dimensions, located in
the area of the enclosures, would have been used for
mass bap-

Page 149
the pool, perhaps at the head of the processional way,
is what seems to have been an open-air shrine, with
stones set up to mark the perimeter and an apse with
an upright stone at the eastern end.70 The suggestion
that the stone served as an altar for the seasonal
shrine should be treated with caution, since it would
be highly unusual to leave an altar exposed to the elements and potential desecrators. More likely it would
have marked the correct orientation for worship.
The seasonal nature of nomadic association with alRamthaniya should be underlined more than it has
been in the published reports, which assume that nomadic presence at a settlement implies a gradual
sedentarization. As we have seen in the case of the
Arab interaction with monasteries recorded in the History of Ahudemmeh and the History of Maruta, the
popularity and influence of certain cult sites in Syria
and Mesopotamia depended to a significant extent on
the mobility of groups in the region, especially the
Arab pastoralists. The evidence at al-Ramthaniya, in
particular, ought to be seen in the context of the site's
close proximity to the important Ghassanid centers at
Jabiya and Jasim, in a strategically important zone

487/754

between the Syro-Mesopotamian steppe and the

Mediterranean coastal cities. The patronage of Flavius
and Balbion may reflect their attempts as
leaders in the region to capitalize on the link between
cult sites and pastoralist movement. Despite the
shortcomings of our present knowledge of alRamthaniya's architectural history, the accumulation
around the site's water supply of a martyrium, enclosures, fields, and, it seems, a small monastery, points
toward its inclusion among sites associated with the
Ghassanid confederation. Like Hassan ibn Thabit's
poems, the evidence from al-Ramthaniya reveals the
wide spectrum of pastoralists and agriculturalists that
flourished in the Hawran and Jawlan. But what is
more, these examples reinforce the importance in nomadic life of permanent centers, especially cult centers, a relationship political leaders in the region were
keen to control.

AL-MUNDHIR
Outside Rusafa's northern gate is a cemetery in which
stands a small cross-in-square stone building, measuring 17 m in width and 20 m in length

488/754

tisms. Without more evidence about the context and

construction of the pool, the suggestion cannot be
evaluated. Even if the pool is more likely to have been
intended for storage of water for animals than for
Christian ritual, the suggestion that mass baptisms occurred should be taken seriously, since these could
easily have been facilitated by a movable font.
70. Dauphin, Archologia 297 (1994): 64; Dauphin
and Gibson, BA-IAS 12 (199293): 2829; and
Dauphin, IEJ 36 (1986): 274, pl. 35c, a photograph of
the stone at the eastern end of the temenos.

Page 150

Figure 11. Rusafa: al-Mundhir building, interior from

west.
(fig. 11).71 The building, positioned 150 m northeast
of the gate, stands on an eminence east of the main
road leading into the town. Entering through the main
door, which is located on the building's west side, one
sees four central piers whose capitals are decorated
with the split leaf design typical of northern Syria and
Mesopotamia. The central space was probably
covered by a pyramidal wooden roof, while barrel
vaults covered the main axes. Pendentive domes still

490/754

today cover the small bays in the building's four

corners. At the eastern end an apse was flanked on
either side by a separate, small rectangular room. The
arch that spans the apse is supported on piers to the
south and north, and the corner capital of the southern pier was adorned with a bird with outstretched
wings, now much damaged.
71. I would like to thank Thilo Ulbert who, over the
course of several years, has responded to my queries
about the excavations at Rusafa and has read and
commented on earlier versions of my discussion of
the al-Mundhir building. I am also indebted to Gunnar
Brands, who provided extensive comments on the discussion of the al-Mundhir building in my dissertation,
"Sergius of Rusafa: Sacred Defense in Late Antique
Syria-Mesopotamia" (Princeton, 1995). The argument
that I advanced there concerning the building's identification remains unchanged here, but is supported by
further evidence. Neither Ulbert nor Brands should be
held responsible for my conclusions. For a much
needed reconsideration of the building's architecture
and decoration, see Brands, DaM 10 (1998), 21135.
Also, Fowden, DaM 11 (2000).

Page 151
Crosses carved on some of the capitals were later
lopped off.72 Other crosses were scratched on the interior walls of the building. A cross in relief marked the
apex of the apse and it is very likely that another was
carved on the now seriously damaged medallion at
the center of the molding over the double apse window. The upper band of the molding is decorated with
an elaborate and finely carved frieze with marine motifs. The lower band bears a Greek inscription:
, "the fortune of Alamoundaros triumphs," or, more loosely, "Long live al-Mundhir."73
The inscription refers to the Ghassanid leader alMundhir, called Alamoundaros in Greek, who held the
office of phylarch from 570 to 581.
Standing just outside the gypsum walls of Rusafa, the
al-Mundhir building occupies a space between the
densely built-up pilgrimage center and the open Syrian steppe (see title page). This building is important to
the study of the Sergius cult for various reasons. First
and foremost, thanks to the inscription, the structure is
a concrete link between the Ghassanid phylarch and
S. Sergius at the martyr's primary cult site. The
building's presence at Rusafa is an indication of the

492/754

geographical range of Ghassanid involvement in

Syria. Secondly, the phylarch's building, I argue, offers another example of S. Sergius's mediating power
in this region. This point in particular requires some
explanation, which must begin with the task of identifying the structure's purpose.
In 1939, Jean Sauvaget published an interpretation of
this building that has been almost universally accepted ever since.74 The success of Sauvaget's views
owes much to the author's familiarity with the landscape and the early Islamic cultural history of Syria.
Excavation of the monument began over a decade
ago, and a full publication of the building is now being
prepared by Thilo Ulbert. While a reexamination of the
evidence may, in the circumstances, seem premature,
there are still many points that can usefully be
72. Guyer in Sarre and Herzfeld, Archologische
Reise, 43; Spanner and Guyer, Rusafa, 66.
73. Guyer discussed Rusafa's architecture in Sarre
and Herzfeld, Archologische Reise, 2: 3943; fig.
156 illustrates the inscription; fig. 152 is a plan and fig.
154 a reconstruction of the building with pyramidal
roof. Spanner and Guyer, Rusafa, also publish a plan,

493/754

pl. 31, and the inscription, fig. 16. For discussion of

the accentuation of the acclamationwhich should
here be taken as the present indicativesee Cameron, Porphyrius, 7680. See now also the discussion
of the inscription by Shahd, BASIC, 1: 5015, esp.
503 n. 354, with photograph opposite 502. On the
basis of the inscription's relationship to the decoration
on the molding, Sauvaget, Mosque omeyyade, 159
n. 2, concludes that the inscription is original and excludes the possibility that it could be a later graffito.
74. Sauvaget, Byzantion 14 (1939): 11530. The thorough dissemination of this theory is remarkable. To
mention only two recent examples from books with a
wide readership, Cameron, Mediterranean World,
181, writes that Rusafa was "the main Ghassanid
centre," while Isaac, Limits of Empire, 256, asserts
that "an inscription shows that a modest building north
of the town served as praetorium for the Ghassanids."
As we shall see, the inscription itself "shows" nothing
with regard to the building's function as either a praetorium or anything else.

Page 152

Figure 12. Rusafa: Plan of al-Mundhir building.

discussed, even if new archeological findings, when
published, may impose some modifications. My interpretation of the function of al-Mundhir's building outside Rusafa's walls owes much to Sauvaget's emphasis on the site's function as a meeting place for the
Ghassanid confederation. However, my understanding of the relationship between the sacred and the
profane, as it is made manifest in the structure in
question, diverges from his and, consequently, leads

495/754

to a difference of opinion regarding the building's identification. Attention must be given to the building's details, some of which were overlooked or misrepresented by Sauvaget. These details can contribute to a
wider understanding of the place of al-Mundhir's building within the context of steppe architecture, but also
in relationship to the history of early Islamic architecture and culture.
Sauvaget's first concern was the building's plan (fig.
12). Architectural historians from Guyer to Lassus understood the structure as a church and drew parallels
with several similar cross-in-square buildings in Syria.
Sauvaget discounted this identification because of
what he saw as three in-escapable problems: (1) its
plan is an anomaly in the context of Rusafa's

Page 153
other churches; (2) no other Syrian church possesses
an identical plan, and the only one that bears comparison, the chapel at Androna, is excluded because its
apse is much larger than that in the al-Mundhir building; (3) the inscription in the apse is out of place in a
church, funerary or not.75
Sauvaget's first objection, as he admits himself, does
not carry much weight. No canon stipulates that
churches in one place must be stylistically related. But
as it happens, excavation at Rusafa has brought to
light a very close parallel to the al-Mundhir building: a
cross-in-square building erected in the southeast part
of the basilica A complex. Like the al-Mundhir building, its orientation is east-west, with the apse at the
eastern end. The building was part of the original
plan, which is now dated to the last quarter of the fifth
century.76 This structure has been identified by Ulbert
as the baptistry of basilica A. As regards Sauvaget's
second point, a few years after he published his article, the conclusion that the building was not a church
was accepted by Lassus in his important book on the
Christian architecture of Syria. However, in his discussion of cross-in-square churches, Lassus included

497/754

sixth-century churches whose kinship with Rusafa's

building is difficult to discount.77 In particular, the
church at al-Tuba, one of the largest ruined towns of
the basalt plateau north of Salaminias, is dated to 582
and strikingly similar in plan, even in the apse's proportion to the rest of the building.78 The al-Mundhir
building and the church at al-Tuba were erected within
a span of ten years at the end of the sixth centuryand both buildings were erected roughly a century after cross-in-square building southeast of basilica A.
The decoration of the al-Mundhir building likewise falls
within the standard repertoire of fifth- and sixth-century church decoration. First, maritime iconography,
such as that carved on the molding above the apse
windows, commonly adorned churches throughout the
late Roman empire,
75. Guyer in Sarre and Herzfeld, Archologische
Reise, 42, suggested that the building was intended
as the phylarch's mausoleum, a view that at least
takes into consideration the building's context. This
view was also considered possible by Tchalenko, Villages antiques, 1: 262 n. 1.

498/754

76. For publication of the "Viersttzenbau I," see Ulbert, Resafa, 2: 98101, 12223, and fig. 62. For the
redating of phase one of basilica A, see above, pp.
8091.
77. Lassus, Sanctuaires, 14350, fig. 61.
78. On al-Tuba, see Butler, Early Churches, 16364,
with n. 299 by Baldwin Smith, who notes the similarity
between the plans of al-Tuba, Androna / Andarin, and
Rusafa's extramuros building. Despite Sauvaget's objection, apse size has not excluded Rusafa's building
from consideration among cross-in-square churches:
see, e.g., Tchalenko, Villages antiques, 1: 262 n. 1.
Mango, Byzantine Architecture, 52, with figs. 7475,
concludes that the size of the apse"a little too small
with regard to the proportions of the building"is not
sufficient to disqualify the building as a church, although for Mango, as for Sauvaget, the inscription
makes the identification as a church impossible. For
other parallels among Syrian churches, see now
Brands, DaM 10 (1998), 21135.

Page 154
including basilica B and the tetraconch at Rusafa.79
At the most general level of interpretation, the scenes
represent the bounty of creation or allude to Paradise.
The bird carved on the southern apse capital is now
damaged, but was most likely an eagle, also characteristic decoration in churches of the period.80 Like
the marine scenes, the representation of an eagle, the
bird that soars higher, closer to the sun than all others, can be variously interpreted.81 The eagle may,
for instance, symbolize the presence of angels, witnesses to the worship offered therea particularly apt
interpretation since the eagle appears on a capital belonging to the apse. So often a symbol of the sun or
the supreme god in pre-Christian art, the eagle in
Christian iconography may represent the fusion of celestial and earthly power, the two sources of authority
that overlap in a princely church. Two contemporary
parallels to the use of the eagle in the al-Mundhir
building, one from Lebanon, the other from Syria and
both from churches, help to illustrate this point. First,
the well-preserved mosaic pavement of a church at alGhina, located in the mountains 11 kilometers southeast of Byblos / Jbayl, is decorated with geometric
designs in the side aisles, while medallions with fish

500/754

and animals cover the floor of the central nave. On a

raised podium in front of the sanctuary is portrayed a
lone eagle, depicted frontally with wings widespread.82 As at Rusafa, the eagle is given a prominent position at the threshold between the nave and
the sanctuary. Presumably, the eagle was chosen because of its own liminal nature between heaven and
earth. Again in the second example, the Adam mosaic
from the north church at Huwirta (Huarte), the eagle
appears near this potent threshold.83 At Hu79. See Kollwitz, AA 1963: 346, fig. 13, for a cornice
in the tetraconch that shows a scallop shell flanked by
dolphins. The marine decoration from basilica B will
be published by Brands in Resafa, vol. 6
(forthcoming). Brands provides further examples of
marine imagery from ecclesiastical contexts in DaM
10 (1998), 22731.
80. A phoenix has also been suggested. Brands, DaM
10 (1998), 227, does not rule out this possibility, but
prefers the eagle interpretation. On eagles in the context of the al-Mundhir building, see ibid. 22527.
81. It has been suggested in a recent discussion of
the reused material in the al-Aqsa mosque in

501/754

Jerusalem that the representation of an eagle on a

capital identifies the sculpture's original context as a
church (Wilkinson in Bayt al-Maqdis, 1: 129, 13435).
For a range of interpretations of the eagle in the
Christian context, see Schneider, RAC, 1: 9194;
Maguire, Earth and Ocean, 6566, with notes. Eagles
appear frequently in both non-Christian and Christian
art in the Hawran and the Jawlan, familiar territory for
the Banu Ghassan. For eagles in the basalt synagogues of the Jawlan, see Hachili in Roman and
Byzantine Near East, 1: 18586, 198201; and see
also Grabar, CArch 11 (1960): 62, figs. 15 and 19.
82. Donceel-Vote, Pavements, 34753, with figs.
33438; for further bibliography on the symbolism of
eagles, see 352 n. 11. Also on the eagle at al-Ghina,
see Stern, CArch 15 (1965): 35, who discusses parallel iconography on both the lintel decoration and floor
mosaic of a synagogue at Jaffa.
83. Donceel-Vote, Pavements, 10216, with figs.
7179 and pl. 5. Note that a phoenix also appears in
the upper southeast corner of the Adam scene.

Page 155
wirta, the eagle takes up its position of honor at the
right hand of Adam, just as the eagle in the al-Mundhir
building hovers over the right side of the sanctuary. At
Huwirta, too, the mosaic ensemble is located in the
central nave, just in front of the sanctuary proper. In
the al-Mundhir building, the eagle, straddling both celestial and terrestrial worlds, would have served to
represent the phylarch's secular authority, while at the
same time alluding to the heavenly authority behind
it.84
Sauvaget's positive identification of the building outside Rusafa's walls grew out of his discussion of the
al-Mundhir acclamation. In his opinion, the victory cry
excluded the possibility that this was a sacred building, and parallels therefore had to be sought in secular architecture. Sauvaget's preferred comparison was
the second-century so-called praetorium at Mismiya
on the northern edge of the Laja, whose apsidal recess bore close resemblance to that at Rusafa.85 The
year before Sauvaget published his article on the
building outside Rusafa, Edmund Weigand published
what is still regarded as the most exhaustive study of
the building at Mismiya.86 Weigand concluded that it

503/754

was a temple, and that the nearby temples at

Sanamayn and Slim offer the closest parallels to Mismiya.87 More recent considerations of the building by
Stephen Hill and
emphasize the similarities
between the Mismiya temple and later church architecture. Photographs of the building, which had been
destroyed by the beginning of this century, reveal a
very grand cross-in-square structure with an intricate
tripartite faade. While the second-century edifice
may have been an important link in the chain of architectural developments from the second to the sixth
centuries, it is difficult to deny that the impression left
on a visitor by the modestly adorned building outside
Rusafa would have been more akin to that of the
humble church at al-Tuba than to that of the building
at Mismiya. And there is one further consideration that
Sauvaget overlooked. Structural changes to the building at Mismiya indicate that it was converted into a
church, probably before 450.88 In the 570s or 580s,
when al-Mundhir had his name inscribed on the
Rusafa building, Mismiya was a church, regardless of
its original function.
Certain details of Sauvaget's theory also require modification. For in-

504/754

84. At the church of the Studite monastery in Constantinople, the eagle was associated with imperial
and consular rank, again as a reflection of heavenly
authority on earth (Kramer, Skulpturen, 8284); see
also Brands, DaM 10 (1998), 226.
85. Spanner and Guyer, Rusafa, 67, had already
drawn attention to the similarity between the Mismiya
and Rusafa buildings. For a recent reexamination of
the monument, see
, DOP 44 (1990): 4146, with
excellent photographs from the Dumbarton Oaks
archive.
86. Weigand in Wrzburger Festgabe, 7192.
87. See Freyberger, DaM 5 (1991): 138, for the type
of Roman temple to which the Mismiya temple originally belonged; see also id., DaM 4 (1989): 89 n. 15.
88. Hill, DOP 29 (1975): 349; de Vog and Waddington, Syrie central, 4546.

Page 156
stance, he assumed that the feast of S. Sergius was
celebrated on 15 November.89 But both the
Chalcedonian and non-Chalcedonian Churches celebrated the feast of S. Sergius in October, on the first or
the seventh. October first is recorded as the feast
date in 514 when Severus of Antioch delivered his
homily in honor of S. Sergius at Chalcis. A date in
October corresponds more accurately with the date of
the seasonal migration than does 15 November, since
already in early October, transhumants near the
steppe's fringe would be preparing to move to winter
pastures.90 Furthermore, Sauvaget's assertion that
the feast would have been celebrated according to the
non-Chalcedonian calendar (which in fact did not differ from the Chalcedonian, at least as regards S. Sergius) betrays a dangerous assumption that the church
hierarchy at Rusafa was dominated by non-Chalcedonians, simply because the site was associated with
the Ghassanids thanks to the al-Mundhir building. As
was noted earlier, the only bishop of Rusafa who was
certainly non-Chalcedonian was the Sergius who attended the council of Ramla under Justin I. After 536,
in Roman Syria and Mesopotamia, monasteries rather
than metropolitan sees were the usual power centers

506/754

for the non-Chalcedonian cause. It is a great misfortune that no clear evidence of everyday interaction
between supporters and opponents of Chalcedon exists from Rusafa.91 But it is reasonable to assume
that as a pilgrimage center, Rusafa would have seen
visitors of all persuasions. The Vita of S. Gulanducht
offers a hint that has not been taken: after emerging
from an instructive vision in which an angel shows
Gulanducht two cups, one (Chalcedonian) resplendent with light and the other (non-Chalcedonian) hidden by shadows, Gulanducht sets off to Jerusalem via
Rusafa, where she worships. The passage suggests
that Rusafa at the end of the sixth century was a holy
place acceptable to supporters of Chalcedon.92 It is
also instructive, in the absence of direct evidence relating to the cohabitation of supporters and opponents
of Chalcedon at Rusafa, that John of Ephesus records
a situation at Amida where both groups at89. In this, Sauvaget followed Charles, Christianisme,
33.
90. Macdonald, JRAS 2 (1992): 4 and 9; Schlumberger, Palmyrne, 131 with n. 2; and Musil, Manners
and Customs of the Rwala, 78.

507/754

91. The important study by Mundell (Mango) in Iconoclasm, 5974, recognizes the contemporary interest in
S. Sergius at Rusafa of both supporters and opponents of Chalcedon. More recently, Walter too has insisted that both Chalcedonians and non-Chalcedonians would have enjoyed Justinian's patronage at
Rusafa: REB 53 (1995): 375. Cf. the otherwise admirable Making of Orthodox Byzantium, where Whittow,
with less care, simply describes the shrine at Rusafa
as Sergius's "great Monophysite shrine" (4748).
92. Vita Gulanducht (Georgian redaction) 14.715.1;
the story exists also in the Greek summary of the Vita
Gulanducht made by the priest Eustratius before 602:
see Peeters, AB 62 (1944): 8889.

Page 157
tended the same church, but did not receive the
Eucharist from the same chalice.93 Another solution
for a mixed congregation at a holy site was found at
the shrine of Cyrus and John at Menuthis near Alexandria. There non-Chalcedonians appear to have imbibed oil taken from the lamp that hung over the holy
sarcophagus, as a sort of antidoron.94 Throughout
the region, S. Sergius was cultivated by
Chalcedonians, non-Chalcedonians, Nestorians, and
even non-Christians. No evidence indicates that any
group managed to monopolize the shrine at Rusafa.
Another important detail in need of further attention
concerns the site of the building. Sauvaget explains
that it lies outside Rusafa's main north gate, but mentions only in passing that it is surrounded by Christian
and Muslim graves.95 On his first visit in 1908, Musil
saw a large, undisturbed Christian cemetery extending east from the building, but when he returned in
1912, he found all the graves opened and looted.
Muslim graves northwest of the city walls had already
suffered the same fate at the time of his first visit.96
None of the ruined graves have been published, but
the only graves to be found near Rusafa are located

509/754

to the north of the city, extending east and west.97 A

ruined mausoleum, also unpublished, north of the alMundhir building has been informally dated by Ulbert
to the fifth century.98 No parallels to the situation outside Rusafa have been discovered at other
Ghassanid-associated sites, in the Hawran, for instance. It must be asked, then, why this building was
set among graves.
To find an answer, we must turn to the Passio of SS.
Sergius and Bacchus and to the overall plan of
Rusafa. Sergius, shod in spiked sandals, led a ducal
procession along the Strata Diocletiana from Sura and
Tetrapyrgium to Rusafa. Approaching Rusafa from the
north, the party would have entered the castrum
through the north gate. Sergius was brought one last
time before the dux Antiochus, who condemned him
to death by the sword. The young officer was then displayed before the people of Rusafa and taken to the
place of execution, where he was allowed a final prayer before his head was cut off. A chasm opened
where his blood fell to spare the holy site from
93. Joh. Eph., Vita. sanct. 1012.
94. Marcos, "Thaumata," 46 with n. 19.

510/754

95. Spanner and Guyer, Rusafa, pl. 1, show the necropolis to the north and northeast of the city walls on
their map. See also Brands, DaM 10 (1998), 220, on
the grave chambers in this region north of the walls.
96. Musil, Palmyrena, 165.
97. Although none of these graves have been systematically excavated, many of the coins found in beduin
digging in this grave area are said to be late antique.
98. Ulbert, letter to the author, 18 October 1996. Thilo
Ulbert and Stephan Westphalen plan to publish the
building in the near future. For a plan of the mausoleum, see Musil, Palmyrena, 211, fig. 83.

Page 158
desecration by unbelievers, and the martyr was buried
at the site of his execution, where countless miracles
occurred, even into the 430s or soon after, when the
Passio was written.99 The attempt by a party of Surans to steal the body was averted by the emergence
of armed defenders from within Rusafa's walls. Even
on the most skeptical reading of the story, it is clear
that in the fifth century, Sergius's grave was considered to have been located outside the city walls.
We are not told on which side of the settlement.100
As stated above, in addition to the presence of Christian graves, the northern cemetery's antiquity is suggested by a mausoleum of late Roman date located
north of the al-Mundhir building. Nor is any other
cemetery visible today. The author of the Passio
clearly implies that the holiness of the site of Sergius's
martyrdom persisted even after the saint's body was
translated within the walls for security reasons and
honored by a magnificent shrine. The memory of the
original site could easily have been preserved in oral
tradition for centuries after the translation. And the
grave of a miracle-working saint would naturally have

512/754

attracted the graves of those who desired to rest near

the holy tomb.101
A clue to the site of Sergius's original tomb may be
provided by the conspicuously large, decorated gate
in Rusafa's north wall. Each of its three portals is
flanked by two columns, which in turn support five
arches upon their exquisitely wrought acanthus capitals. The carving on the arch moldings is of the highest
quality and is dense with luxuriant floral and vine motifs. A carved cross marks the keystone of the central
arch. Considerable skill and effort clearly went into the
adornment of the north gate of this heavily fortified
settlement. Rusafa did occupy a strategically important site for the security of the frontier zone, and the
entrance was no doubt designed to impress all visitors
to the fortress and pilgrims to the popular shrine. The
significance of Rusafa as fortress and shrine might be
thought sufficient explanation for the grandeur of the
settlement's main entrance. But the northward orientation toward the Euphrates (or "the pacified parts of the
empire," as Krautheimer saw it)102 is a weak explanation of why the north gate in particular was chosen
for special elaboration. The elaboration of the gate

513/754

becomes more intelligible, however, if Sergius's original tomb was lo99. Pass. gr. 2829. On the date, see chapter 1.
100. The text simply does not support the curious assertion made by Woods, JECS 5 (1997): 36465, that
the grave was located midway between Rusafa and
Sura, more or less near Tetrapyrgium: see Pass. gr.
27, 29; Pass. syr., pp. 317, 32021.
101. On ad sanctos burials, see Duval, Loca sanctorum Africae, 50116. The belief that holiness lingered
at places that formerly housed relics is expressed in
the inscription from basilica B, which identifies the site
of the church that was begun in 518 by its earlier
status as the house of Sergius's relics.
102. Krautheimer with
antine Architecture, 262.

, Early Christian and Byz-

Page 159
cated north of the city. Then the impressive entrance
would have marked out Rusafa's sacred orientation.
All the gates, including the northern one, were built at
the same time as the walls, in the first third of the sixth
century.103 These walls embraced the buildings within the older walls and, in particular, the shrine built for
Sergius's relics by Alexander of Hierapolis, which has
been located beneath basilica B.104 Given the antiquity of the graves north and northeast of the north
gate, it is certainly possible that this northern graveyard grew up around the traditional site of the tomb of
Sergius, and that al-Mundhir's building stood on an
eminence among the graves in this area is because it
was at or near the site that local memory held sacred
as Sergius's original resting place.105
With the architectural background and position of the
building now placed more firmly in context, the meaning of the inscription can be addressed. Sauvaget's attempt to explain the common Greek acclamation at
Rusafa as a translation of an Arabic phrase is unnecessary in the context of al-Mundhir's deliberate adoption in the building of other Roman conventions, such

515/754

as the plan, which is strikingly similar to the baptistry

within Rusafa's greatest architectural complex surrounding basilica A. The acclamation underscores alMundhir's skillful adoption of the cultural language of
Rome. Alan Cameron's study of acclamations has
demonstrated that the phrase in question was entirely
standard, even hackneyed, in the eastern provinces
and attested epigraphically from the late fifth to the
seventh centuries.106 The name given in the genitive
indicated a favored faction, or else an individualsometimes an emperorusually in association
with a faction. The phylarch's inscription imitated the
simplicity of the well-worn acclamation formula, omitting, for example, any titles. Its attractiveness must
have lain precisely in its simplicity and familiarity. And,
of course, in its association with success.
To reconstruct the cultural context in which these acclamations were used, Alan Cameron refers to a characteristic passage from the De ceremoniisa tenthcentury document that here preserves sixth-century
prac103. Karnapp, Stadtmauer, 3749, 5253.
104. Konrad, DaM 6 (1992): 313402.

516/754

105. Alois Musil, whose interpretations always deserve serious consideration, suggested, in passing,
that the al-Mundhir building was the church originally
built to house Sergius's relics by the Surans and "rebuilt and richly endowed" by the Ghassanid phylarch
(Palmyrena, 264). This observation has escaped the
notice of subsequent students of the building, including myself in my dissertation (see above, n. 71).
106. Cameron, Porphyrius, 7475. Already at the time
of the publication of Sauvaget's article, the editor of
Byzantion, Henri Grgoire, had noted that the acclamation of al-Mundhir "est du type byzantin le plus
banal" (Sauvaget, Byzantion 14 [1939]: 117, below n.
1).

Page 160
tice.107 The following excerpt describes hippodrome
ceremonial during which the emperor might have
been present, surrounded by the sacred symbols of
his power:
And again the cheerleader cries in a loud voice: "Holy,
thrice holy"; and they all shout back: "Victory for the
Blues" . . .; the cheerleader cries: "Lady, Mother of
God"; and the people cry: "Victory for the Blues" . . .;
and the cheerleader cries: "The power of the Cross";
and the people respond: "Victory for the
Blues." . . .108
Modern students may not feel immediately at home
with this antiphonal interplay of sacred and profane
that, as Cameron observes, "is so characteristic of the
unselfconsciously religious Byzantine." But, I suggest,
it is precisely this interplay that informed al-Mundhir's
choices as an Arab ally of Rome, with pronounced
political and religious interests. Acclamations using
sacred language were inscribed in the auditorium of
the hippodrome at Alexandria, an example from the
field that complements the De ceremoniis.109 Acclamations also played a prominent role in

518/754

ecclesiastical assemblies, many of which were held in

churches and related buildings.110 The acclamation
"The fortune of the Christians triumphs" (
) has been found on a column in the
courtyard of a church at Zenobia, east of Rusafa, on
the Euphrates. Closer to home, "The faith of the
Christians triumphs" (
) is carved on
a column inside the north gate of Rusafa itself.111
Two questions that naturally arise next are: who was
the intended audience, and what was the intended
purpose of the inscription? The latter question is
closely linked with the more general issue of the
building's function and so will be left until later. Obviously, since the text is in Greek, only those who knew
that language would be able to read it. Whether it
would in fact have been legible is less apparent. Although the apse is the most symbolically potent place
in the building, the inscription's location in the innermost recess of the apse is certainly not the most
prominent position from which to proclaim the
phylarch's success.112 The acclamation is legible if
one stands in front of the apse, but not from much further away.

519/754

107. Cameron, Porphyrius, 69.

108. Const. Porph., De cer. 1.69, p. 311.614
(Reiske)
109. Borkowski, Inscriptions, 7596; see also 14446
and the companion sketch of the auditorium, which illustrates the location of the different acclamations.
110. Rouech, JRS 74 (1984): 18689.
111. For Halabiya: Lauffray, Halabiyya-Zenobia, 2:
23137, no. 6; SEG 41 (1991), no. 1531d; Rusafa:
Seyrig in Tchalenko, Villages antiques, 3: 3334, with
drawing in vol. 2, pl. 148, no. 36. Cf. Gatier, DaM 10
(1998): 241.
112. The apse measures ca. 4.35 m in width and ca.
3.35 m in depth.

Page 161
Approximately 1.20 m long, the text follows the curved
molding over the double window (fig. 13). Set in the
back of the apse, with letters standing only 12 cm high
and no sign of color, we cannot take it for granted that
many, in fact, would have read it. And what effect
would the inscription have had on any of al-Mundhir's
Arab visitors who were not literate in Greek? For
them, it is likely that the Greek letters would have expressed the phylarch's link with the Roman authority
that had invested him with special privileges among
the Arab allies. In this respect the carved characters
were a cultural marker that operated in a fashion similar to the familiar architectural style of the building itself. Even among those for whom a text is inaccessible either because of its language or location, writing
can nonetheless fulfill a nonliterary function by symbolizing authority, especially religious authority.113
That different visitors would interpret it at different
levels is only to be expected and does not necessarily
conflict with the intended purpose, or better, purposes, of the building. The phylarch may well have
behaved in a different manner according to the audience, but we must leave this issue for later, after

521/754

considering the structure's possible functions within

the Syro-Mesopotamian cultural milieu.
For a fuller picture of the range of religious, social,
and political uses to which churches in late antique
Syria were put, we must consider some parallel situations in the region. As a framework within which to
assemble these examples, it is helpful to think of the
tribal church, which was common in predominantly
Arab Hira. For example, at Hira the
(church) Bani
Mazin was named after an ancestor of the Banu Zimman, a clan of the large al-Azd tribal group, which included the Banu Mazin and the Banu Ghassan.114
The tribal church would be a structure where the tribe
would gather, both its settled members and those who
did not live permanently at Hira. It may be recalled
that in the mid sixth century, the missionary Ahudemmeh had made a point of assigning the names of clan
chiefs to the new churches he established, explicitly
"so that they would assist in every affair and matter of
business, as much as they were needed."115
The stone martyrium dedicated to the "holy martyr
Thomas" outside the city walls of Anasartha provides
another useful parallel to the al-Mundhir

522/754

113. For examples of this nonliterary function in the

early Roman Empire, see Beard in Literacy in the Roman World, 48, and id., PBSR 53 (1985): 139.
114. For Hira, see al-Baladhuri, Futuh al-buldan, p.
281 (tr. Hitti and Murgotten, 1: 442); Ibn Hazm, Jamharat ansab
37475;
al-Ghani,
al-Hira 44.
Likewise at Kufa, after the original establishment at
the settlement's center of a mosque, government
building, and open space for the market and camels,
various tribal groups such as the Banu Makhzum, a
clan of the
tribe, built separate tribal mosques. See
al-Tabari,
2.734.
115. Hist. Ahud. 4, p. 27.

Page 162

Figure 13. Rusafa: al-Mundhir building, apse. Photograph by R. Anderson. Courtesy of Dumbarton Oaks.

Page 163
building at Rusafa. Its Greek inscription states that in
425 a woman with the unusual Arab name Mavia
(Mo in the inscription) dedicated the martyrium.116 Southeast of Chalcis, Anasartha was a major
market center for the cultivated fringe of the steppe,
where pastoral and agricultural life met, and still do
today. It was the region controlled by the Tanukhids in
the fourth century. Located east of Anasartha is
Zabad, where a famous early sixth-century inscription
identified a Sergius church in Greek, Syriac, and Arabic.117 The Arab dedicant of the Anasartha martyrium used Greek alone. But the Arab association of the
martyrium is undisputed thanks to the name Mavia
(Arabic Mawiya) in the inscription. This Arab building
in honor of a Christian saint was constructed outside
the walls of a regional center on the steppe's edge,
foreshadowing the role of the al-Mundhir building at
Rusafa. Both suggest the importance of permanent
buildings as unifying points along the wide spectrum
between settled and fully pastoral Arabs.118
At Harran in the Laja, a bilingual Greek and Arabic inscription carved on a lintel identifies a martyrium dedicated to S. John by "Sharahil son of Zalim, phylarch,"

525/754

in the Greek. The first line of the Arabic reads "I,

Sharahil, son of Zalim, have built this martyrium in the
year 463."119 The year, reckoned here according to
the era of Bostra, is A.D. 568. The inscription offers a
glimpse into the phylarchal system, although it is not
certain whether Sharahil would have been associated
with the Ghassanid phylarchate, since this name is
otherwise unattested. But the prominent
116. Mouterde and Poidebard, Limes, 19495. The
building has now been destroyed. Shahd, BAFOC,
22238, and BASIC, 1: 243, argues that the Mavia in
the inscription should be identified as the famous Tanukhid queen.
117. IGLS, no. 310, with drawing. See the discussion
in chapter 4.
118. Shahd's insistence, with reference to the Tanukhids, that "they [the foederati] were not nomads
but were sedentaries who had their own permanent
establishments outside the big cities" (BAFOC, 227)
seems here to sacrifice an opportunity to appreciate
how Arab buildings served to mediate between the
very diverse population that inhabited this region. The
statement that follows is characteristic of Shahd's

526/754

thought on Arab building projects throughout his work

and deserves a brief commentary: "Far from being a
nomadizing and unsettling element, they [again, the
foederati] were the watchmen of the imperial frontier
against the inroads of the nomads from the Arabian
Peninsula, and as builders of castles, palaces,
churches, and monasteries, they were active participants in the development of settled life in the Byzantine limitrophe." First of all, pairing of "nomadizing"
with "unsettling" implies that the nomadic way of life is
unsettling. Secondly, the deliberate contrast of "nomadizing" and "unsettling," which represent mobility
and impermanence, with the static image of the
"watchman" and his permanently built environment
sweeps away all the gray area in between, which includes, e.g., ephemeral architecture and seminomads.
119. Dussaud with Macler, Nouvelles archives 10
(1902): 72627, for the Arabic; Waddington, Inscriptions, no. 2464, for the Greek.

Page 164
display of the phylarch's association with the church
again bears witness to the close, personal links of
Arab leaders to particular churches, perhaps especially martyria.
Perhaps the most interesting and persuasive evidence for the existence of tribal churches within the
Ghassanid orbit is the church referred to as the
"church of the glorious, Christ-loving and patrician
Mundhir" that appears in the "Letter of the Archimandrites," dated to ca. 570.120 Also included in this list
is the monastery of 'Isnaya, convincingly identified by
Irfan Shahd as a Syriac form of the Arabic "Ghassaniya."121 The churches and monasteries affiliated with
the Ghassanid confederation, and with specific
phylarchs, would have served to affirm their Christian
identity, as well as to provide a permanent base for
the spread and maintenance of non-Chalcedonian
Christianity in the region. In similar fashion the building outside Rusafa, with its ecclesiastical architecture
and the prominently placed inscription linking the
structure with the Ghassanid phylarch, could have
served as a tribal church.

528/754

The greatest obstacle to our understanding of the alMundhir building and its inscription is the underdeveloped nature of our knowledge of how churches
workedespecially tribal churches. What may seem
to us an impossibly blasphemous combination, such
as a personal acclamation in the apse of a church,
may have seemed laudable to some contemporaries,
tolerable to others and unforgivable to still others. Response to what we may choose to assume were abrasive actions would depend on the context, and particularly on the balance of power among those
involved.
An incident recorded in the Vita Euthymii by Cyril of
Scythopolis illustrates the confusion over boundaries
that animated relationships between secular and clerical authorities in this period. During Euthymius's twoyear absence from his monastery in Palestine, the
Arab phylarch Terebon had established the habit of
standing near the altar, just behind the chancel screen
(
) during the liturgy.122 After Euthymius's return, Terebon, in his usual manner, was resting his
hands on the chancel screen during the liturgy celebrated by Euthymius. Suddenly, during the anaphora,
the phylarch saw a veil of fire descend in front of the

529/754

altar. In terror, he retreated to the back of the church.

This incident first makes clear that
120. In Documenta 223 (tr. Chabot, 155). The location
of this church, named as
, is still uncertain. See
the discussion of 'Uqabta by Shahd, BASIC, 1:
83031, 83435, and also 86265. Nldeke's tentative suggestion that it may be identified with
,
south of Damascus, has received cautious acceptance from Shahd.
121. In Documenta 224 (tr. Chabot, 156). See the discussion in Shahd, BASIC, 1: 83338, with an important parallel from a letter of Simeon Beth Arshan.
122. Cyr. Scyth., Vita Euthym. 28.

Page 165
Terebon's adopted place was next to the altar and
that in Euthymius's absence this preference had been
accepted. Secondly, with the return of Euthymius's
authority came the ultimatum that the phylarch's behavior was not acceptable for a secular ruler. It was abruptly terminated, thanks to divine intervention.
Terebon's behavior could in this case be curtailed because of the pressure exerted by a more powerful authority than himselfEuthymius and his God.
Other evidence survives from the late sixth and early
seventh centuries to illustrate a range of activities that
took place in churches of the region.
ibn Zayd, a
Christian Arab counselor of Khusrau II was reported
to have held a banquet in a church in order to mark an
alliance with some Arab clients.123 As we have seen,
the histories of the sixth-and early seventh-century
missionaries Ahudemmeh and Maruta also refer specifically to the Syro-Mesopotamian Arabs and their social and charitable activities centered on churches and
monastic complexes.124 What we need to explore
further is how the region's leaders used the space
within a church, whether such use varied according to
local geography, climate, and culture, and whether

531/754

this would have differed from standard use in, for example, a highly urban context. In the misunderstanding between Terebon and Euthymius we see the gaps
that could easily open up in practice. The episode
may also point to varying practices from region to region, even from church to church.
An incident at Dara a decade after al-Mundhir's fall
from grace provides a helpful context for interpreting
the phylarch's building outside Rusafa. Theophylact
recounts how in early 591 Khusrau was proclaimed
king and, surrounded by an armed Roman contingent,
advanced with the Roman general Narses to Dara.
There, "the barbarian king entered the city and enshrined himself at once within the walls in the city's
celebrated shrine, in which the Romans conducted
the mysteries of religion."125 This act of Khusrau
galled the city's inhabitants, who protested that even
upon taking the city Khusrau I had not committed
such an outrage against their religious observances (
). Bishop Domitianus's threat to withdraw the
Roman alliance quickly chastened Khusrau, who sent
his highest-ranking representatives to plead with
Domitianus on his behalf. Once he had evicted

532/754

123. al-Isfahani, Kitab al-aghani 2.100.

124. See chapter 4 above.
125.

Theoph.

Sim.,

Hist.

5.3.4:

can be interpreted as either "in" or "belonging to." In his translation of Theophylact, Schreiner finds the meaning of
the sentence so opaque that he emends
, which
appears in the MSS, to
, from
, producing an
unnecessarily innovative translation: "Der barbarische
Knig trat in die Stadt ein und schritt in Waffenrstung
in die berhmte Kirche der Stadt."

Page 166
Khusrau from the shrine
Domitianus relented.126

),

The episode is a prime example of crossed cultural

signals. Over the course of two years Khusrau had
been raised from exile to the throne again, thanks to
the assistance of his traditional enemy, Rome.
Khusrau's cause had been advanced by Roman bishops and commanders; he had prayed to S. Sergius;
Roman forces rallied, and a month later he received
the head of Zadesprates; Khusrau grew impressed
with the Christian God, encouraged by close contact
with Bishop Domitianus of Melitene and Patriarch
Gregory of Antioch;127 the emperor Maurice granted
reinforcements and a loan for Khusrau to keep the Iranian forces faithful to him. In this buoyant atmosphere,
Khusrau arrived at Dara with a new Roman general
and took what to him, a king with a brilliant future,
must have seemed the most natural stepto display
his personal authority in the city's most prestigious
space, the sanctuary of the Great Church. It is even
possible that Khusrau intended the gesture as a compliment to Domitianus and Gregory, whose

534/754

enthusiasm for his interest in Christianity must have

been transparent to the exiled king.
By occupying the site most sacred to Christians,
Khusrau was allying himself with the powerful Christian God. But Khusrau was not a Christian king. That
Khusrau established himself in the sanctuary itself
and not, as Michael and Mary Whitby suggest, in "part
of the complex of ecclesiastical buildings attached to
the Great Church,"128 is certain from Theophylact's
significant repetition of
and
. Khusrau
did not raise a tempest simply by moving into the
bishop's guest rooms without an invitation. The Sasanian king was no fool, and given his dependence on
Roman support (emphasized by Theophylact), the
fiasco must have taken Khusrau by surprise just as
much as it did Domitianus and the Darans. Already
while staying at Edessa Khusrau had offended his
host, John of Rusafa, by insisting that the hostess
serve wine to their foreign guest by her own fair
handas, he attested, was the Sasanian custom.129
We can only conclude that Khusrau's self-installation
at the church was a ritual gesture of power based on
a precedent familiar to the Sasanian king at home in
Iran, but which backfired in this particular context.

535/754

In Khusrau's time, though, there still persisted a traditional function of

126. Theoph. Sim., Hist. 5.3.7.
127. On the role of Domitianus and Gregory during
Khusrau's exile, and in particular Domitianus's correspondence with Pope Gregory the Great with regard to
what the clerics believed was Khusrau's imminent
conversion, see Paret, REB 15 (1957): 4550;
Peeters, AB 65 (1947): 1117.
128. Theoph. Sim., Hist. 5.3.4 (tr. Whitby and Whitby,
135; see also n. 12 for comment).
129. Mich. Syr., Chron. 2.38081.

Page 167
the Zoroastrian fire temple that is highly relevant to
Khusrau's action at Dara. On the occasion of a Sasanian monarch's coronation, the king of kings lit a fire
in what then became his fire temple. The king's reign
was reckoned from that moment.130 And although
this is the most powerful testimony to the joining of
political rule with divine authority in Khusrau's immediate cultural environment, there are other examples
from late antique Syria and Mesopotamia where sacred spaces witnessed the long-term demonstration of
political authority.
While the examples of Terebon and Khusrau are helpful in that they illustrate the often uneasy interaction
between priests and kings in sacred space, in many
ways it is more illuminating to find evidence of the
day-to-day established use of holy places for functions other than the explicitly liturgical. It cannot be
overstressed that the transaction of judicial matters in
churches was a long-established practice in Syria and
Mesopotamia. Already in 497/98 Alexander, the governor of Edessa, established a reform by sitting in
judgment every Friday in the martyrium of S. John the
Baptist and S. Addai the Apostle.131 This is perhaps

537/754

the most important key to our understanding of how

the secular and the sacred were integrated in the alMundhir building outside Rusafa. Sasanian precedent
also existed, since judicial proceedings relating to the
temple treasury took place inside fire temples.132 As
we have seen, the authority of Sasanian kings was
also intimately linked with the fire temple: some early
Sasanian coins show the fire altar on the reverse, together with a legend which describes the fire as belonging to the king on whose coin the image appears.133 The throne-altar appears on several Sasanian and later pseudo-Sasanian silver plates, again
as a fusion of the religious and secular power of the
king. In his Chronology of the Ancient Nations, alBiruni describes the appearance and function of
thrones in fire temples. For example, the emperor
Peroz, seeking relief for his country from a severe
drought, entered a fire temple, embraced the fire, and
then "descended from the altar, left the cupola, and
sat down on the (seat) made of gold, similar to a
throne, but smaller."134
The example set in churches and fire temples may
also have had some influence on the similar use
made of mosques in early Islamic Syria and

538/754

Mesopotamia. For instance, not long after the Muslim

conquest of Meso130. Lukonin, CHIran, 3: 695; Chaumont, JA 252
(1964): 5975, esp. 6575.
131. Chronicle of Edessa of 506 29; see also Morony,
Iraq, 8385.
132. Lukonin, CHIran, 3: 734. On the relationship
between fire temples and palaces, see Schippmann,
Iranischen Feuerheiligtmer, 499515.
133. Harper, Iran 17 (1979): 5152, 63 with pl. 4b, illustrating a Sasanian throne in front of the royal fire
altar.
134. al-Biruni, Kitab al-athar al-baqiya 228 (tr.
Sachau, 215). On the altar-throne, see Harper, Iran
17 (1979): 63.

Page 168
potamia, mosques became the favored venue where
public announcements were made and where qadis
would hear lawsuits.135 Of course, this practice is
traced back to the Prophet himself, who prayed,
preached, and settled disputes in his house-mosque
at Medina. In late antique Syria and Mesopotamia, examples of the use of mosques for political and social
purposes are not difficult to find. For example, in early
Umayyad Syria, the Caliph
held audiences in
both mosque and palace, but at the mosque he cultivated a more informal atmosphere with his public.
Then he would come out and say: "Boy, bring my
chair." He would go into the mosque and humble himself [before God], after which, leaning his back against
his stall [maqsura], he sat on his chair, attended by
his guards. There would approach him the weak, the
Arabs of the desert, children and women, and whoever had no [protector], and they would say: "I have
been wronged," to which he replied: "Console him."
Or: "I have been attacked," and he replied: "Send
[men to occupy themselves] with him." Or: "Somebody swindled me," and he gave the order: "Go and
look into his affairs."136

540/754

The cumulative effect of these examples is to suggest

that to discuss the al-Mundhir building as if in late antique Syria and Mesopotamia church and audience
hall were mutually exclusive categories is simply to
miss the point. The eastward orientation, plan, and
decoration of al-Mundhir's building belong to the world
of ecclesiastical architecture. When considering the
structure, we must remember the powerful interplay
between sacred and imperial symbolism that enlivened church architecture at this period.137 One
may recall the prominence of the monograms of
Justinian and Theodora in their church dedicated to S.
Sergius in Constantinople. An instructive parallel
comes from Huwirta's Michaelion, also known as the
north church, mentioned earlier for its representation
of an eagle. The Michaelion is a three-aisled basilica
dedicated to the archangel and dated to the end of the
fifth century.138 Unlike the al-Mundhir building, where
the complete absence of a pavement creates serious
difficulties of interpretation,139 sections of the mosaic
floors of the Huwirta basilica are well preserved. At
the east end of the central aisle, just in front of the
apse, Adam is depicted in the act of naming the animals. His throne, cushion, and footstool allude to

541/754

standard Roman imperial images, for God had given

Adam royal au135. According to hadith, some qadis sat next to the
minbar (Pedersen, EI2, 6: 67071, with examples of
churches used for litigation in the Islamic period).
136.

, Muruj al-dhahab 1833.

137. Krautheimer with

antine Architecture, 262.
138. Canivet and Canivet,

, Early Christian and Byz, 1: 31517.

139. Ulbert, AAS 33 (1983): 71, reports that no trace

of the floor was discovered during excavation. On the
rarity of finding clear archeological evidence of altar
installations in apses, see Donceel-Vote, Pavements, 112 n. 38.

Page 169
thority over Creation. But at the same time the image
anticipates the throne of Christ the heavenly kingthe
altar, in other words, that stood within the apse.140
We might also compare the contemporary Chrysotriclinos built by Justin II (56578) in the Great Palace at
Constantinople, which could even have been seen by
al-Mundhir on his official visit to the capital two years
after the structure's initial construction. The decoration, executed under Tiberius II (57882) and restored
by Michael III (84267), depicts Christ enthroned in
the vault of the eastern apse, in which the emperor
would preside from his throne over a wide range of
ceremonies, many of explicitly religious character. In
Krautheimer's words, "it served as audience hall for
the Emperor's Majesty and palace chapel."141 One
example of a multilayered religious ceremony celebrated in this space took place each Palm Sunday.
After receiving high dignitaries in the Chrysotriclinos
early in the day, the emperor would return after a
series of processions outside the Palace, and take his
place on the right-hand side of the Chrysotriclinos.
The priests then entered and stood in the center,
while a deacon placed the Gospel on the royal throne
and recited the litany.142

543/754

Bearing in mind this cultural context where sacred and

secular power were commonly fused, we can also
speculate about how the building outside Rusafa's
walls would have been used. In addition to attending
services there when he was present in the area, we
can imagine that the phylarch might also have used
the space as the governor Alexander of Edessa had,
as a potent venue for settling disputes. Perhaps, like
Adam in the Michaelion, the phylarch would have sat
before the apse as God's earthly regent, the phylarch
over the Arab confederacy who would have held
meetings in this tribal church with Arab encampments
spread out in the plain around Rusafa (fig. 14).143
Seasonal rhythms would have dictated the building's
use. For the Ghassanids whose primary hirta, Jabiya,
was in the Hawran, Rusafa was a far-flung northeast
extension of their influence. But as we have seen, the
settlement was well placed to bring together, once a
year, the Ghassanid-allied Arabs whose pastures
were far from the Hawran. The building constructed
outside the walls served the Arab pastoralists who
came to Rusafa en route to winter pastures. It
provided a permanent struc-

544/754

140. My husband kindly drew my attention to this

parallel.
141. Krautheimer with
antine Architecture, 78.

, Early Christian and Byz-

142. Const. Porph., De cer. 1.32 (Reiske); 1.41

(Vogt), with commentary 1.810 (Vogt); see also Lavin, Art Bull. 44 (1962): 127, for discussion of the interplay of sacred and profane in late Roman architecture, esp. 2223 on the Chrysotriclinos.
143. Cf. Cyr. Scyth., V. Euthym. 15, where the Christianized Arabs under the leadership of AspebetusPeter camp in their tents around the church founded
for them by Euthymius at some distance from his
monastery. Sauvaget, Byzantion 14 (1939): 130 n. 1,
draws attention to the late antique placement of Tanukh encampments outside the walls of Aleppo. See
also Mich. Syr., Chron 12.8 (tr. 3: 31).

Page 170

Figure 14. The fourteenth-century Armenian monastery of the Apostle Thaddeus between Lake Van, Turkey, and Lake Urmiya, in northwest Iran, surrounded
by pilgrims' tents for the patronal festival. Courtesy of
Domas Publications.
ture where the phylarch could be present during the
festal period of a favorite saint to hear suits and meet
with other leaders within the confederation. A gathering of these leaders under the auspices of a single
saint would have served to reaffirm and strengthen
the no doubt sometimes fragile federate bonds.144

546/754

The importance of fixed, congregational centers in nomadic society is also expressed in early Arabic poetry,
as in the
of Labid b.
, a younger contemporary of the Prophet Muhammad and a convert to Islam:
They have built for us a house whose elevation is high,
and the old and the young of the tribe have ascended to it.145
The association of the al-Mundhir building with Rusafa
and S. Sergius would have served to reinforce the authority of the phylarch in the same way that a holy
man buttressed the authority of the ruling clan at a
haram.
144. For comparable federate gatherings at a temple
dedicated to a single deity in South Arabia, see, e.g.,
Doe, Monuments, 207.
145. Labid b.
84 (tr. and comm. Jones, Early
Arabic Poetry, 2: 200).

Page 171
Especially if, as seems likely, the building was near
the place of Sergius's martyrdom, and perhaps even
stood upon the very ground sanctified by it, alMundhir's own authority would have been further
sanctioned by the saint's prior presence. If the collective memory of Rusafa recalled that al-Mundhir's building rested on or near the site where according to tradition Sergius's blood had been spilt, then the various
religious and political purposes that merged at the
building would also be stamped by the saint's authority. The inscription would have served as another sign
of al-Mundhir's intimate association with the saint, this
time with the authority of the written word behind it. It
was less important that all who entered the building
could see or read his name in the inscription than that
the words were simply present in the most sacred
space of the building.
If we are to understand the al-Mundhir building as a
church, it should be seen within the broad spectrum of
church functions sketched here. Perhaps the most important precedent is the use of churches, such as that
at Edessa, for judicial proceedings. And as we have
seen, Edessa was not an isolated example, but part of

548/754

a long-standing tradition in west Asia. Inevitably conflicts of interest and interpretation arose. In the cases
of Terebon and Khusrau II, clerical authorities possessed power enough to override the political leaders'
claims within the sacred space of the church. Our written sources for Rusafa do not preserve any clerical
response to al-Mundhir's inscription in the apse, or to
his actual use of the building. Nor, as we have seen,
do the sources provide any clear impression of the
nature of either the Chalcedonian or the nonChalcedonian hierarchy at Rusafa. The record leaves
no trace of a dynamic religious leader like Euthymius
or Domitianus who would have been in a position to
challenge al-Mundhir. Certainly a non-Chalcedonian
cleric would have thought twice about opposing his
most powerful patron. And Chalcedonians too would
have been acutely aware of the critical role played by
the Ghassanids in the protection of the frontier zone,
despite al-Mundhir's tumultuous relationship with Roman authority. The Roman portrait of al-Mundhir as
headstrong and independent-minded is an exaggeration of the qualities that made him an undeniably distinguished commander and a leader capable of buttressing his own authority by close association with S.
Sergius.

549/754

al-Mundhir and his forces had been highly successful

in defending Roman territory, especially north and
east of the Hawran, from Iranian and Arab attacks. In
570, they had routed the Arabs of Qabus, Iranian allies who had provoked an encounter with the Ghassanids by invading Roman territory.146 After 575, alMundhir took the initiative in attacking and plundering
Hira and was said to have distributed the booty
among monasteries,
146. Men. Prot. fr. 17; Joh. Eph., HE 3.6.3; Mich. Syr.,
Chron. 10.8. (tr. 2: 3089).

Page 172
churches, and the needy.147 After an official visit to
Constantinople in 580, he returned to find the area he
protected exposed to the raids of the Iranians and
their Arab allies, a situation he soon put straight;148
and later that year, after an unsuccessful campaign
with Maurice, then the emperor Tiberius's magister
utriusque militiae per orientem, he routed the Iranian
and Arab forces that attacked him on his return
home.149
At least as early as May 570, al-Mundhir held the title
of phylarch, but only two years later Tiberius was plotting his death. The emperor sent a letter to Marcianus,
the magister utriusque militiae per orientem, with instructions how to ensnare the Arab leader; and he
sent another letter to al-Mundhir himself. But somehow, each letter was delivered to the other's addressee, and al-Mundhir, alerted in this way to Tiberius's
treachery against him, withdrew his protection against
raids in the frontier zone for three years. When, in
early 575, al-Mundhir agreed to a reconciliation, he
and Justinianus, Marcianus's successor, sealed their
alliance with oaths at the tomb of S. Sergius at
Rusafa.150 The commanding personality that had

551/754

caused Romans to distrust al-Mundhir almost from the

start of his career as phylarch, was perhaps also what
inspired him to link his authority so intimately with that
of Rusafa's saint. If the building outside Rusafa's
north gate was built during the first part of alMundhir's service, between 570 and 575, he may
have first met Justinianus there before entering
Rusafa together. But if, as is also possible, the inscription
was carved after 575, the
building would have been intended as a trophy of his
restoration as protector of the frontier zone and as a
symbol of thanksgiving for Sergius's aid in that
achievement. The site and function of the building
would then have represented the inalienable alliance
between the soldier saint Sergius and the phylarch alMundhir. Before the reconciliation in early 575,
Justinianus was preparing a force made up in part by
"barbarian" mercenaries for engagement with Iranian
forces, and al-Mundhir must have perceived that his
estrangement was by then only working against him.
The walled settlement of Rusafa was conveniently
near the front and enshrined the region's most
revered saint, a powerful guarantor of the two parties'
promises.

552/754

In 581, Roman imperial plotting against al-Mundhir finally bore fruit, and at the consecration of a friend's
church at Huwwarin, al-Mundhir was arrested, "like a
lion of the wilderness shut up in a cage."151 The
phylarch was taken to Constantinople, where he and
his immediate family were
147. Joh. Eph., HE 3.6.4.
148. Ibid., 4.42.
149. Ibid., 6.18.
150. Ibid., 6.4.
151. Ibid., 41. See Shahd, BASIC, 1: 45961.

Page 173
placed under house arrest. The phylarch's son
stayed on to raise havoc in the frontier zone his father
had once protected. After the accession of Maurice,
who had tarred al-Mundhir as a traitor in the reign of
Tiberius, the captive phylarch was exiled to Sicily. Not
long afterward, probably in 584,
too was arrested and exiled to the same island. The leadership vacuum left behind was soon filled by fifteen splinter
groups, each controlled by an unnamed prince. Syriac
sources refer to an Arab commander allied with Rome
again in 587: Abu Jafna ibn al-Mundhir, who was living at Rusafa.152 To judge from his name he was
probably another of al-Mundhir's sons, who remained
in Syria during the former phylarch's exile.
S. Sergius embodied qualities, such as military
prowess and healing powers, that were required for
survival in the environment where his cult flourished.
What is so striking about his cult is how tightly enmeshed it was with the place where it thrived. As we
have seen, the focal point of the relationship that had
developed between the saint and the Arab population
during the course of the fifth and sixth centuries was
the frontier shrine, where the political, economic, and

554/754

religious interests of Rome, Iran, and the Arabs converged. But it should never be overlooked, despite the
site's impressive architecture and its famous saint,
that this shrine lay fully in the steppe. It was by no
means inevitable that the Roman fortress would grow
into a celebrated city in the sixth century. Without belief in the saint's ability to benefit his wide range of devotees through the powers he embodied, the site
would have remained a modest watering point.
152. For
revolts, see Joh. Eph., HE 3.4243; for
the splinter groups, see Mich. Syr., Chron. 10.19 (tr.
2: 34951); Chron. 1234 74 (tr. 1: 166). For Jafna at
Rusafa: Chron. 1234 90 (tr. 1: 169), which preserves
the secular history of Dionysius of Tell Mahre (d. 845);
and also for Jafna, see Agapius, Kitab
, p. 182,
which refers to Jafna as a qadi, here best translated
as leader. Cf. also Yaqut,
3.156, for later Ghassanid associations with Rusafa. For one account of
the Ghassanid phylarchate after the reign of alMundhir, see Shahd, BASIC, 1: 54063, and id. in
ARAM 5 (1993): 491503.

Page 174

SIX
The Cult of S. Sergius after the Islamic
Conquest
UMAYYAD RUSAFA
ibn
, the warrior-poet who fought at Yarmuk,
wrote of the Ghassanids: "They were lords in the
jahiliya and stars in Islam."1 Certainly Ghassanids
were among the Arab allies of Rome who fought at
the protracted battle of Yarmuk in August 636. After
this signal Roman defeat, the victors collected their
booty at "Jabiya of the kings," the site of the famed
Ghassanid hirta.2 Jabiya was established as the
region's main military camp, and in 638, the caliph
met there with his principle advisers to hammer
out the terms under which the newly conquered lands
would be governed. With the Islamic conquests and
the collapse of Sasanian and Roman control in Syria
and Mesopotamia, the Syrian steppe ceased to be a
frontier zone. Some of the Ghassanids moved to
Anatolia, while others remained in Syria. Little or

556/754

nothing is heard of Rusafa during the first decades of

Islamic rule.
The Umayyads, Syria's rulers from 661 to 750, left behind a legacy characterized most of all by its synthesis of the region's traditions: political, economic, architectural, and religious.3 Ghassanid rule was obliterated, but the path carved out by the Ghassanid confederation as Arab allies of the Roman empire was often followed by the new Arab elite. Through their
formal alliance with Rome, the Ghassanid elite had
been drawn into the political
1. Quoted by Shahd in Islamic World, 329; and see
generally 32336 for the fate of the Ghassanids after
Yarmuk.
2. For an account of the battle and its background,
see Kaegi, Byzantium and the Early Islamic Conquests, esp. 11246.
3. For general discussions of Umayyad rule and its
cultural context, see Kennedy, The Prophet and the
Age of the Early Caliphates, 82123; Grabar, Formation, passim.

Page 175
establishment with its titles and rituals. Of particular
importance in the context of Umayyad Syria was the
Ghassanid integration of Graeco-Roman architectural
forms into the framework of Arab culture in Syria. As
we have seen, this is strikingly manifest at alMundhir's building outside the walls of Rusafa. Many
of the surviving and identifiably Ghassanid buildings,
such as Qasr al-Hayr al-Gharbi, continued to be used
in the Umayyad period, some certainly by the Umayyads themselves.4 Some of the fortified settlements
near the Euphrates also continued to be inhabited under the Umayyads, including Tetrapyrgium and
Rusafa.5 Rusafa was chosen by the Umayyad caliph
Hisham as the site of a country residence outside the
walls, and came to be known by Arabic writers as
Rusafat Hisham.6 During Hisham's sojourn at Rusafa,
literary Arabic flourished, and countless poets passed
through his court, reciting their poems beside the reflecting pool and awaiting praise or censure.7 To
judge by the number of poems recounted in the setting of Hisham's court in the Kitab al-aghani, the potential reward was deemed worth the risk.8 After the
Umayyad conquest of Spain in 711, the first amir of
al-Andalus,
al-Rahman I, built a palace (qasr) and

558/754

gardens northwest of Cordova and baptized it Rusafa

(Arruzafa).9
al-Rahman was the grandson of
Hisham and had spent part of his childhood at Syrian
Rusafa. The gardens of Arruzafa boasted plants from
Syria, and the palace was ranked among
alRahman's most remarkable constructions, along with
the Grand Mosque and the palace at Cordova. Like
his grandfather's residence, Arruzafa became a center
for poets and country pleasures such as hunting.10
Hisham's contribution to Syrian Rusafa included the
mosque built due north of basilica A, which
encompassed one-third of the church's monumental
courtyard (fig. 15).11 The worshipper could enter this
courtyard
4. Shahd, Byzantium and the Arabs in the Sixth Century, vol. 2 (forthcoming), will discuss the buildings attributed to the Ghassanids by Hamza al-Isfahani. See
Gaube, Ein arabischer Palast, who argues a Ghassanid origin for Khirbat al-Bayda.
5. Konrad, letter to the author, 14 October 1996. See
Konrad's forthcoming Resafa, vol. 5: Der sptrmische Limes in Syrien.

559/754

6. For the Arabic sources on Rusafa, see Honigman

in EI1, 6: 118386; Haase in EI2, 8: 63031; and
Kellner-Heinkele in Sack, Resafa, 4: 13354.
7. Hisham's palace has not been excavated. It is believed to be located in the ruins that occupy an area of
4 km2 south of the walls (Sack, Resafa, 4: 4952). Ulbert has excavated what he interprets as an Umayyad
pavilion outside the walls of Rusafa to the southeast
(Ulbert, DaM 7 [1993]: 21331).
8. See, e.g., al-Isfahani, Kitab al-aghani 3.21617;
4.41315; 5.19192.
9. For Spanish Rusafa, see Marin, EI2, 8: 63133.
10. The suburb that spread between Cordova and the
residential complex also became known as al-Rusafa
(Marin, EI2, 8: 632).
11. The mosque is dated by numismatic evidence
from the foundations to Hisham's reign (Sack, Resafa,
4: 4749).

Page 176

561/754

562/754

Figure 15. Rusafa: Plan of basilica A and mosque.

Page 177
directly from the mosque's prayer hall, through a door
in the south, qibla, wall. The result of this architectural
innovation was that the mosque shared the courtyard
and its stoa with the city's main church, which housed
the martyrium of S. Sergius. Two mihrabs, as well as
the stone minbar that later replaced the original
wooden one, are still visible on the interior qibla wall.
North of the prayer hall is another open-air courtyard,
with the formal entrance to the whole complex in its
north perimeter wall.
In order to understand the relationship between basilica A and the Umayyad mosque at Rusafa, it is
helpful to consider briefly the Dome of the Rock and
the Church of the Resurrection in Jerusalem, and the
shared temenos of S. John at Damascus. The Muslim
sanctuary in Jerusalem, ranking third in importance
after Mecca and Medina, was built by
al-Malik.12
It clearly upstaged the Church of the Resurrection, the
holiest monument of Christendom, expressing Muslim
sovereignty over the Holy City without direct, physical
impingement on the Christian shrine. The location
chosen instead for the Dome of the Rock, the Temple
Mount, was among the most symbolically potent

564/754

within the monotheist tradition. What is of interest

here, though, is not Muslim use of sacred topography,
but more specifically the form that Muslim interaction
with Christianity took. In terms of cult, the Church of
the Resurrection presented little of interest to adherents of the Muslim faith. There was, naturally enough,
no relic of the ascended Jesus there to reverence.
What was important in Jerusalem was not so much to
absorb current religious practice, but rather to construct a symbol of Muslim hegemony in the ancient
holy city of Judaism and Christianity, now adopted by
Islam. The message is powerfully conveyed by the
monument's position, but also by the lengthy monumental inscription inside the building. This elegant
compilation of
texts includes a rebuttal of the
Christian doctrine of the Trinity. But since Muslims
rather than Christians would have frequented the
building, the inscription was aimed more at cementing
Islamic faith and averting conversion from Islam to
Christianity, than at proselytizing among Christians.
Early Muslim leaders were not unaware of the spell
cast over the Arabs by Christianity, with its liturgical,
charitable, and festal traditions, often based at
churches and monasteries. In 706, on the order of al-

565/754

Walid I, a mosque finally usurped the entire site of

Damascus's main church, dedicated to John the
Baptist, which had stood there since at least the time
of Theodosius I.13 From 635 until 706, however,
Christians and Muslims had
12. On the Dome of the Rock, see Grabar, Shape of
the Holy, esp. 21134; id, Formation, 4771; and
Raby and Johns, ed., Bayt al-Maqdis.
13. For the architectural history of the pre-Christian
temenos and the buildings it enshrined, see Sack,
Damaskus, 1619; see also Freyberger, DaM 4
(1989): 8284 on the reuse of pre-Christian architectural fragments in the church complex.

Page 178
worshipped in separate areas within the same
temenos. The church housed as its most cherished
relic the head of John the Baptist, revered by Christian and Muslim alike as one of the greatest prophets.
Even though we can assume that the church would
still have been a great attraction in its own right in the
early eighth century, it should not be forgotten that it
was located in the heart of the political capital of the
Umayyad empire. This fact must have militated
against the sharing of the city's most important shrine
by Christians and Muslims. During the period of symbiosis, the visual balance within the temenos would
have been tipped toward the preexisting Christian basilica, renowned for its grandeur. The mosque seems
to have taken the form of an open-air arrangement,
and occupied the southeast corner of the temenos.14
What we see at both Jerusalem and Damascus is the
upstaging of a Christian holy place. In the case of
Damascus, the price was total absorption of what had
once been a shared precinct. The tenth-century historian Muqaddasi records an interchange about
Walid's decision to build the Great Mosque that recaptures the late antique cultural climate in which the

567/754

belief that architecture had the power to hold hearts

and minds flourished:
"O my uncle, verily it was not well of the Caliph alWalid to expend so much of the wealth of the Muslims
on the mosque at Damascus. Had he expended the
same on making roads, or for caravanserais, or in the
restoration of the frontier fortresses, it would have
been more fitting and more excellent of him." But my
uncle said to me in answer, "O my little son, thou hast
no understanding! Verily al-Walid was right, and he
was prompted to a worthy work. For he beheld Syria
to be a country that had long been occupied by the
Christians, and he noted herein the beautiful churches
still belonging to them, so enchantingly fair, and so
renowned for their splendor, even as are the al-Qumama15 and the churches of Lydda and Edessa. So he
sought to build for the Muslims a mosque that should
prevent their regarding these, and that should be
unique and a wonder to the world. And in like manner
is it not evident how the Caliph
al-Malik, noting the
greatness of the dome of the al-Qumama and its magnificence, was moved lest it should dazzle the minds
of the Muslims, and hence erected above the Rock
the Dome which now is seen there?"16

568/754

The situation at Rusafa offers another possible form

of cohabitation, not far from the original arrangement
in Damascus. By incorporating the three-aisled
mosque into the preexisting north courtyard of basilica
A, Hisham's architects linked their building to the
shrine of S. Sergius, but without at14. See Creswell, Early Muslim Architecture, 1:
15196, esp. 18096.
15. Qumama (which means garbage) is a derogatory
name for al-Qiyama, the Church of the Resurrection
at Jerusalem.
16. Muqaddasi, Ahsan al-taqasim 159 (tr. Le Strange,
Palestine under the Moslems, 11718).

Page 179
tempting to replace the Christian cult and its material
apparatus. Judging from the unadorned structure that
remains, he did not attempt to rival the dazzling
beauty of basilica A. Despite its important role as a
pilgrimage center, Rusafa was not a leading political
center in Umayyad Syria. Even though it served as a
refuge for Hisham, Rusafa never carried Damascus's
political weight, so it did not require the same show of
Muslim hegemony, which was what imposed the destruction of the Christian basilica of S. John and the
erection on its site of a monument to Umayyad rule.
Acknowledgment of the martyr's importance is implied
in the mosque's very location. Its proximity to basilica
A suggests an effort to benefit from the saint's
miracle-working presence and to provide Muslims with
a place nearby to worshipeven to participate in the
cult of Sergius.17 The close symbiosis of church and
mosque might also be interpreted as a Muslim attempt to ease Christians into Islam by not rejecting
outright the patterns of Arab involvement in Christian
establishments, especially in the steppe, where, as
we have seen, churches and monasteries were important fixed points in pastoral Arab life. It would not

570/754

be surprising if the mosque were known as the masjid

of Mar Sarjis, since dedications of mosques to local
saints were not uncommon even in the early years of
Islam.18 The Masjid Nabi Allah Jarjis mosque at Tarim in the Wadi Hadramawt of South Arabia is also
popularly known still today as the holy place of the
prophet Sarjis.19 The appearance of both Jarjis and
Sarjis in variants of the mosque's appellation is not
surprising. It represents another instance of the frequent confusion of these two popular military saints,
both of whom were also blurred with the Muslim Khidr.
By the time of the Islamic conquest of Syria in 636, in
the minds of many Arabs, Sergius had become so
closely identified with the Christian religion that he
stood as a symbol of the faith itself. This tendency
was strong also among those Arabs who adopted the
new religion of Islam. The early
17. For modern examples of shared cult between
Christians and Muslims, see Curtiss, Ursemitische
Religion, passim, esp. 1035 and 283, on the cult of
S. Thecla and S. Sergius; Hasluck, Christianity and
Islam, passim, esp. 6397; and Bowman, Man 28
(1993): 43160. In Urmiya in northwest Iran, both

571/754

Christians and Muslims sought healing by incubation

and anointment at the church of SS. Sergius and Bacchus (Fiey, Mossoul chrtienne, 14647, n. 2).
18. See, e.g., Palmer, Monk and Mason, 23.
19. Serjeant, Studies in Arabian History II.57475,
concludes that the dedication was to S. Sergius, popular among the Arabs of Syriaalthough Serjeant
strangely confuses Rusafa with Babisqa. His comments are extremely brief and include no attempt to
explain the obvious connection between the Arabic
Jarjis and Greek Georgios, or the relationship
between S. George and S. Sergius. In July 1997, a
banner hung outside the mosque clearly identified it
as the Masjid Sarjis. Shahd in DOP 33 (1979):
8587, favored S. George as the dedicatee, but has
more recently considered the identification with S.
Sergius more likelyin fact, he simply calls the building the Masjid Sarjis: Shahd, BASIC, 1: 956 n. 35. On
the frequent conflation of the two saints, see, e.g.,
Hasluck, Christianity and Islam, 57071.

Page 180
Islamic poet Jarir accused one of the remaining Christian tribes in Syria of having been deluded by the
Cross and by S. Sergius.20 This tribe, the Taghlib,
was famous for carrying into battle "a cross held high
and Mar Sarjis."21 Most commentators have taken
this to refer to either an icon of S. Sergius or a banner
bearing the martyr's image. Both suggestions are
plausible. Another possibility might be a relic of the
warrior saintrecalling the famed eastern king who
led his forces with Sergius's thumb attached to his
own.22 This close association of Sergius and the
Cross anticipates and helps explain a telegraphic
story about Christianity's adoption by a Roman king
that was recorded by the historian Ibn Ishaq in the
eighth century. This king, upon hearing about the miracles of a man among the Israelites, and his death at
the hands of the Jews, summons the holy man's
apostles to preach their master's message to him.
Embracing the religion of Jesus, the king then "took
down Sergius and concealed him."23 Then he proceeded to preserve and revere the cross on which
Sergius was crucified. A key to un-raveling this chronologically confused passage is that some Muslim traditions provide a double for Jesus, in order to avoid

573/754

the (to them) unacceptable view that he was crucified.

That Sergius was chosen to stand in for Christ in this
instance further highlights the saint's prominence in
Muslim understanding of Christianity. Sergius's inclusion as one of the essential features that even a
boiled-down version of early Christianity must retain is
perfectly consistent with the Taghlib talismans of Sergius and the Cross, and with Jarir's jibe. The known
dedications at Rusafato S. Sergius and to the Holy
Crossunderscore once again the fortress-shrine's
prominent position in the cultural and religious developments of the late antique world.
The Cross was widely perceived by Muslim Arabs as
a symbol not only of Christianity but of Rome,24 and
Sergius seems to have shared this distinction.
Umayyad reaction to the Cross was not uniform, ranging from outright proscription to the adoption of more
subtle alterations, as seen in the coins
20. Jarir and Farazdaq,
2. 904, ll. 8590. For additional brief references to this Christian Arab weakness for S. Sergius, see Trimingham, Christianity 236
n. 60.
21. al-Akhtal, Diwan 309 (tr. Lammens, Syrie, 73).

574/754

22. For banners associated with South Arabian saints,

see Serjeant, Studies in Arabian History III.5354.
The Christian Arab history of northern Syria is remembered with pride by modern Christian Arabs: both
the schools operated by the Syrian Orthodox archdiocese of Aleppo are named after the Banu Taghlib.
For the relic-bearing king, see Greg. Tur., HF 7.31,
and above, p. 129.
23. Ibn Ishaq, preserved in Tabari,
1.739 (tr. Perlmann 4.124, who mistakenly makes "Jesus" rather
than "Sergius" the subject of the passive verb suliba).
Cf. Newby, Making of the Last Prophet 21011, translating this passage from the
, and 2059, discussing and translating other passages from Tabari's Tafsir that refer to "Sergius" as Jesus's double.
24. King, BSOAS 48 (1985): 269

Page 181
of
al-Malik.25 Treatment of Sergius also varied,
and the case of his shrine at Rusafa reveals a sequence of responses over time. What is certain is that
Hisham confronted head on this issue of the saint's
conspicuous cult establishment at Rusafa. In the
choice of the mosque's location, proximity to the
shrine clearly overrode more practical factors, such as
the fact that the ground under the mosque was riddled
with dolines. At the time of the mosque's construction,
subsidence had already damaged the nearby basilica
A.26 The three-aisled mosque absorbed one-third of
the north courtyard attached to basilica A, so that the
courtyard was thereafter shared by those who visited
the church and the mosque. The entrance to the
courtyard, which led directly through the qibla wall, is
a highly unusual feature in Umayyad architecture. The
arrangement appears, for example, in Damascus and
Kufa, but only when connecting a mosque to a
palace.27 At Rusafa, there is no sign that the qibla
door was reserved for the caliph, imam, or any other
authority entering from a palace. On the contrary the
juxtaposition of the public courtyard and the prayer
hall at Rusafa suggests that the qibla door was designed to facilitate public movement between the

576/754

mosque and the shrine at the southeast corner of the

courtyard.
In order to explain the impractical location chosen for
the mosque and its architectural links with basilica A,
the site's excavator, Dorothe Sack, has suggested
that Hisham must have been personally devoted to S.
Sergius.28 In saying this, Sack follows Yaqut, who
claims that Hisham built his residence at Rusafa in order to be near its monastery.29 Certainly, the Umayyads were famous for the catholicity of their religious
tastes. But the choice to tie the mosque architecturally
to the famous church sprang from more than the personal fancy of Hisham's. The intimate association of
the mosque with the shrine should be seen in the context of the wider political arena in which the cult of S.
Sergius operated in late antique Syria and Mesopotamia. Although the frontier between Rome and Iran
had dissolved in the wake of the Islamic conquests,
the new rulers of Syria and Mesopotamia still needed
to exert control over the populations who inhabited the
region, the mobile Arabs in particular.30 As a place
for worship and arbitration, a space in which the identity of the community was affirmed, the mosque at

577/754

Rusafa would have been a powerful site for mediation, thanks to the divine au25. For a recent discussion of the coins, see Blair, in
Bayt al-Maqdis, 1: 6468.
26. Sack, Resafa, 4: 4142.
27. Ibid., 5354, with n. 192, and 69, with n. 239; also,
Hamilton, Khirbat, 1067.
28. Sack, Resafa, 4: 156.
29. For Yaqut, see below.
30. For the political situation, especially with regard to
the Arabs of Iraq, see, e.g., Kennedy, Prophet,
82123.

Page 182
thority of S. Sergius. At Rusafa, Hisham would have
stepped into the shoes of Anastasius, Justinian, Khusrau, and, perhaps in particular, the Ghassanid
phylarch al-Mundhir, who had all allied themselves in
various ways with the mediating influence of S. Sergius among the region's population. Hisham would have
hoped to benefit from the flow of pilgrims to the neighboring shrine of the miracle-working saint, but the
Christian cult would simply have been too well established and influential beyond the bounds of Rusafa to
have been clothed fully in Muslim garb.
At a much later phase, possibly after the earthquake
in the twelfth century, a tomb was built in the courtyard north of the mosque.31 Sack has suggested that
it was a Muslim tomb, or mashhad, of S. Sergius.32 It
seems unlikely that the Muslim tomb had replaced the
Christian holy site, since Christians were still dedicating offerings at the shrine of S. Sergius at Rusafa in
the thirteenth century. If the tomb in the mosque's
courtyard was also considered a tomb of S. Sergius, it
might signify a separation of Christian and Muslim
cult, of which there are otherwise no such definite
signs, either literary or archaeological. However,

579/754

another interpretation can be offered, based on parallel evidence of cultic continuity in northern Syria and
on a twelfth-century literary tradition about Rusafa. alHawari (d. 1215), an ascetic of
leanings who
composed a guide to pilgrimage sites, includes in his
entry for Rusafa "tombs of the companions and followers" of the Prophet. He adds that he has not seen the
names himself, and that God alone knows the truth.33
Yaqut, in the early thirteenth century, reports that a
mashhad had been built in the place of an abandoned
Christian monastery outside Aleppo, in honor of ,
who had been seen there by several
.34 What
these two examples may point toward is Rusafa's
place in the wider Islamicization of holy sites in
twelfth-and thirteenth-century Syria. At Rusafa,
Muslim accommodation of S. Sergius represented the
first phase that is clearly reflected in the mosque's intimate proximity to the Sergius church. The second
phase that we may postulate here moves beyond the
shared holy man Sergius / Sarjis to the installation at
the holy site of purely Muslim holy men. The degree to
which the Prophet's companions would have cohabited comfortably with S. Sergius must have varied
over time, in accordance with historical circumstance
and religious function.35

580/754

31. Evidence also exists of a small chapel built to the

south of basilica A at an unspecified, but late, date
(Ulbert, Resafa, 2; 152; id. in Syrien, 124).
32. Sack, Resafa, 4: 4546.
33. al-Hawari, Kitab al-isharat, p. 61 (tr. SourdelThoumine, p. 136).
34. Yaqut,
al-buldan 2.691 (tr. Le Strange,
Palestine under the Moslems, 43031).
35. Different views of the Prophet Sarjis at the
mosque of Nabi Allah Sarjis at Tarim in South Arabia
offer an illustration of how understanding of a holy
man can be influenced by contemporary political and
religious developments. In the early 1950s, Serjeant
reported that "the prophet Sarjis is generally reckoned
in Tarim to be
" (Serjeant, Studies in

Page 183
We should be careful to observe the real boundary
that did stand between Christianity and Islam, despite
what appears to have been a willingness, especially in
the early Islamic period, to recognize common
ground. Muslims were discouragedon pain of
deathfrom conversion to Christianity. To underscore
the reality of this boundary, Hisham created a new
Rusafan martyr in 737, when, outside the city's walls,
he executed an Armenian prince named Vahan who
had converted to Islam and then reverted to Christianity.36 The martyr Vahan is remembered by Armenian
Christians and not usually by students of Rusafa, but
it is our only explicit literary evidence pertaining to
Hisham's relations with a Christian at the site. Although Vahan did not reside at Rusafa and was martyred there simply because he was brought to appear
before Hisham, his fate would have been a lesson to
all potential converts to Christianity.

RUSAFA AFTER HISHAM

First the obliteration of the Syro-Mesopotamian arena
of Roman and Iranian confrontation, and then the
shift of the caliphal capital from Syria to Iraq,

582/754

meant increasing political irrelevance for Rusafa, and

consequently a lack of interest in it by contemporary
historians. Rusafa's peripheral location between two
empires had boosted the circulation and prestige of its
saint on both sides of the frontier. But once it was absorbed into the Islamic commonwealth, Rusafa seems
to have been demoted to the status of a minor regional center that still bore signs of its luminous past.
From the time of the conquest onward, evidence for
the survival of the Sergius cult at Rusafa must be inferred from the continuing prosperity of the fortified
town itself. Around 1154, al-Idrisi described the settlement as the goal of travelers and merchants; and attractive place, vibrant with bazaars, merchandise,
craftsmen, and its own flourishing inhabitants.37
Three Arabic inscriptions in Hebrew script discovered
in a much-damaged khan at the junction of the town's
main east-west artery and the street leading to the
north gate suggest that some of the merchants doing
business at Rusafa were Jewish. The oldest commemorates the establishment of a room in the southwest corner of the courtyard as cult place in 1105.38
Ibn Butlan commented half a century later that most of
the inhabitants were Christian and most were

583/754

Arabian History III.574). Some forty-five years later, in

a climate of strong Wahhabi influence in Hadramawt,
the bearded, middle-aged man standing watch outside the mosque assured me that Sarjis is not in the
and should be regarded with suspicion.
36. Peeters, AB 57 (1939): 33033.
37. al-Idrisi, Nuzha 649 (tr. Kellner-Heinkele in Sack,
Resafa, 4: 14445).
38. Caquot, Syria 32 (1955): 7074. On the khan, see
Karnapp, AA 1978: 13650.

Page 184
occupied with the caravan trade.39 According to alIdrisi, Rusafa still occupied its place on the important
route between the Euphrates and Homs that skirted
the steppe to the north.40
Ibn Shaddad, a native of Aleppo who died in Cairo as
a refugee after the Mongol invasion of Syria in 1259,
records that the Mongols spared Sergius's city. Until
1269, a Mamluk governor was stationed at Rusafa,
now on the front line in the confrontation between the
Mamluks and the Mongols. Perhaps in connection
with this tension, and perhaps also because of related
damage to the settlement, which shows signs of fire
damage in its last phase, the inhabitants soon abandoned Rusafa for towns further to the west, such as
Salamiya and Hama.41 The city would not be reinhabited, aside from a goat-headed snake that haunted
the ruins thereafter and terrified passing beduin.42
Rusafa's buildings, rebuilt manifold times over the
centuries, had suffered from fire, earthquake, and
subsidence. Basilica B, damaged most likely by an
earthquake or subsidence, was allowed to be quarried
as early as the late sixth century for materials used in
basilica A and, later, in the Umayyad mosque.

585/754

Eventually, houses were built within its ruins. The

main church, basilica A, was frequently restored and
reworked to accommodate changing needs. An inscription set into its western wall records restoration in
1093 by Symeon, metropolitan of Sergiopolis, perhaps after the earthquake of 1068 that devastated
Palmyra.43 Ibn Butlan, in a letter written in 1051, says
that inside Rusafa's walls there was "a mighty church,
the exterior of which is ornamented with gold mosaics,
begun by the order of Constantine, the son of
Helena."44 No traces of such decoration have come
to light, but the statement captures the legendary allure that Rusafa's religious architecture still held in the
eleventh century. It seems that the southern or southwestern part of the basilica A complex housed monks,
although neither archeological nor literary evidence
helps pin down the chronology of such a community.45 Yaqut in 1225 describes "a convent in the
city of Rusafa Hisham ibn
al-Malik on the western
bank of the Euphrates and in the desert, a day's
march from Raqqa for those who are laden . . . I myself have seen this monastery, and it is one of the
wonders of the world as regards its beauty and its
structure. I have heard that Hisham built his city to be
near

586/754

39. Ibn Butlan in Yaqut 2.955 (tr. Le Strange,

Palestine under the Moslems, 522; cf. tr. KellnerHeinkele in Sack, Resafa, 4: 14647).
40. al-Idrisi, Nuzha 649 (tr. Kellner-Heinkele in Sack,
Resafa, 4: 145).
41. Ibn Shaddad,
al-Hatira 394 (tr. Edd-Terrasse,
2122; cf. tr. Kellner-Heinkele in Sack, Resafa, 4:
151). See also Ulbert, Resafa, 2: 15153; and Sack,
Resafa, 4: 132.
42. Musil, Manners and Customs, 414.
43. Gatier in Ulbert, Resafa, 2: 169, no. 19; see also
Musil, Palmyrena, 271.
44. Ibn Butlan in Yaqut,
al-buldan 3.48 (tr. Le
Strange, Palestine under the Moslems, 522; cf. tr.
Kellner-Heinkele in Sack, Resafa, 4: 14647).
45. For evidence of a monastery at Rusafa, see p.
80n.97, above.

Page 185
this monastery, and that it existed before the city.
There are monks in it and religious men. It stands in
the middle of the town of Rusafa."46

BETWEEN EAST AND WEST

Some glimmer of the sanctuary's beauty shines
through the five silver objects discovered in 1982 during the excavation of the western end of the northern
courtyard. The impressive assemblage includes a
hanging lamp, a eucharistic chalice, the foot of another chalice, a paten, and a nonliturgical drinking cup.
The objects were buried in a clay jar just before the
Mongol onslaught of 1259/60, and represent the only
finds in precious metal, other than a few coins, that
have been uncovered and recorded as part of systematic investigation at this once famously wealthy site.47
All five can be dated to the thirteenth century and all
attest to the enduring cultural diversity of those who
revered Rusafa's martyr. Apart from their exquisite
decoration, the objects are particularly fascinating for
the phases of use and influence they reveal. The
lamp, wrought in silver and adorned with gold and intricate niello friezes, was designed originally as a

588/754

drinking vessel about 1200 A.D. The style suggests a

Syrian origin. Sometime before 1258/59, the cup was
fitted out as a hanging lamp and presumably intended
for use at Rusafa in honor of S. Sergius. The paten introduces another type of reuse, or rather rededication,
most likely by a secondary owner of the object. A
central medallion depicts the hand of God. The
paten's form and iconography indicate a western origin, either in northern Germany, northern France, or
England, where depiction of the hand of God was a
common feature of patens dating from the early thirteenth century. After it had been transported from its
native land, a man of Edessene descent donated the
paten to "the church of Mar Sargis of Rusafa," according to the Syriac inscription.48 The nonliturgical drinking cup too originated in the west, and in this case the
central heraldic figure identifies it with the noble
Coucy family of Picardy.49 Most likely, the vessel
would have been part of the baggage of Raoul I, sire
de Coucy (d. 1191), who had come to Syria with the
Third Crusade. This connection helps more generally
to explain the proliferation of western-produced objects in Syria at this time. The elegant silver cup
shows signs of numerous repairs and a damaged secondary Arabic in-

589/754

46. Yaqut,
al-buldan 2.510 (tr. after Le Strange,
Palestine under the Moslems, 432; cf. tr. KellnerHeinkele in Sack, Resafa, 4: 146).
47. For their publication, see Ulbert, Resafa, 3, esp.
164, on the objects; and Degen in ibid., 3: 6576, on
their inscriptions.
48. Degen in Ulbert, Resafa, 3: 6874.
49. See de Pinoteau in Ulbert, Resafa, 3: 7778.

Page 186

Figure 16. Silver chalice found at Rusafa. National Archaeological Museum, Damascus.
scription explains that it was dedicated at a church,
probably at Qalat
on the Euphrates. How it
reached Rusafa is unknown.50

591/754

While the single chalice foot seems also to come from

France or England, a subtler fusion of East and West
is represented by the one intact chalice (fig. 16). The
half-round shape of the cup and the large ribbed
nodus that links the cup and foot are characteristic of
late Romanesque chalices in Germany, France, and
England. But the delicate niello ornament and figural
representations are clearly native to northern Syria,
suggesting that the chalice was made in Syria and its
form inspired by a western model.51
50. Degen in Ulbert, Resafa, 3: 7476.
51. Ulbert, Resafa, vol. 3, and Degen in ibid., 3:
6768.

Page 187
The estrangela dedicatory inscription of the priest
Iwannis (John) is original, in contrast to that inscribed
on the western paten. The representation of the Theotokos with Christ child in the tondo, and of Christ the
Pantokrator on the external border of the cup, belong
to the Byzantine iconographical tradition, and are
even accompanied by the typical Greek abbreviations
in Greek characters. Although the subject requires further investigation, the combination of standard Byzantine religious iconography and a dedicatory inscription in Syriac by a priest named John suggests that
perhaps the chalice was commissioned by a
Chalcedonian whose liturgical language was Syriac.
While, as we have seen, no depictions of S. Sergius
have been preserved on the walls of Rusafa's
churches, frescoes of the saint dating to the twelfth
and eighteenth centuries adorn three other Syrian
churches. The monastery of Mar Musa al-Habashi,
near Nabak, the church of Mar Sarjis at Qara and the
church of Mar Sarjis at Sadad are all located west of
Rusafa, roughly between Damascus and Homs.52 As
in the case of the thirteenth-century silver paten found
at Rusafa, the Latin influence is evident in the twelfth-

593/754

century paintings and is in fact more pronounced than

any Byzantine debt. At both Mar Musa and Qara, Sergius is mounted on a white horse. The rider saint
wears typical Crusader dress and turns his torso to
face the viewer. He carries in one hand the characteristic Crusader pennant, a red cross on white ground,
and in the other he holds a martyr's cross. A third portrait of this type is the famous Crusader icon of S. Sergius at Sinai.53 Western influence in this medieval
portrait of Sergius as a rider saint is undeniable.
However, it was not the only source for the rider portrait of Sergius, which, as we have seen, has Syrian
roots planted in the early phases of the cult's development.54
At Sadad, in the church dedicated to SS. Sergius and
Bacchus, the exuberant wall-paintings seem to reflect
an idiosyncratic style more than a particular iconographic tradition. The saints' portraits occupy the
south wall next to the sanctuary and are the only unlabeled depictions of rider saints in the church's highly
animated iconography. Their prominent position and
the building's dedication would have made their identity obvious. Sergius is shown crowned, but with a
much more elaborate headpiece than he wears at Mar

594/754

Musa and Qara. And at Sadad, the young martyr has

acquired a beard. Also exceptionally for Sergius, but
within the wider tradition of soldier-saints, he spears a
demon who struggles beneath the forefeet of the
martyr's horse. His accomplice's mount also rears
above the pros52. Cruikshank Dodd, Arte Medievale 1 (1992):
61130; Velmans, CArch 42 (1994): 13438; Johann
Georg, Herzog zu Sachsen, OC 2 (1927): 23342.
53. Hunt, BMGS 15 (1991): 96145.
54. See above, pp. 3544.

Page 188
trate figure. This smaller rider, no doubt Bacchus, is
also crowned. Sergius carries a diminutive passenger
seated just behind him on the horse. Given the eccentricity of these late paintings, such deviation from
the standard iconography of SS. Sergius and Bacchus
is not terribly surprising. The Sadad painter's Sergius
is a collage of attributes characteristic of several
powerful eastern saintsthe riders Sergius and Bacchus; the beloved George, spearing the dragon and
carrying a tiny figure behind him; and the other popular Sergius, whose son Martyrius accompanies him on
his steed, much in the manner of George's companion. Sergius and Martyrius may be lesser known
today, but they were sufficiently esteemed in the
eighth century for the poet Jarir to reprove the Christian Taghlib for embracing "Sergius and his son" instead of Muhammada barb that suggests that the
Sadad painter was not alone in confusing the two
powerful Sergii.55
In the wall just in front of the head of Sergius's horse a
stone reliquary still containing bones was discovered.56 That the relics came from Rusafa cannot
be excluded, since the cult spread earliest precisely in

596/754

this southwestern direction from Rusafa. It may be recalled in this context that the conclusion of the Passio
notes that miracle-working relics of S. Sergius could
be found elsewhere in addition to Rusafa.57 But there
is also the question of what happened to the relics
after Rusafa's abandonment. Some of Rusafa's inhabitants were said to have fled westward and taken
refuge in the Hama region and at Salamiya.58 No explicit mention is made of who was responsible for the
relics. Miscellaneous references to the relics of SS.
Sergius and
55. Jarir, Diwan 474. A few of the Sadad wall paintings (but not SS. Sergius and Bacchus) are illustrated
by Johan Georg zu Sachsen, OC 2 (1927), pls. 34.
See Littmann, OC 34 (1930): 28891, and Littmann
in PAES 4.B.5662, for the Garshuni (Arabic written
with Syriac letters) legends on the wall paintings. For
the iconography of SS. Sergius and Martyrius, see
Mirzoyan, REArm 20 (198687), 44155, esp. fig. 1. I
would like to thank Mat Immerzeel, who prefers to
identify the painting as S. George with Amiras, for his
correspondence about this painting.

597/754

56. In May 1996, the Syrian Orthodox priest of Sadad

showed me a small stone sarcophagus-shaped
reliquary and a silver reliquary containing small pieces
of bone. He claimed that both had been found in the
wall next to the painting of S. Sergius. The hole in the
wall is clearly visible. The recently published history of
this Christian village does not mention the reliquary,
but it describes the paintings in detail (
, Sadad fi
, 29396). On a similar stone reliquary now in the
Aleppo Archeological Museum, the name Mar Sargis
is scratched lightly on the lid. This Syriac inscription is
considered to be secondary; cf. Hunter, Studia Patristica 25 (1993): 30910.
57. Pass. gr. 30.
58. Ibn Shaddad,
2122 (tr. Edd-Terrasse). Jabbur, Bedouins, 46566, quotes Ibn Butlan on the
Christian makeup of Rusafa as late as the eleventh
century and considers it possible that the Slayb tribe
may be descended from Christians of Rusafa and the
Euphrates region.

Page 189
Bacchus are nonetheless to be found: a Georgian
translation of the original Greek Life of John, abbot of
the S. Sergius monastery in Constantinople in the
later ninth century, claims, for example, that John discovered the relics of S. Sergius in the S. Sergius
church in Constantinople. According to this account,
the relics had been hidden for many years in the
ground by heedless men.59 In the eleventh century, a
Russian pilgrim to the church of S. Sergius in Constantinople, Stephen of Novgorod, claims to have reverenced the heads of Sergius and Bacchus there; and
the martyrs' chlamydes were reportedly housed there
also since before the thirteenth century.60 At the beginning of the fourteenth century, relics of SS. Sergius
and Bacchus were kept in the Nestorian monastery at
Maragah just west of Lake Urmiya, in northwest
Iran.61 In 1377, Armenian raiders are said to have
scattered bones of S. Sergius at the Sergius monastery of Sergisiyah, in the region of Melitine.62 Today
the monasteries of Vatopedi and Simonopetra on
Mount Athos both preserve parts of the skull of S.
Sergius.63

599/754

BETWEEN CHRISTIANITY AND ISLAM.

Besides Hisham's activities at Rusafa, there is some
modern evidence that Muslims too embraced Sergius
as their saint. Charles Doughty, the Victorian traveler
in Syria and Arabia, observes of Sergius that he "was
an old Christian saint at Damascus, 'but he is now of
Islam'; you may see his shrine in the sk by the street
fountain, the bars of his windows are all behanged
with votive rags." And he relates a story he had heard
about S. Sergius in connection with the Muslim massacre of Christians at Damascus in 1860. "Upon a
morrow, in the beginning of the rebellion, Sergius his
lamps were found full of gore, also his fountain ran
blood, prodigies which great learned turbans interpreted to presage 'great destruction of Christian
blood!'"64 Despite his adoption into Muslim cultic life,
Sergius still retainedas the shaykhs' interpretations
revealhis Christian heritage. These mixed associations allowed the saint to find a place simultaneously
in both traditions.

600/754

Today, perhaps the best-known Syrian church dedicated to S. Sergius is located at

, not far from
Sadad, in the Anti-Lebanon mountains
59. See van Esbroeck, OC 80 (1996): 163, 15.
60. Majeska, Russian Travellers, 3839; cf. 16565,
for the fourteenth century; see also 26465, with n. 5,
for the chlamydes.
61. Monks of

24546 (tr. Budge).

62. Mich. Syr., Chron. 15.1 (tr. Chabot, 3: 162).

63. In contemporary Greece, S. Sergius's healing
powers are sought especially for head injuries, inspired, it is said, by the gash in the cranium kept at
the monastery of Simonopetra.
64. Doughty, Travels 1.522. For the Sergius shrine in
Damascus, see p. 105n.22, above.

Page 190
northwest of Damascus.65
today is famous for
the dialect of Aramaic that is still spoken by its inhabitants, both Christian and Muslim. Two monasteries,
one dedicated to S. Sergius and the other to S.
Thecla, perch in the mountains above the flat-roofed
houses, among which there is also a church dedicated to S. George. The church of the Sergius monastery houses an exquisite icon of SS. Sergius and Bacchus on horseback by Michael of Crete, dated to
1813. Another by the same hand depicts S. Thecla
enthroned.66 The monastery of S. Thecla is built
around a cave revered according to local tradition as
Thecla's tomb.
trio of martyrs, Sergius, Thecla,
and George, are among the select few who were
reckoned by John Moschus as the greatest saints of
the East, a distinction that they still claim not only
among Christians but also among Muslims.67 Pilgrims of both faiths visit the shrine of Thecla for healing.
is today the best-known example in Syria of
the overlapping of Christian and Muslim practice.
In the East, both S. George and S. Sergius have often
been identified with the Muslim prophet Khidr, thanks
to the healing and protective powers associated with

602/754

both the Christian and the Muslim figures. It has been

thought that this identification was encouraged by
Muslims in order to induce a gradual assimilation of
the two faiths, which would eventually lead to the conversion of Christians to Islam.68 More likely in most
cases, it simply underlines the shared hopes and aspirations for supernatural healing and protection of
both groups. S. Sergius is especially identified with
Khidr in areas with strong Armenian influence, since
Sarkis is one of the most highly revered saints in Armenian Christianity.69 There can be no doubt that the
appeal of S. Sergius to both Christian and Muslim had
roots at Rusafa, where the church and mosque stood
side by side for six centuries.
The cult of S. Sergius was established in late antiquity
around the relics of a martyr whose power as a healer
and defender was sought by the Romans, Iranians,
and Arabs who shared an interest in Syria and
Mesopo65. The monastery of S. Sergius preserves reused architectural elements that date back as early as the
fourth century, although the present church belongs to
the Middle Ages (Ulbert in Pietas, 55962).

603/754

66. On Michael of Crete and the large number if icons

by him at this monastery, see Candea and Agmian,
eds., Icones Melkites, 196202, and Hatzidakis and
Drakopoulou,
, 2: 3012, with figs. 202, 204.
67. At
, the feasts of SS. Sergius and Thecla are
both celebrated on 7 October (Reich, tudes sur les
villages aramens, 146).
68. See Hasluck, Christianity and Islam, 145, 335 n.
1; and Canaan, Mohammedan Saints, 1415.
69. On S. Sergius in Armenian culture, see Hasluck,
Christianity and Islam, 57071; Peeters, Recherches,
1: 2536; and Thierry, Monuments Armniens 508.
Further research needs to be done on the Armenian
life of S. Sergius and his son Martyrius (Peeters, ed.,
Bibliotheca hagiographica orientalis, 23132) in the
context of frontier politics and culture.

Page 191
tamia. The spread of belief in S. Sergius's power and
subsequently of the influence of the saint and his cult
center drew much of its distinctive strength from
Rusafa's location in the fluid frontier zone, where
knowledge traveled widely and among varied groups.
By the time of the Islamic conquests, when the Barbarian Plain lost its political significance, the cult was
widely disseminated in Syria and Mesopotamia, and
even further afield, where Sergius could not anyhow
be so closely associated with frontier defense. What
started as a frontier religion very much part of a place
and a time, had acquired the momentum to cross other boundaries, less mindful of geography, between
Christianity and Islam.

Page 192
This page intentionally left blank.

Page 193

BIBLIOGRAPHY
Articles in encyclopaedias, the Bulletin pigraphique,
and The Cambridge History of Iran are not included in
the bibliography.

PRIMARY SOURCES
Acta Anastasii Persae. Edited and translated by B.
Flusin. In id., Saint Anastase le Perse et l'histoire de
la Palestine au dbut du VIIe sicle, 1: 391. Paris,
1992.
Acta conciliorum oecumenicorum. Edited by E.
Schwartz. Strasbourg, 1914.
Acta Mar Kardagh martyris. Edited by J. B. Abbeloos.
AB 9 (1890): 5106
Acta S. Theclae. Edited by G. Dagron in Vie et miracles de Sainte Thcle, 167283. Brussels, 1978.
Agapius of Manbij. Kitab
. Edited and translated
by A. Vasiliev. PO 5.4, 7.4, 8.3, 11.1.

607/754

Agathias. Historiae. Edited by R. Keydell. Berlin,

1967.
al-Akhtal. Diwan. Edited by A. Salhani. Beirut,
18911925.
Ammianus Marcellinus. Res gestae. Edited by C. U.
Clark. Berlin, 191015.
ibn Matta. Revision of Mari ibn Sulayman, Kitab almijdal. Edited by E. Gismondi. In Maris Amri et Slibae
de patriarchis nestorianorum commentaria. Rome,
189699.
Anthologia graeca. Edited by H. Beckby. Munich,
195758.
Arab Jacobite Synaxarion. Coptic recension. Edited
and translated by R. Basset. PO 1.3, 3.3, 11.5, 16.2,
17.3, 20.5.
Armenian Synaxarion of Ter Israel. Edited and translated by G. Bayan. PO 5.3, 6.2, 15.3, 16.1, 18.1, 19.1,
21.16.

608/754

Babai the Great. History of Mar Giwargis. In Histoire

de Mar Jabalaha, de trois autres patriarches et de
deux laques nestoriens, ed. P. Bedjan, 416571.
Leipzig, 1895. Translated by O. Braun under the title
Ausgewhlte Akten persischer Mrtyrer mit einem Anhang: Ostsyrisches Mnchsleben (Kempten, 1915),
22177.

Page 194
al-Baladhuri, Ahmad ibn Yahya. Futuh al-buldan.
Edited by M. J. de Goeje. Leiden, 1866. Translated by
P. K. Hitti and F. C. Murgotten under the title The Origins of the Islamic State (New York, 191624).
Bar Hebraeus. Chronicon ecclesiasticum. Edited and
translated by J. B. Abbeloos and Th.-J. Lamy. Louvain, 187277.
Basil of Caesarea. Epistolae. Edited and translated by
Y. Courtonne. Paris, 195766.
. Homiliae. PG 31.164617.
al-Biruni,
Muhammad ibn Ahmad. Kitab alathar al-baqiya. Edited by C. E. Sachau. Leipzig,
187678. Translated by C. E. Sachau under the title
The Chronology of the Ancient Nations (London,
1879).
al-Buhturi. Diwan. Edited by R. Sarkis. Beirut, 1911.
Calendar of Edessa of 411. Edited and translated by
F. Nau under the title Un Martyrologue syriaque. PO
10.1.1128.

610/754

Cedrenus, George. Compendium historiarum. Edited

by I. Bekker. Bonn, 183839.
Choniates, Nicetas. Historia. Edited by J. L. van Dieten. Berlin, 1975.
Choricius of Gaza. Laudatio Marciani 1. Edited by R.
Foerster and C. Richtsteig in Choricii Gazaei opera,
126. Leipzig, 1929.
Chronicle of Amida of 569 = Historia ecclesiastica
Zachariae Rhetori vulgo adscripta 2. Edited by E. W.
Brooks. Paris, 1921. Translated by F. J. Hamilton and
E. W. Brooks under the title The Syriac Chronicle
Known as That of Zachariah of Mitylene (London,
1899).
Chronicle of Edessa of 506. In Chronicle of Zuqnin
775. Edited by J.-B. Chabot in Incerti auctoris chronicum anonymum pseudo-Dionysianum vulgo dictum, 1:
235317. Paris, 1927. Latin translation by J.-B. Chabot in Incerti auctoris chronicum anonymum pseudoDionysianum vulgo dictum, 1: 174233 (Paris, 1949).
English translation by W. Wright under the title The
Chronicle of Joshua the Stylite, Composed in Syriac
A.D. 507 (Cambridge, 1882).

611/754

Chronicle of Seert. Edited and translated by A. Scher

and others. PO 4.3, 5.2, 7.2, 13.4.
Chronicle of Zuqnin of 775. In Incerti auctoris chronicum anonymum pseudo-Dionysianum vulgo dictum,
ed. J.-B. Chabot, vol. 2. Paris 1933. French translation by R. Hespel in Incerti auctoris chronicum anonymum pseudo-Dionysianum vulgo dictum, vol. 2
(Louvain, 1989).
Chronicon ad annum Christi 1234 pertinens. Edited by
J.-B. Chabot. Louvain, 191620. Latin translation by
J.-B. Chabot, 1 (Louvain, 1937). French translation by
A. Abouna, 2 (Louvain, 1974). Partial English translation in A. Palmer, The Seventh Century in the West
Syrian Chronicles, 111221 (Liverpool, 1993).
Chronicon Pascale. Edited by L. Dindorf. Bonn, 1832.
Cinnamus, John. Epitome. Edited by A. Meineke.
Bonn, 1836.
Codex Justinianus. Edited by P. Krueger. Berlin,
1877.

612/754

Constantine Porphyrogenitus. De ceremoniis aulae

Byzantinae. Edited by J. J. Reiske. Bonn, 182930.
Edited and translated by A. Vogt in Le Livre des crmonies, 1: 183 (Paris, 193539).
Cyril of Scythopolis. Vita Euthymii. Edited by E.
Schwartz. Leipzig, 1939.
Denha. History of Maruta. Edited and translated by F.
Nau. PO 3.1.6396.
Digenis Akritis. Edited and translated by E. Jeffreys.
Cambridge, 1998.

Page 195
Documenta ad origines monophysitarum illustrandas.
Edited and translated by J.-B. Chabot. Paris, 1907,
1933.
Elias, patriarch of Jerusalem. De SS. Sergio et Baccho. In Analecta Sacra, ed. J. B. Pitra, 1: 28991.
Paris, 1876.
Ephrem the Syrian. Carmina nisibena. Edited and
translated by E. Beck. Louvain, 196163.
Eusebius. Historia ecclesiastica.
Schwartz. Leipzig, 19039.

Edited

by

E.

. Liber de martyribus Palaestinae. In Historia

ecclesiastica, ed. E. Schwartz, 2: 907502. Leipzig,
1908.
Evagrius. Historia ecclesiastica. Edited by J. Bidez
and L. Parmentier. London, 1898.
George of Cyprus. Descriptio orbis romani. Edited by
H. Gelzer. Leipzig, 1890.
Gregory of Nazianzus. Orationes. PG 3536.

614/754

Gregory of Nyssa. De sancto Theodoro. Edited by J.

P. Cavarnos. In Gregorii Nysseni sermones, ed. G.
Heil et al., 2: 6171. Leiden, 1990.
. Hom. in XL martyres II. Edited by O. Lendle.
In Gregorii Nysseni sermones, ed. G. Heil et al., 2:
15969. Leiden, 1990.
Gregory of Tours. Historia francorum. Edited by B.
Krusch and W. Levison. MGH, SRM 1.1.
Liber in gloriam martyrum. Edited by B.
Krusch. In MGH, SRM, 1: 484561. Berlin, 1885.
Translated by R. Van Dam (Liverpool, 1988).
. De virtutibus sancti Martini episcopi. Edited by
B. Krusch. In MGH, SRM, 1: 584661. Berlin, 1885.
Translated by R. Van Dam in Saints and Their Miracles in Late Antique Gaul (Princeton, 1993),
199303.
. Liber vitae patrum. Edited by B. Krusch. In
MGH, SRM, 1: 661744. Berlin, 1885.
al-Harawi, . Kitab al-isharat ila
al-ziyarat. Edited
by J. Sourdel-Thomine. Damascus, 1953. Translated

615/754

by id. under the title Guide des lieux de plerinage

(Damascus, 1957).
al-Harawi, Hassan ibn Thabit. Diwan. Edited by W. N.
. London, 1971.
History of Ahudemmeh. Edited and translated by F.
Nau. PO 3.1.751.
History of the Patriarchs of Alexandria. Edited and
translated by B. Evetts. PO 1.2, 1.4, 5.1, 10.5.
Hugeburc. Life of Willibald. Edited by O. Holder-Egger. MGH 15.1: 86106. Berlin, 1887.
Ibn Hazm, Abu Muhammad
ibn Ahmad. Jamharat
ansab
. Edited by . Harun. Cairo, 1962.
Ibn Ishaq, Muhammad. Sirat rasul Allah. Recension of
Abu Muhammad
al-Malik ibn Hisham. Edited by F.
Wstenfeld. Gttingen, 185860. Translated by A.
Guillaume under the title The Life of Muhammad (London, 1955).
Ibn Shaddad,
al-Din.
al-khatira fi dhikr
alSham wa 'l-Jazira. Translated by A.-M. Edd-Terrasse

616/754

under the title Description de la Syrie du Nord

(Damascus, 1984).
al-Idrisi, Abu
Muhammad. Nuzhat al mushtaq fi
ikhtiraq al-afaq. Edited by E. Cerulli et al. In Opus
Geographicum. Leiden, 197084.
al-Isfahani,
. Kitab al-aghani. Edited by A.
Muhanna and S. Y. Jabir. Beirut, 1982.

Page 196
Jacob of Sarug. Mimra on the Victorious Sergius and
Bacchus. Edited by P. Bedjan. In AMS, 6: 65061.
Translated by A. Palmer (unpublished).
. Mimra on the Victorious Sergius and Bacchus. Translated by Andrew Palmer. N.d.
James of Edessa. The Hymns of Severus of Antioch
and Others. Edited and translated by E. W. Brooks.
PO 6.1, 7.5.
Jarir and al-Farazdaq.
. Edited by A. A. Bevan under the title The
of Jarir and al-Farazdaq. Leiden,
190512.
John Chrysostom.
50.66266.

Homilia

in

martyres.

PG

John Diacrinomenus. Fragmenta. Edited by G. C.

Hansen. In Theodore Anagnostes, Historia ecclesiastica, 15257. 2d ed. Berlin, 1995.
John of Ephesus. Historia ecclesiastica. Pars tertia.
Edited and translated by E. W. Brooks. Louvain,
193536.

618/754

. Vita sanctorum orientalium. Edited and translated by E. W. Brooks. PO 17.1, 18.4, 19.2.
John of Epiphania. Fragmenta. Edited by C. Mller. In
Fragmenta historicorum graecorum, 4: 27376. Paris,
1851.
John Lydus. De magistratibus populi romani. Edited
and translated by A. C. Bandy under the title On
Powers, or, The Magistracies of the Roman State.
Philadelphia, 1983.
John Malalas. Chronographia. Edited by L. Dindorf.
Bonn, 1831.
John Moschus. Pratum spirituale. PG 87.28513112
John the Persian. The History of Rabban bar . Metrical version by Abraham of Zabde. Edited and translated by E. A. W. Budge under the title The Histories
of Rabban
the Persian and Rabban Bar
(London, 1902).
Joseph the Scevophylax. Life of S. John, Abbot of S.
Sergius's Monastery. Edited by K. Kekelidze. In Etiudebi jveli kartuli literaturis istoriidan, 3: 26070.

619/754

Tbilisi, 1955. Translated by M. van Esbroeck under

the title "La Vie de Saint Jean higoumne de Saint
Serge par Joseph le Skevophylax," OC 80 (1996):
15366.
Joshua the Stylite. See Chronicle of Edessa of 506.
Lactantius. De mortibus persecutorum. Edited by J.
Moreau. Paris, 1954.
Leontius of Jerusalem. Contra monophysitas. PG 86.
17691902.
Malchus. Fragmenta. Edited and translated by R. C.
Blockley. In The Fragmentary Classicizing Historians
of the Later Roman Empire, 2: 40262. Liverpool,
1983.
Mari ibn Sulayman. Kitab al-mijdal. Edited by E. Gismondi. Maris Amri et Slibae de patriarchis nestorianorum commentaria. Rome, 189699.
Edited
1898.

by

A.

,
Papadopoulos-Kerameus.
In
, 5: 36067. Saint Petersburg,

620/754

Martyrum Persarum acta. Edited by H. Delehaye. PO

2.4.
. Muruj al-dhahab. Edited and translated by C.
Barbier de Meynard and J.-B. Pavet de Courteille under the title Maoudi: Les Prairies d'or. Paris,
186177. Revised by C. Pellat. Beirut, 196679 (edition). Paris, 1962 (translation).
Menander Protector. Fragmenta. Edited and translated by R. C. Blockley under the title The History of
Menander the Guardsman. Liverpool, 1985.
Michael the Syrian. Chronicle. Edited and translated
by J.-B. Chabot. Paris, 18991910.

Page 197
Miracula S. Theclae. Edited and translated by G. Dagron in Vie et miracles de Sainte Thcle, 284421.
Brussels, 1978.
The Monks of
, Emperor of China. Translated
by E. A. Budge. London, 1928.
Notitia dignitatum. Edited by O. Seeck. Berlin, 1876.
Olympiodorus. Fragmenta. Edited and translated by
R. C. Blockley in The Fragmentary Classicising Historians of the Later Roman Empire, 2: 152209. Liverpool, 1983.
Pachymeres, George. De Michaele et Andronico Paleologis. Edited by A. Failler. Paris, 1984.
Palladius. Historia Lausiaca. Edited by G. J. M. Bartelink. N.p., 1974.
Passio antiquior SS. Sergii et Bacchi. Edited by I. van
de Gheyn. AB 14 (1895): 37195. Latin redaction, ed.
C. Byeus, Acta sanctorum Octobris 3: 86370. Metaphrastic redaction, ed. C. Byeus, Acta sanctorum

622/754

Octobris 3: 87182. Syriac redaction, ed. P. Bedjan,

AMS 3: 28322.
Patria Konstantinoupoleos. In Scriptores originum
Constantinopolitanarum, ed. Th. Preger, 118. Leipzig
19017.
Paulinus of Nola. Carmina. Edited by G. de Hartel.
CSEL 30. Vienna, 1894.
Paulus Silentiarius. Descriptio Sanctae Sophiae.
Edited by P. Friedlnder. Leipzig, 1912.
Photius. Bibliotheca. Edited by R. Henry. Paris,
195977.
Piacenza Pilgrim. Itinerarium. In Itinera Hierosolymitana saeculi IIIIVII, ed. P. Geyer. CSEL 39. Vienna,
1890.
Procopius. De aedificiis. Edited by J. Haury. Revised
by G. Wirth. Leipzig, 1964.
. De bello persico. Edited by J. Haury. Revised
by G. Wirth. Leipzig, 1962.

623/754

Prudentius. Peristephanon. Edited by J. Bergman. Vienna, 1926. Translated by H. J. Thomson

(Cambridge, Mass., 1953).
Ptolemy. Geographia. Edited by C. Mller. Paris,
18831901.
Ravenna Geographer. In Ravennatis Anonymi Cosmographia et Guidonis Geographica. Edited by M.
Pinder and G. Parthey, 1445. Berlin, 1860.
Sebeos.
. Edited by V. G. Abgarian. Erevan,
1979. Translated by C. Gugerotti under the title
,
Storia (Verona, 1990).
Severus of Antioch. Homiliae cathedrales. Edited and
translated by R. Duval and others. PO 4.1, 8.2, 12.1,
16.5, 20.2, 22.2, 23.1, 25.1, 26.3, 29.1, 35.3, 36.1,
36.3.
Scylitzes, John. Synopsis historiarum. Edited by H.
Thurn. Berlin, 1973.
Socrates Scholasticus. Historia ecclesiastica. Edited
by G. C. Hansen. Berlin, 1995.

624/754

Sophronius. Miracula Cyri et Ioannis. Edited by N. F.

Marcos. In Los "thaumata" de Sofronio: Contribucion
al estudio de la "incubatio" cristiana. Madrid, 1975.
Sozomen. Historia ecclesiastica. Edited by J. Bidez.
Berlin, 1960.
Stephanus of Byzantium. Ethnica. Edited by A.
Meineke. Berlin, 1849.
Synaxarium ecclesiae Constantinopolitanae. Propylaeum ad Acta Sanctorum Novembris. Edited by H.
Delehaye. Brussels, 1902.
Synodicon orientale ou Recueil de synodes nestoriens. Edited and translated by J.-B. Chabot. Paris,
1902.
Syrian Calendar. Edited and translated by F. Nau under the title Douze mnologues syriaques. PO
10.1.2956.
al-Tabari, Abu
Muhammad ibn Jarir.
wa 'l-muluk. Edited by

al-rusul

Page 198
M. J. de Goeje et al. Leiden, 18791901. Partial translation by T. Nldeke under the title Geschichte der
Perser und Araber zur Zeit der Sasaniden: Aus der arabischen Chronik des Tabari (Leiden 1879). English
translation edited by E. Yar-Shater (New York 1985).
Tabula Peutingeriana. Facsimile edition with commentary by E. Weber. Graz, 1976.
Tarafa ibn
al-Bakri.
. Edited by
. In
Diwan Tarafa ibn
. Abu Dhabi, n.d. Translated by
A. J. Arberry in The Seven Odes: The First Chapter in
Arabic Literature, 8389 (London, 1957)
Theodoret. Graecarum affectionum curatio. Edited
and translated by P. Canivet. Paris, 1958.
. Historia religiosa. Edited and translated by P.
Carnivet and A. Leroy-Molinghen under the title Histoire des moines de Syrie. Paris, 197779.
Theophanes. Chronographia. Edited by C. de Boor.
Leipzig, 1883.

626/754

Theophylact Simocatta. Historia universalis. Edited by

C. de Boor. Revised by P. Wirth. Stuttgart, 1972. German translation by P. Schreiner (Stuttgart, 1985).
English translation by M. and M. Whitby (Oxford,
1986).
Thomas of Marga. The Book of Governors. Edited
and translated by E. A. Budge. London, 1893.
Translatio reliquiarum [Anastasii Persae]. Edited and
translated by B. Flusin in Saint Anastase le Perse, 1:
97107. Paris, 1992.
Vita et institutum piisimi patris nostri Alexandri. Edited
and translated by E. de Stoop. PO 6.5.
Vita Gulanducht. Georgian redaction. Edited and
translated by G. Garitte. AB 74 (1956): 42640. Greek
redaction by Eustratios presbyteros,
,
,
Kerameus,
Petersburg, 1897).

ed.

A. Papadopoulos, 4: 14979 (Saint

Vita Maruthae. Armenian edition. In

(Vitae et passiones sanctorum), 2: 1732. Venice,

627/754

1874. Translated by R. Marcus, HThR 25 (1932):

4771.
Vita Maruthae (BHG 2265). Edited and translated by
J. Noret. AB 91 (1973): 8091.
Vita Maruthae (BHG 2266). Edited and translated by
J. Noret. AB 91 (1973): 8091.
Vita Petri Iberi. Edited and translated by R. Raabe.
Leipzig, 1895.
Vita Simeonis Stylitae (Syriac). Translated by R. Doran. Kalamazoo, Mich., 1992. Based on Vat. Syr. 160
and the edition by S. E. Assemani, Acta sanctorum
martyrum orientalium et occidentalium, 2: 268398
(Rome, 1748).
West Syrian Synodicon. Edited and translated by A.
Vbus under the title The Synodicon in the West
Syrian Tradition. Louvain, 197576.
Xanthopoulus, Nicephorus Callistus. Ecclesiastica historia. PG 145. 559147.448.

628/754

Yaqut b.
195557.

al-Hamawi.

al-buldan. Beirut,

Zachariah of Mitylene. Chronicle. Edited by E. W.

Brooks as Historia ecclesiastica Zachariae Rhetori
vulgo adscripta. Paris, 191921. Translated by F. J.
Hamilton and E. W. Brooks under the title The Syriac
Chronicle Known as That of Zachariah of Mitylene
(London, 1899).
Zosimus. Historia nova. Edited and translated by F.
Paschoud. Paris, 197189.

Page 199

SECONDARY LITERATURE
al-Ghani,
Damascus, 1993.

al-Hira fi al-jahiliya wa 'l-islam.

Aigrain, R. "Sur quelques inscriptions d'glises de

l'poque byzantine." OCP 13 (1947): 1827.
, H. Sadad fi
Assemani,
171928.

J.

. Damascus, 1995.
S.

Bibliotheca

orientalis.

Rome,

Bakirtzis, Ch., ed.

.

Athens,

1997.
Balty, J. Mosaques antiques de la Syrie. Brussels,
1977.
Barnes, T. D. The New Empire of Diocletian and Constantine. Cambridge, Mass., 1982.
Barnish, S. J. B. "Late Roman Prosopography Reassessed." JRS 84 (1994): 17177.

630/754

Baumstark, A. Geschichte der syrischen Literatur mit

Ausschlu der christlich-palstinensischen Texte.
Bonn, 1922.
Baur, P. V. C., and M. I. Rostovtzeff, eds. The Excavations at Dura-Europos: Preliminary Report of Second
Season of Work, October 1928April 1929. New
Haven, Conn., 1931.
Bauzou, Th. "pigraphie et toponymie: Le Cas de la
Palmyrne du sud-ouest." Syria 70 (1993): 2750.
. "Routes romaines de Syrie." In Archologie et
histoire de la Syrie: La Syrie de l'poque achmnide
l'avnement de l'Islam, ed. J.-M. Dentzer and W.
Orthmann, 2: 20521. Saarbrcken, 1989.
. "Les Voies de communication dans le Hauran
l'poque romaine." In Hauran 1: Recherches
archologiques sur la Syrie du sud l'poque hellnistique et romaine, ed. J.-M. Dentzer, 13765. Paris,
1985.
Beard, M. "Writing and Religion: Ancient Literacy and
the Function of the Written Word in Roman Religion."

631/754

In Literacy in the Roman World, ed. J. H. Humphrey,

3558. Ann Arbor, Mich., 1991.
. "Writing and Ritual: A Study of Diversity and
Expansion in the Arval Acta." PBSR 53 (1985): 1462.
Beck, H.-G. Kirche und theologische Literatur im byzantinischen Reich. Munich, 1959.
Beck, R. "The Mithras Cult as Association." Studies in
Religion 21 (1992): 313.
Bell, G. The Churches and Monasteries of the Tur
. Edited and revised by M. Mundell Mango. London,
1982.
Bernbeck, R. Steppe als Kulturlandschaft. Das
Ostsyriens vom Neolithikum bis zur islamischen Zeit.
Berlin, 1993.
Bishop, E. Liturgica historica: Papers on the Liturgy
and Religious Life of the Western Church. Oxford,
1918.

632/754

Blair, S. "What Is the Date of the Dome of the Rock?"

In Bayt al-Maqdis: al-Malik's Jerusalem, ed. J. Raby
and J. Johns, 1: 5987. Oxford, 1992.
Blanck, H. Wiederverwendung alter Statuen als
Ehrendenkmler bei Griechen und Rmer. Rome,
1969.
Blockley, R. C. East Roman Foreign Policy: Formation
and Conduct from Diocletian to Anastasius. Leeds,
1992.
Blunt, Lady Anne. Bedouins of the Euphrates. London, 1879.
Bonner, C. Studies in Magical Amulets, Chiefly
Graeco-Egyptian. Ann Arbor, Mich., 1950.

Page 200
Borkowski, Z. Inscriptions des factions Alexandrie.
Warsaw, 1981.
Boswell, J. Same-Sex Unions in Pre-Modern Europe.
New York, 1994.
Boucheman, A. de. Une Petite Cit caravanire:
Damascus, 1937.

Bowersock, G. W. Roman Arabia. Cambridge, Mass.,

1983.
. "Social and Economic History of Syria under
the Roman Empire." In Archologie et histoire de la
Syrie, vol. 2: La Syrie de l'poque achmnide
l'avnement de l'Islam, ed. J.-M. Dentzer and W.
Orthmann, 6380. Saarbrcken, 1989.
. "Syria under Vespasian." JRS 63 (1973):
13340.
Bowman, G. "Nationalizing the Sacred: Shrines and
Shifting Identities in the Israeli-occupied territories."
Man 28 (1993): 43160.

634/754

Brands, G. "Architektur und Bauornamentik von

Resafa-Sergiupolis." Diss., Freie Universitt Berlin,
1994. Forthcoming as Resafa, vol. 6: Die Bauornamentik von Resafa-Sergiupolis.
. "Die Entstehung einer StadtBeobachtungen
zur Bauornamentik von Resafa." In Sptantike und
byzantinische Bauskulptur, ed. U. Peschlow and S.
Mllers, 5974. Stuttgart, 1998.
. "Martyrion und BischofskircheAnmerkungen
zur Architektur und Bauornamentik des Zentralbaus
von Resafa." In Akten des 12. Internationalen Kongresses fr Christliche Archologie, 59097. Mnster,
1995.
. "Der sogenannte Audienzsaal des al-Mundhir
in Resafa." DaM 10 (1998): 21135.
Brire, M. "Introduction gnrale toutes les
homlies." In Severus of Antioch, Hom. cath.
CXXCXXV. PO 29.1.572.
Brinker, W. "Zur Wasserversorgung von Resafa-Sergiupolis." DaM 5 (1991): 11946.

635/754

Brinkmann, A. "Der rmische Limes im Orient." BJ 99

(1896): 25257.
Brock, S. "The Syriac Background." In Archbishop
Theodore: Commemorative Studies on His Life and
Influence, ed. M. Lapidge, 3053. Cambridge, 1995.
. "Syriac Culture, 337425." Cambridge Ancient History, 13: 70819. Cambridge, 1998.
Brown, P. The Cult of the Saints: Its Rise and Function in Latin Christianity. Chicago, Il., 1981.
. "The Diffusion of Manichaeism in the Roman
Empire." JRS 59 (1969): 92103. Reprinted in Religion and Society in the Age of Saint Augustine,
94118 (London, 1972).
Brnnow, R. E., and A. von Domaszewski. Die Provincia Arabia. Strasbourg, 19049.
Bryce, J. "Life of Justinian by Theophilus." EHR 2
(1887): 65786.

636/754

Bulliet, R. The Camel and the Wheel. Cambridge,

Mass., 1975. Reprinted with a new preface. New
York, 1990.
Butler, H. C. Early Churches in Syria: Fourth to
Seventh Centuries. Edited by E. Baldwin Smith. Princeton, N.J., 1929.
Cameron, Alan. Porphyrius the Charioteer. Oxford,
1973.
Cameron, Averil, "Agathias on the Sassanians." DOP
2324 (196970): 67183.
. The Mediterranean World in Late Antiquity,
A.D. 395600. London, 1993.
Canaan, T. Mohammedan Saints and Sanctuaries in
Palestine. Jerusalem, n.d. Reprinted from the Journal
of the Palestine Oriental Society, 1927.
Candea, V., and S. Agmian, eds. Icones Melkites.
Beirut, 1969.

Page 201
Candemir, H., and J. Werner. "Christliche Mosaiken in
der nrdlichen Euphratesia." In Studien zur Religion
und Kultur Kleinasiens: Festschrift fr F. K. Drner,
ed. S.
, E. Schwertheim, and J. Werner, 192231.
Leiden, 1978.
Canivet, M.-T. "Le Reliquaire huile de la grande glise de
(Syrie)." Syria 55 (1978): 15362.
Canivet, P., and M.-T. Canivet.
: Sanctuaire chrtienne d'Apamne (IVeVIes.) Paris, 1987.
Caquot, A. "Inscriptions judo-arabes de
opolis)." Syria 32 (1955): 7074.

(Sergi-

Caussin de Perceval, A. P. Essai sur l'histoire des

Arabes avant l'Islamisme, pendant l'poque de Mahomet, et jusqu' la rduction de toutes les tribus sous la
loi musulmane. Paris, 184748.
Chabot, J.-B. Littrature syriaque. Paris, 1934.
. "Notes d'pigraphie et d'archologie orientale." JA 16 (1900): 24988.

638/754

Charles, H. Le Christianisme des Arabes nomades

sur le limes et dans le dsert syromsopotamien aux
alentours de l'Hgire. Paris, 1936.
Chaumont, M.-L. "O les rois sassanides taint-ils
couronns?" JA 252 (1964): 5975.
Chesney, F. R. The Expedition for the Survey of the
Rivers Euphrates and Tigris, 183537. London, 1850.
Christensen, A. L'Iran sous les Sassanides. 2d ed.
Copenhagen 1944.
Ciggar, K. "Description de Constantinople traduite par
un plerin anglais." REB 34 (1976): 21167.
Clarke, G. W. "Syriac Inscriptions from the Middle Euphrates." Abr-Nahrain 23 (198485): 7382.
Conrad, L. "Epidemic Disease in Central Syria in the
Late Sixth Century: Some New Insights from the
Verse of
ibn
." BMGS 18 (1994): 1258.
. "Historical Evidence and the Archaeology of
Early Islam." In Quest for Understanding: Arabic and

639/754

Islamic Studies in Memory of Malcom H. Kerr, ed. S.

Seikaly et al., 26382. Beirut, 1991.
Contenson, H. de. "Quatrime et cinquime campagnes Tell Ramad, 19671968." AAS 19 (1969):
2530.
Contini, R. "Il Hawran preislamico: Ipotesi di storia linguistica." Felix Ravenna 4 (1987): 2579.
Cook, M. "The Heraclian Dynasty in Muslim Eschatology."
: Revista de Estudios Arabes 13 (1992):
323.
Creswell, K. A. C. Early Muslim Architecture. 2d ed.
Vol. 1: Umayyads. Oxford, 1969.
Croke, B., and J. Crow. "Procopius and Dara." JRS 73
(1983): 14359.
Cruikshank Dodd, E. "The Monastery of Mar Musa alHabashi, near Nebek, Syria." Arte medievale 1
(1992): 61130.
Cumont, F. tudes syriennes. Paris, 1917.

640/754

Curtiss, S. I. Ursemitische Religion im Volksleben des

heutigen Orients. Forschungen und Funde aus Syrien
und Palstina. Translated into German by H. Stocks,
with additional material by the author. Leipzig, 1903.
Originally published as Primitive Semitic Religion
Today: A Record of Researches, Discoveries and
Studies in Syria, Palestine and the Sinaitic Peninsula
(Chicago, Il., 1902).
Dagron, G. Vie et miracles de Sainte Thcle. Brussels, 1978.

Page 202
Daumas, F., and A. Guillaumont. Kellia 1. Kom 219:
Fouilles excutes en 1964 and 1965. Cairo, 1969.
Dauphin, C. "Encore des judo-chrtiennes au
Golan?" In Early Christianity in Context: Monuments
and Documents, ed. F. Manns and E. Alliata, 6984.
Jerusalem, 1993.
. "Golan Survey, 19811982." IEJ 33 (1983):
11213.
. "Golan Survey, 1983." IEJ 34 (1984): 26869.
. "Golan Survey, 19841986." IEJ (1986):
27375.
. "Paysages antiques du Golan." Archologia
294 (1993): 5057.
. "Plerinage ghassanide au sanctuaire byzantin de Saint Jean-Baptiste er-Ramthaniyye en
Gaulanitide." In Akten des XII. internationalen Kongresses fr christliche Archologie, 66773. Mnster,
1995.

642/754

. "Er-Ramthaniyye: Surveying an Early Bedouin

Byzantine Pilgrimage Centre in the Golan Heights."
BA-IAS 8 (198889): 8284.
. "Villages dserts juifs et chrtiens du Golan."
Archologia 297 (1994): 5264.
Dauphin, C., S. Brock, R. C. Gregg, and A. F. L. Beeston. "Paens, juifs, judo-chrtiens, chrtiens, et musulmans en Gaulanitide: Les Inscriptions de
, Kafr
Naffakh, Farj et er-Ramthaniyye." P-OChr 46 (1996):
30540.
Dauphin, C., and S. Gibson. "Ancient Settlements in
Their Landscapes: The Results of Ten Years of Survey on the Golan Heights (19781988)." BA-IAS 12
(199293): 731.
. "Landscape Archeology at er-Ramthaniyye in
the Golan Heights." In Archologie et espaces: Actes
des Xe rencontres internationales d'archologie et
d'histoire d'Antibes, 192021 octobre 1989, ed. J.-L.
Fiches and S. van der Leeuw, 3545. Juanles-Pins,
1990.

643/754

Degen, R. "Die Inschriften." In T. Ulbert, Resafa, vol.

3: Der Kreuzfahrerzeitliche Silberschatz aus ResafaSergiupolis, 6576. Mainz, 1990.
Delbrueck, R. Die Consulardiptychen und verwandte
Denkmler. Berlin, 1929.
Delehaye, H. Les Lgendes grecques des saints militaires. Paris, 1909.
. Les Lgendes hagiographiques. Brussels,
1927.
. "Les Martyrs d'Egypte." AB 40 (1922): 5154.
. Mlanges d'hagiographie grecque et latine.
Brussels, 1966.
. Les Origines du culte des martyrs. 2d ed.
Brussels, 1933.
. Les Passions des martyrs et les genres littraires. Brussels, 1921.

644/754

. Review of Die Chronologie der altchristlichen

literatur bis Eusebius, by A. Harnack. AB 23 (1904):
47680.
, ed. and tr. Les Versions grecques des actes
des martyrs persans sous Sapor II. PO 2.4.
Delehaye, H., et al., eds. Propylaeum ad Acta Sanctorum Decembris. Martyrologium romanum ad formam editionis typicae. Brussels, 1940.
Descoeudres, G. Die Pastophorien im syro-byzantinischen Osten: Eine Untersuchung zu architektur- und
liturgie-geschichtlichen Problemen. Wiesbaden, 1983.
Devreesse, R. Le Patriarcat d'Antioche depuis la paix
de l'glise jusqu' la conqute arabe. Paris, 1945.
Dillemann, L. Haute Msopotamie orientale et pays
adjacents: Contribution la gographie historique de
la rgion, du Ve s. avant l're chrtienne au VIe de
cette re. Paris, 1962.

Page 203
Doe, B. Monuments of South Arabia. Naples, 1983.
Donceel-Vote, P. "La Chapelle des martyrs: Dcor,
archologie et liturgie." In J. Noret, "La Socit belge
d'tudes byzantines depuis 1990." Byzantion 63
(1993): 43740.
. Les Pavements des glises byzantines de
Syrie et du Liban: Dcor, archologie et liturgie.
Louvain-la-Neuve, 1988.
. "Le Rle des reliquaires dans les plerinages." In Akten des XII. internationalen Kongresses
fr christliche Archologie, 184205. Mnster, 1995.
Donner, F. Mc. The Early Islamic Conquests. Princeton, N.J., 1981.
. "The Role of Nomads in the Near East in Late
Antiquity (400800 C.E.)." In Tradition and Innovation
in Late Antiquity, ed. F. Clover and S. F. Humphreys,
7385. Madison, Wis., 1989.
Doughty, C. M. Travels in Arabia Deserta. 3d ed. London, 1936.

646/754

Downey, G. A History of Antioch in Syria from Seleucus to the Arab Conquest. Princeton, N.J., 1961.
Dunand, M. "Nouvelles inscriptions du Djebel Druze et
du Hauran." In Mlanges syriens offerts Monsieur
Ren Dussaud, par ses amis et ses lves. Paris,
1939.
Durliat, J. Les Ddicaces d'ouvrages de dfense dans
l'Afrique byzantine. Rome, 1981.
Dussaud, R. La Pntration des Arabes en Syrie avant l'Islam. Paris, 1955.
. Topographie historique de la Syrie antique et
mdivale. Paris, 1927.
Dussaud, R., and M. F. Macler. "Rapport sur une mission scientifique dans les rgions dsertiques de la
Syrie moyenne." Nouvelles archives des missions scientifiques et littraires 10 (1902): 411744.
Duval, Y. Loca sanctorum Africae: Le Culte des martyrs en Afrique du IVe au VIIe sicle. Rome, 1982.

647/754

Ebersolt, J., and A. Thiers. Les glises de Constantinople. Paris, 1913.

Esbroeck, M. van. "La Naissance du culte de Saint
Barthlmy en Armenie." REArm 17 (1983): 17195.
. "La Vie de Saint Jean higoumne de Saint
Serge par Joseph le Skevophylax." OC 80 (1996):
15366.
Essen, C. van, and M. Vermaseren. The Excavations
in the Mithraeum of the Church of Sta Prisca in Rome.
Leiden, 1965.
Ettinghausen, R., and O. Grabar. The Art and Architecture of Islam, 6501250. Harmondsworth, Middlesex, 1987.
Fars, B. Vision chrtienne et signes musulmans.
Cairo, 1961.
Farioli Campanati, R. "Per un'introduzione sul
tardoantico." Felix Ravenna 14144 (199192):
177232.

648/754

Feissel, D. "Magnus Mgas et les curateurs des

'maisons divines' de Justin II Maurice." T&MByz 9
(1985): 46576.
. "Noms de villages de Syrie du nord: Elements
grecs et smitiques." In O
, 287302.
Athens, 1991.
Feissel, D., and J. Gascou. "Documents d'archives romains indits du Moyen Euphrate (IIIe sicle aprs J.C.)." CRAI 1989: 53561.
Fiey, J. M. Assyrie chrtienne: Contribution l'tude
de l'histoire et de la gographie ecclsiastiques et
monastiques du nord de l'Iraq. Beirut, 1965.
. "Identification of Qasr Serej." Sumer 14
(1958): 12527.

Page 204
. "Martyropolis syriaque." Le Muson 89
(1976): 538.
. "Maruta de Martyropolis d'aprs ibn al-Azraq
(1181)." AB 94 (1976): 3545.
. Mossoul chrtienne: Essai sur l'histoire,
l'archologie et l'tat actuel des monuments chrtiennes de la ville de Mossoul. Beirut, 1959.
. "Les Saints Serge de l'Iraq." AB 79 (1961):
10214.
. "Tagrt: Esquisse d'histoire
L'Orient Syrien 8 (1963): 289341.

chrtienne."

Flusin, B. Saint Anastase le Perse et l'histoire de la

Palestine au dbut du VIIe sicle. Paris, 1992.
Fowden, E. K. "An Arab Building at Rusafa-Sergiupolis." DaM 11 (2000).
Fowden, G. "'Desert kites': Ethnography, Archaeology
and Art." In The Roman and Byzantine Near East, vol.
2, ed. J. Humphrey, 10736. Portsmouth, R.I., 1999.

650/754

. Empire to Commonwealth: Consequences of

Monotheism in Late Antiquity. Princeton, N.J., 1993.
Franchi de' Cavalieri, P. "Dei SS. Gioventino e
Massimino." In id., Note agiographiche, 9: 169200.
Studi e testi, 175. Vatican City, 1953.
Frank, R. I. Scholae palatinae: The Palace Guards of
the Late Roman Empire. Rome, 1969.
French, D. "A Road Problem: Roman or Byzantine?"
IstMit 43 (1993): 44554.
Frend, W. H. C. The Rise of the Monophysite Movement: Chapters in the History of the Church in the
Fifth and Sixth Centuries. Cambridge, 1972. Corrected reprint, 1979.
Freyberger, K. S. "Der Tempel in
DaM 5 (1991): 938.

: Ein Bericht."

. "Untersuchungen zur Baugeschichte des

Jupiter-Heiligtums in Damaskus." DaM 4 (1989):
6186.

651/754

Galtier, E. "Contribution l'tude de la littrature

arabe-copte." BIFAO 4 (1905): 105221.
Garitte, G. "La Passion gorgienne de Sainte
Golindouch." AB 74 (1956): 40540.
Garsoan, N. "The Iranian Substratum of the
Cycle." In East of Byzantium: Syria and Armenia in
the Formative Period, ed. N. Garsoan, T. Mathews,
and R. Thomson, 15174. Washington, D.C., 1982.
. "Le Rle de l'hirarchie chrtienne dans les
relations diplomatiques entre Byzance et les Sassanides." REArm 10 (197374): 11938.
Gatier, P.-L. "Inscriptions grecques de Rsafa." DaM
10 (1998): 23741.
. Inscriptions de la Jordanie, vol. 2: Rgion
centrale (Amman, Hesban, Madaba, Main, Dhiban).
Paris, 1986.
. "Un Moine sur la frontire: Alexandre
l'Acmte en Syrie." In Frontires terrestres,
frontires clestes dans l'antiquit, ed. A. Rousselle,
43557. Paris, 1995.

652/754

. "Les Traditions de l'histoire du Sina du IVe au

VIIe sicle." In L'Arabie prislamique et son environnement historique et culturel, ed. T. Fahd, 499523.
Leiden, 1989.
Gatier, P.-L., and Th. Ulbert. "Eine Trsturzinschrift
aus Resafa-Sergiupolis." DaM 5 (1991): 16982.
Gaube, H. Ein arabischer Palast in Sdsyrien:
. Beirut, 1974.
Gawlikowski, M. "Palmyra as a Trading Centre." Iraq
56 (1994): 2733.
. Review of The Roman Near East, 31
B.C.A.D. 337, by F. Millar. JRS 84 (1994): 24446.
Gero, S. "The Legend of the Monk
Cross and Iconoclasm."

, the Cult of the

Page 205
In La Syrie de Byzance l'Islam VIIeVIIIe sicles,
ed. P. Canivet and J.-P. Rey-Coquais, 4758. Damascus, 1992.
Gessel, W. "Das l der Mrtyrer: Zur Funktion und Interpretation der lsarkophage von Apamea in Syrien."
OC 72 (1988): 182202.
Gilles, P. De topographia Constantinopoleos, et de illius antiquitatibus. Leiden, 1562.
Gilliot, C. "Une Leon magistrale d'Orientalisme:
L'Opus magnum de J. van Ess." Arabica 40 (1993):
345402.
Gogrfe, R. "Die Datierung des Tempels von Isriye."
DaM 7 (1993): 4561.
Goldziher, I. Muslim Studies. Edited by S. M. Stern.
Translated from the German by C. R. Barber and S.
M. Stern. Chicago, Il., 196771.
Gordon, R. "Reality, Evocation and Boundary in the
Mysteries of Mithras." JMS 3 (1980): 1999.

654/754

Grabar, A. Martyrium: Recherches sur le culte des

reliques et l'art chrtien antique. Paris, 1946.
. "Recherches sur les sources juives de l'art
palo-chrtien." CArch 11 (1960): 4071.
Grabar, O. The Formation of Islamic Art. 2d ed. New
Haven, Conn., 1987.
. The Shape of the Holy: Early Islamic Jerusalem. Princeton, N.J., 1996.
Graf, D. "Camels, Roads and Wheels in Late
Antiquity." Electrum 1 (1997): 4349.
. "Rome and the Saracens: Reassessing the
Nomadic Menace." In L'Arabie prislamique et son
environnement historique et culturel, ed. T. Fahd,
341400. Leiden, 1989.
Graf, G. Geschichte der christlichen arabischen Literatur. Studi e testi, 118, 133, 146, 147, 172. Vatican
City, 194453.
Grant, C. P. The Syrian Desert: Caravans, Travel and
Exploration. London, 1937.

655/754

Gregory, S. Roman Military Architecture on the

Eastern Frontier. Amsterdam, 199597.
Grimme, H. Palmyrae sive Tadmur urbis fata. Mnster, 1886.
Groom, N. A Dictionary of Arabic Topography and
Place Names. Beirut, 1983.
Hachili, R. "Late Antique Jewish Art from the Golan."
In The Roman and Byzantine Near East, Vol. 1, ed. J.
Humphrey, 183212. Ann Arbor, Mich., 1995.
Halkin, F. "Un Second Saint Gordius?" AB 79 (1961):
515.
, ed. Bibliotheca hagiographica graeca. Brussels, 1957.
Hamilton, R. Kirbat al Mafjar: An Arabian Mansion in
the Jordan Valley. Oxford, 1959.
. Walid and His Friends: An Umayyad Tragedy.
Oxford, 1988.

656/754

Hammer [-Purgstall], J. von. Constantinopolis und der

Bosporus, rtlich und geschichtlich beschrieben.
Pesth, 1822.
Harper, P. O. "Thrones and Enthronement Scenes in
Sasanian Art." Iran 17 (1979): 4965.
Harrauer, H. "'OYBPOM, Abrasax, Jahwe u.a. aus
Syrien." Tyche 7 (1992): 3944.
Harrison, M. Review of Die Stadtmauer von Resafa in
Syrien, by W. Karnapp. CR 34 (1984): 1056.
Hartmann, M. "Beitrge zur Kenntniss der syrischen
Steppe." ZDPV 22 (1899): 12749, 15377; 23
(1901): 177, 97158.
Harvey, S. A. Asceticism and Society in Crisis: John
of Ephesus and the Lives of the Eastern Saints.
Berkeley, Ca., 1990.
Hasluck, F. W. Christianity and Islam under the
Sultans. Edited by M. M. Hasluck. Oxford, 1929.

Page 206
Hatzidakis,

M.,
and
E.
. Athens, 198797.

Drakopoulou.

Helms, S., with contributions by A. V. G. Betts, W. and

F. Lancaster, and C. J. Lentzen. Early Islamic Architecture of the Desert: A Bedouin Station in Eastern
Jordan. Edinburgh, 1990.
Hendy, M. F. Studies in the Byzantine Monetary
Economy c. 3001450. Cambridge, 1985.
Hepper, F. N., and S. Gibson. "Abraham's Oak of
Mamre: The Story of a Venerable Tree." PEQ 126
(1994): 94105.
Higgins, M. J. "Chosroes II's Votive Offerings at Sergiopolis." BZ 48 (1955): 89102.
Hill, S. "The 'praetorium' at Mismiya." DOP 29 (1975):
34749.
Hgel, C. "The Redaction of Symeon Metaphrastes:
Literary Aspects of the Metaphrastic Martyria." In
Metaphrasis: Redactions and Audiences in Middle

658/754

Byzantine Hagiography, ed. C. Hgel, 721. Oslo,

1996.
Hoffmann, G. Auszge aus syrischen Akten persischer Mrtyrer bersetzt und durch Untersuchungen
zur historischen Topographie erlutert. Leipzig, 1880.
Hoffmann, P. Das sptrmische Bewegungsheer und
die Notitia Dignitatum. Dsseldorf, 196970.
Holum, K. "Pulcheria's Crusade, A.D. 42122, and the
Ideology of Imperial Victory." GRBS 18 (1977):
15372.
. Theodosian Empresses: Women and Imperial
Dominion in Late Antiquity. Berkeley, Ca., 1982.
Honigmann, E. Evques et vchs monophysites
d'Asie antrieure au VIe sicle. Brussels, 1951.
. Die Ostgrenze des byzantinischen Reiches
von 363 bis 1071 nach griechischen, arabischen, syrischen und armenischen Quellen. Brussels, 1935.

659/754

. Review of Le Patriarcat d'Antioche depuis la

paix de l'glise jusqu'a la conqute arabe, by R.
Devreesse. Traditio 5 (1947): 13561.
Horak, U. "Ein amulett aus dem 7. Jahrhundert mit der
Darstellung des Hl. Sergius." In Syrien: Von den
Aposteln zu den Khalifen, ed. E. M. Ruprechtsberger,
19495. Mainz, 1993.
Howard-Johnston, J. "The Two Great Powers in Late
Antiquity: A Comparison." In The Byzantine and Early
Islamic Near East. Vol. 3: States, Resources and
Armies, ed. A. Cameron, 157226. Princeton, N.J.,
1995.
Hunt, L.-A. "A Woman's Prayer to St Sergius in Latin
Syria." BMGS 15 (1991): 96145.
Hunter, E. C. D. "The Cult of Saints in Syria during the
Fifth Century A.D." Studia Patristica 25 (1993):
30812.
. "Syriac Inscriptions from a Melkite Monastery
on the Middle Euphrates." BSOAS 52 (1989): 117.

660/754

Isaac, B. "The Army in the Late Roman East: The Persian Wars and the Defence of the Byzantine
Provinces." In The Byzantine and Early Islamic Near
East. Vol. 3: States, Resources and Armies, ed. A.
Cameron, 12555. Princeton, N.J., 1995.
. The Limits of Empire: The Roman Army in the
East. Rev. ed. Oxford, 1992.
. "An Open Frontier." In Frontires d'empire:
Nature et signification des frontires romaines, ed. P.
Brun et al., 10514. Nemours, 1993.
Jabbur, J. S. The Bedouins and the Desert: Aspects
of Nomadic Life in the Arab East. Translated by L.
Conrad. Albany, N.Y., 1995.

Page 207
Janin, R. La Gographie ecclsiastique de l'empire
byzantin, 1: Le Sige de Constantinople et le patriarcat oecumenique, 3: Les glises et les monastres.
2nd ed. Paris, 1969.
, C. "Das christliche Element in der topographischen Nomenklatur der Balkanlnder." Sitzungsberichte der kaiserlichen Akademie der Wissenschaften 136 (1897): 198.
Johann Georg, Herzog zu Sachsen. "Sadad, Karjaten
und Hawarim." OC 2 (1927): 23342.
Jones, A. Early Arabic Poetry. Reading, 199296.
Jones, A. H. M. The Cities of the Eastern Roman Empire. 2d ed. Oxford, 1971.
. The Later Roman Empire, 284602. Oxford,
1964.
. The Roman Economy. Oxford, 1974.
Kalla, G. "Das ltere Mosaik des byzantinischen
Klosters in Tall ." MDOG 123 (1991): 3539.

662/754

Karnapp, W. "Der Khan in der syrischen Ruinenstadt

Resafa." AA 1978: 13650.
. Die Stadtmauer von Resafa in Syrien. Berlin,
1976.
. "Der Vorplatz vom Zentralbau in Resafa mit
den beiden Torbauten." In Studien zur sptantiken
und byzantinischen Kunst Friedrich Wilhelm Deichmann gewidmet, ed. O. Feld and U. Peschlow, 1:
12532. Bonn, 1986.
Kaufhold, H. "Notizen ber das Moseskloster bei
Nabk und das Julianskloster bei Qarytain in Syrien."
OC 79 (1995): 48119.
Kazhdan, A. "Post-hoc of two Byzantine miracles."
Byzantion 52 (1982): 42022.
Kellner-Heinkele, B.
in den arabischen Quellen."
In D. Sack, Resafa, vol. 4: Die Grosse Moschee von
, 13354. Mainz, 1996.
Kennedy, H. The Prophet and the Age of the Caliphates: The Islamic Near East from the Sixth to the
Eleventh Century. Harlow, Essex, 1986.

663/754

Khoury, R. G. "Die arabischen Inschriften." In T. Ulbert, Resafa, vol. 2: Die Basilika des Heiligen Kreuzes
in Resafa-Sergiupolis, 17980. Mainz, 1986.
Kiilerich, B. Late Fourth Century Classicism in the
Plastic Arts: Studies in the So-called Theodosian
Renaissance. Odense, Denmark, 1993.
King, G. R. D. "Islam, Iconoclasm, and the Declaration
of Doctrine." BSOAS 48 (1985): 26777.
. "Some Churches of the Byzantine Period in
the Jordanian Hawran." DaM 3 (1988): 3575.
. "Two Byzantine Churches in Northern Jordan
and Their Re-Use in the Islamic Period." DaM 1
(1983): 11136.
Kislinger, E., and J. Diethart, "'Hunnisches' auf einem
Wiener Papyrus." Tyche 2 (1987): 510.
Kleinbauer, W. E. "The Origins and Functions of the
Aisled Tetraconch Churches in Syria and Northern
Mesopotamia." DOP 27 (1973): 89114.

664/754

Knowles, D. Great Historical Enterprises: Problems in

Monastic History. London, ca. 1962.
Koder, J., and M. Restle. "Die ra von Sakkaia (Maximianupolis) in Arabia." JB 42 (1992): 7981.
Kolb, F. Diocletian und die erste Tetrarchie. Improvisation oder Experiment in der Organisation monarchischer Herrschaft? Berlin, 1987.
Kollwitz, J. "Die Grabungen in Resafa." In Neue
Deutsche Ausgrabungen im Mittelmeergebiet und im
Vordern Orient, 4570. Berlin, 1959.

Page 208
. "Die Grabungen in Resafa Frhjahr 1959 und
Herbst 1961." AA 1963: 32960.
. "Die Grabungen in Resafa Herbst 1954 u.
Herbst 1956." AA 1957: 64109.
. "Die Grabungen in Resafa Herbst 1952." AAS
45 (195455): 7783.
. "Die Grabungen in Resafa 1952." AA 1954:
11938.
Kollwitz, J., W. Wirth, and W. Karnapp. "Die Grabungen in Resafa: Herbst 1954 und 1956." AAS 89
(195859): 2154.
Konrad, M. "Flavische und sptantike Bebauung unter
der Basilika B von Resafa-Sergiupolis." DaM 6 (1992):
313402.
. "Frhkaiserzeitliche Befestigungen an der
Strata Diocletiana? Neue Kleinfunde des 1. Jahrhunderts n.Chr. aus Nordsyrien." DaM 9 (1996): 16380.

666/754

. Resafa. Vol. 5:Der sptrmische Limes in

Syrien. Archologische Untersuchungen an den Grenzkastellen von Sura, Tetrapyrgium, Cholle, und in
Resafa. Forthcoming.
Koukoules, Ph.

. Athens, 194855.

Kraemer, C. J. Excavations at Nessana. Vol. 3: NonLiterary Papyri. Princeton, N.J., 1958.

Kramer, J. Skulpturen mit Adlerfiguren an Bauten des
5. Jahrhunderts n. Chr. in Konstantinopel. Cologne,
1968.
Krautheimer, R. "Again Saints Sergius and Bacchus
at Constantinople." JB 23 (1974): 25153.
Krautheimer, R., with S.
. Early Christian and Byzantine Architecture. 4th ed. Harmondsworth, Middlesex, 1986.
Krebernik, M. "Schriftfunde aus Tall
123 (1991): 4170.

1990." MDOG

Kugener, M. A. "Note sur l'inscription trilingue de

Zbed." JA 9 (1907): 50924.

667/754

. "Nouvelle note sur l'inscription trilingue de

Zbed." RSO 1 (1907): 57786.
. "Sur l'emploi en Syrie, au VIe sicle de notre
re, du mot 'barbare' dans le sens de 'arabe.'" OC 7
(1907): 40812.
Labourt, J. Le Christianisme dans l'empire Perse sous
la dynastie Sassanide (224632). 2d ed. Paris, 1904.
Lammens, H. "Le Chantre des Omiades: Notes biographiques et littraires sur la pote arabe chrtien
." JA 9/4 (1894): 94176, 193241, 381459.
. La Syrie: Prcis historique. Beirut, 1921.
Lamy, Th.-J. "Profession de foi adresse par les abbs des couvents de la province d'Arabie Jacques
Barade." In Actes du onzime congrs international
des orientalistes, 11737. Paris, 1898.
Lancaster, W., and F. Lancaster. "Thoughts on the
Bedouinisation of Arabia." Proceedings of the Seminar for Arabian Studies 18 (1988): 5162.

668/754

Lassus, J. Sanctuaires chrtiens de Syrie: Essai sur

la gense, la forme et l'usage liturgique des difices
du culte chrtien, en Syrie, du IIIe sicle la conqute
musulmane. Paris, 1947.
Lauffray, J. Halabiyya-Zenobia: Place forte du limes
oriental et la Haute Msopotamie au VIe sicle. Paris,
198391.
Lavin, I. "'The house of the Lord': Aspects of the Role
of Palace Triclinia in the Architecture of Late Antiquity
and the Early Middle Ages." Art Bulletin 44 (1962):
127.
Lee, A. D. Information and Frontiers: Roman Foreign
Relations in Late Antiquity. Cambridge, 1993.

Page 209
LeStrange, G., Palestine under the Moslems: A Description of Syria and the Holy Land from A.D. 650 to
1500. Boston, 1890.
Lewis, N. N. Nomads and Settlers in Syria and
Jordan, 18001980. Cambridge, 1987.
Liebeschuetz, J. H. W. G. Barbarians and Bishops:
Army, Church and State in the Age of Arcadius and
Chrysostom. Oxford, 1990.
Ligt, L. de. Fairs and Markets in the Roman Empire:
Economic and Social Aspects of Periodic Trade in a
Pre-Industrial Society. Amsterdam, 1993.
Litsas, F. K. "Choricius of Gaza and His Description of
Festivals at Gaza." JB 32 (1982): 42736.
Littmann, E. "Die Gemlde der Sergios Kirche in
Sadad." OC 34 (1930): 28891.
. "Osservazioni sulle iscrizioni di
Zebed." RSO 4 (1911): 19398.

e di

670/754

Logar, N. "Kleinfunde aus dem mittelalterlichen

Wohnkomplex in Resafa." DaM 8 (1995): 26992.
Lucius, E. Die Anfnge des Heiligenkults in der christlichen Kirche. Tbingen, 1904.
Luther, A. Die syrische Chronik des Josua Stylites.
Berlin, 1997.
Macdonald, M. C. A. "Nomads and the
in the
Late Hellenistic and Roman Periods: A Reassessment
of the Epigraphic Evidence." Syria 70 (1993):
303413.
. "The Seasons and Transhumance." JRAS 2
(1992): 111.
Mackensen, M. "Eine limeszeitliche Scharnierarmfibel
aus Resafa, Nordsyrien." Germania 61 (1983):
56578.
MacMullen, R. Christianity and Paganism in the
Fourth to Eighth Centuries. New Haven, Conn., 1997.
Maffei, F. de'. "Fortificazioni di Giustiniano sul limes
orientale: monumenti e fonti." In The Seventeenth

671/754

International Byzantine Congress: Major Papers. New

York, 1986.
Magoulias, H. J. "The Lives of the Saints in the Sixth
and Seventh Centuries as Sources for the Internal
and External Enemies of the Byzantine Empire."
48 (199093): 281316.
Maguire, H. Earth and Ocean: The Terrestrial World in
Early Byzantine Art. University Park, Pa., 1987.
Majcherek, G., and A. Taha. "A Selection of Roman
and Byzantine Pottery from Umm el Tlel (Syria)."
Cahiers de l'Euphrate 7 (1993): 10717.
Majeska, G. P. Russian Travelers to Constantinople in
the Fourteenth and Fifteenth Centuries. Washington,
D.C., 1984.
Mango, C. "L'Attitude byzantine l'gard des
antiquits grco-romaines." In Byzance et les images,
ed. A. Guillou and J. Durand, 95120. Paris, 1994.
. Byzantine Architecture. London, 1979.

672/754

. "The Church of Saints Sergius and Bacchus

at Constantinople and the Alleged Tradition of Octagonal Palatine Churches." JB 21 (1972): 18993.
. "The Church of Saints Sergius and Bacchus
Once Again." BZ 68 (1975): 38592.
. "Constantine's Mausoleum and the Translation of Relics." BZ 83 (1990): 5161, with addendum
434.
. "Deux tudes sur Byzance et la Perse sassanide." T&MByz 9 (1985): 91118.
. "Discontinuity with the Classical Past in Byzantium." In Byzantium and the Classical Tradition, ed.
M. Mullett and R. Scott, 4857. Birmingham, 1981.

Page 210
, Z. U. "The 'praetorium' at Musmiya, Again." DOP
44 (1990): 4146.
Maraval, P. Lieux saints et plerinages d'Orient: Histoire et gographie des origines la conqute arabe.
Paris, 1985.
Marcos, N. F. Los "thaumata" de Sofronio: Contribucion al estudio de la "incubatio" cristiana. Madrid,
1975.
Marcus, R. "The Armenian Life of Marutha of
Maipherkat." HThR 25 (1932): 4771.
Marx, E. "The Ecology and Politics of Nomadic
Pastoralists in the Middle East." In The Nomadic Alternative: Modes and Models of Interaction in the
African-Asian Deserts and Steppes, ed. W.
Weissleder, 4174. The Hague, 1978.
Mathews, T. F. "Architecture et liturgie dans les
premires glises palatiales de Constantinople."
Revue de l'art 24 (1974): 2229.

674/754

Matthews, J. "The Tax Law of Palmyra: Evidence for

Economic History in a City of the Roman East." JRS
74 (1984): 15780.
Meimaris, Y., with K. Kritikakou and P. Bougia. Chronological Systems in Roman-Byzantine Palestine and
Arabia: The Evidence of the Dated Greek Inscriptions.
Athens, 1992.
Meischner, J. "Der Hochzeitskameo des Honorius."
AA 1993: 61319.
Mercati, S. G. "Epigraphica." Rendiconti: Atti della
pontificia accademia romana di archeologia 3
(192425): 191214.
Millar, F. The Emperor in the Roman World, 31
B.C.A.D. 337. Corrected reprint. Ithaca, N.Y., 1992.
. The Roman Near East, 31 B.C.A.D. 337.
Cambridge, Mass., 1993.
Miller, E. Mlanges de philologie et d'pigraphie. Paris, 1876.

675/754

Miller, K. Itineraria romana: Rmische Reisewege an

der hand der Tabula Peutingeriana. Stuttgart, 1916.
Millingen, A. van. Byzantine Churches in Constantinople: Their History and Architecture. London,
1912.
Mingana, A. "The Early Spread of Christianity in Central Asia and the Far East: A New Document." BRL 9
(1925): 297371.
Mirzoyan, A. "Les Reprsentations des saints militaires dans le manuscrit No. 6305 Matenadaran M.
." REArm 20 (198687): 44155.
Mitchell, S. Anatolia: Land, Men and Gods in Asia
Minor. Oxford, 1993.
. "Maximus and the Christians in A.D. 312."
JRS 78 (1988): 10524.
Mondsert, C. "Inscriptions et objets chrtiens de
Syrie et de Palestine." Syria 37 (1960): 11630.

676/754

Montgomery, J. E. "The Deserted Encampment in Ancient Arabic Poetry: A Nexus of Topical Comparisons." JSS 40 (1995): 283316.
Morony, M. Iraq after the Muslim Conquest. Princeton,
N.J., 1984.
Mouterde, R. "Dieux cavaliers de la rgion d'Alep:
Collections Guillaume Poche et Franois Marcopoli."
MUSJ 11 (1926): 30722.
. Review of Topographie historique de la Syrie
antique et mdivale, by R. Dussaud. MUSJ 12
(1927): 274.
Mouterde, R., and A. Poidebard. Le Limes de Chalcis:
Organisation de la steppe en haute Syrie romaine.
Paris, 1945.
. See also A. Poidebard.

Page 211
Moyses

(Monachos
Agioreites).
. Thessaloniki, 1989.

Mller, A. "Die Strafjustiz im rmischen Heere." NJKA

17 (1906) 55077
Mundell, M. "Monophysite Church Decoration." In
Iconoclasm: Papers Given at the Ninth Spring Symposium of Byzantine Studies, ed. A. Bryer and J. Herrin, 5974. Birmingham, 1977.
Mundell Mango, M. Silver from Early Byzantium: The
Kaper Koraon and Related Treasures. Baltimore,
1986.
Musil, A., et al.

. Vienna, 1907.

. The Manners and Customs of the Rwala Bedouins. New York, 1928.
. The Middle Euphrates: A Topographical Itinerary. New York, 1927.
. Northern Arabia. New York, 1928.

678/754

. Palmyrena: A Topographical Itinerary. New

York, 1928.
Naguib, S.-A. "The Martyr as Witness: Coptic and
Copto-Arabic Hagiographies as Mediators of Religious Memory." Numen 41 (1994): 22354.
Nau, F. Les Arabes chrtiens de Msopotamie et de
Syrie du VIIe au VIIIe sicle. Paris, 1933.
, ed. and tr. Un Martyrologe et douze mnologes syriaques. PO 10.1.
Newby, G. D. The Making of the Last Prophet: A Reconstruction of the Earliest Biography of Muhammad.
Columbia, S.C., 1989.
Nicholson, O. "The 'pagan churches' of Maximinus
Daia and Julian the Apostate." JEH 45 (1994): 110.
. "Two Notes on Dara." AJA 89 (1985): 66371.
Nicolle, D. "Arms of the Umayyad Era: Military Technology in a Time of Change." In War and Society in
the Eastern Mediterranean, Seventh through Fifteenth
Centuries, ed. Y. Lev, 9100. Leiden, 1997.

679/754

Nldeke, Th. Die Ghssanischen Frsten aus dem

Hause Gafna's. Berlin, 1887.
. "Zur Topographie und Geschichte des damaszenischen Gebiets und der Haurangegend." ZDMG
29 (1875): 41944.
Nogaret, M. "Quelques problmes archologiques et
topographiques dans la rgion de Maiyafarikin."
REArm 18 (1984): 41133.
Noret, J. "La Vie grecque ancienne de S.
." AB 91 (1973): 77103.

de

Norris, H. T. Islam in the Balkans: Religion and Society between Europe and the Arab World. Columbia,
S.C., 1993.
Oates, D. Studies in the Ancient History of Northern
Iraq. London, 1968.
Okada, Y. "Early Christian Architecture in the Iraqi
South-Western Desert."
12 (1991): 7183.
. "Reconsideration of Plaque-type Crosses
from Ain
near Najaf."
11 (1990): 10312.

680/754

Oliphant, L. "New Discoveries." PEQ 1886: 7381.

Oppenheim, M. F. von. Die Beduinen. Leipzig,
19391968.
Oppenheim, M. von, and H. Lucas. "Griechische und
lateinishe Inschriften aus Syrien, Mesopotamien und
Kleinasien." BZ 14 (1905): 174.
Orselli, A. M. Santit militare e culto dei santi
nell'impero dei Romani (secoli VIX). Bologna, 1993.
Passios Hagioreitis.
1987.

. Thessaloniki,

Page 212
Palmer, A. "The Messiah and the Mahdi: History
Presented as the Writing on the Wall." In Polyphonia
byzantina: Studies in Honour of Willem J. Aerts, ed.
H. Hokwerda et al., 4584. Groningen, 1993.
. Monk and Mason on the Tigris Frontier: The
Early History of
. Cambridge, 1990.
Palmer, A., with S. Brock and R. Hoyland. The
Seventh Century in the West Syrian Chronicles. Liverpool, 1993.
Paret, R. "Dometianus de Mlitne et la politique religieuse de l'empereur Maurice." REB 15 (1957):
4272.
Pazdernik, C. "'Our most pious consort given us by
God': Dissident Reactions to the Partnership of
Justinian and Theodora, A.D. 525548." CA 13
(1994): 25681.
Peeters, P. "Les Ex-Voto de Khosrau Aparwez Sergiopolis." AB 65 (1947): 556.

682/754

. "Le Martyrologe de Rabban Sliba." AB 27

(1908): 129200.
. L'Oeuvre des Bollandistes. 2d ed. Brussels,
1961.
. "La Passion armnienne de S. Serge le
Stratlate." In Recherches d'histoire et de philologie
orientales, 1: 2536. Brussels, 1951.
. "La Passion de S. Pierre de Capitolias (13
Janvier 715)." AB 57 (1939): 299333.
. "Sainte Golindouch, martyre Perse (13 Juillet
591)." AB 62 (1944): 74125.
. Le Trfonds oriental de l'hagiographie byzantine. Brussels, 1950.
, ed. Bibliotheca hagiographica orientalis. Brussels, 1910.
Peters, F. E. "Romans and Bedouin in Southern
Syria." JNES 37 (1978): 31526.

683/754

Peterson, E.
. Epigraphische, formgeschichtliche
und religionsgeschichtliche Untersuchungen. Gttingen, 1926.
Pigulevskaya, N. V. Byzanz auf den Wegen nach Indien. Berlin, 1969.
Pinoteau, H. de. "Heraldische Untersuchungen zum
Wappenpokal." In T. Ulbert, Resafa, vol. 3: Der Kreuzfahrerzeitliche Silberschatz aus Resafa-Sergiupolis,
7786. Mainz, 1990.
Pognon, H. Inscriptions smitiques de la Syrie, de la
Msopotamie et de la rgion de Mossoul. Paris, 1907.
Poidebard, A. La Trace de Rome dans le dsert de
Syrie: Le Limes de Trajan la conqute arabe.
Recherches ariennes (19251932). Paris, 1934.
Poidebard, A., and R. Mouterde. " propos de Saint
Serge: Aviation et pigraphie." AB 67 (1949): 10916.
Polites, N. G.

4 (191314): 60169.

Pringle, D. The Defence of Byzantine Africa from

Justinian to the Arab Conquest: An Account of the

684/754

Military History and Archaeology of the African

Provinces in the Sixth and Seventh Centuries. Oxford,
1981.
Raby, J., and J. Johns, eds. Bayt al-Maqdis:
Malik's Jerusalem. Oxford, 1992.

al-

Reich, S. tudes sur les villages aramens de l'AntiLiban. Damascus, n.d.

Rets, J. "The Road to Yarmuk: The Arabs and the
Fall of the Roman Power in the Middle East." In
Aspects of Late Antiquity and Early Byzantium, ed. L.
Rydn and J. O. Rosenqvist, 3141. Istanbul, 1993.
Rey-Coquais, J.-P. Arados et sa pre aux poques
grecque, romaine, et byzantine. Paris, 1974.
. "Syrie romaine de Pompe Diocltien." JRS
68 (1978): 4473.

Page 213
Robin, C. "Le Royaume
, dit 'royaume de Kinda',
entre Himyar et Byzance." CRAI 1996: 665702.
Robinson, C. F. "Ibn al-Azraq, His
Early Islam." JRAS 6 (1996): 727.

and

Rmer, C. "Die griechischen Graffiti." In T. Ulbert,

Resafa, vol. 2: Die Basilika des Heiligen Kreuzes in
Resafa-Sergiupolis, 17177. Mainz, 1986.
Roey, A. van, and P. Allen, eds. and trs. Monophysite
Texts of the Sixth Century. Leuven, 1994.
Ronzevalle, P. "Dieu cavalier sur un bas-relief syrien."
CRAI 1904: 812.
Rosen, S. A. "A Roman-period Pastoral Tent Camp in
the Negev, Israel." JFieldArch 20 (1993): 44151.
Rossi, I. B. de, and L. Duchesne, eds. Martyrologium
Hieronymianum. In Acta Sanctorum Novembris, 2, pt.
1: [I195]. Brussels, 1894.
Rostovtzeff, M. I. "The Caravan Gods of Palmyra."
JRS 22 (1932): 10516.

686/754

. "Dura and the Problem of Parthian Art." YCS

5 (1935): 157304.
. "Graffiti Showing Parthian Warriors." In The
Excavations at Dura-Europos: Preliminary Report of
Second Season of Work, October 1928April 1929,
ed. P. V. C. Baur and M. I. Rostovtzeff, 194200. New
Haven, Conn., 1931.
Rothstein, G. Die Dynastie der
: Ein Versuch zur arabisch-persischen Geschichte zur Zeit der
Sasaniden. Berlin, 1899.
Rouech, C. "Acclamations in the Later Roman Empire: New Evidence from Aphrodisias." JRS 74 (1984):
18199.
Rubin, Z. "Byzantium and Southern Arabia: The Policy
of Anastasius." In The Eastern Frontier of the Roman
Empire, ed. D. H. French and C. S. Lightfoot,
383420. Oxford, 1989.
. "Diplomacy and War in the Relations between
Byzantium and the Sassanids in the Fifth Century
A.D." In The Defence of the Roman and Byzantine

687/754

East, ed. P. Freeman and D. Kennedy, 67795. Oxford, 1986.

. "The Mediterranean and the Dilemma of the
Roman Empire in Late Antiquity." MHR 1 (1986):
1362.
Ruprechtsberger, E. M., ed. Syrien: Von den Aposteln
zu den Khalifen. Mainz, 1993.
Sack, D. Damaskus: Entwicklung und Struktur einer
orientalisch-islamischen Stadt. Mainz, 1989.
. Resafa. Vol. 4: Die Grosse Moschee von
. Mainz, 1996.
Sako, L. Le Rle de la hirarchie syriaque orientale
dans les rapports diplomatiques entre la Perse et Byzance aux VeVIIe sicles. Paris, 1986,
Sarre, F., and E. Herzfeld. Archologische Reise im
Euphrat und Tigris Gebiet. Berlin, 191120.
Sartre, M. "Les Arabes nomades d'Arabie du nordouest d'Alexandre au haut-empire." In Le Dsert:

688/754

Image et ralit, ed. Y. Christie et al., 13960.

Leuven, 1989.
. Bostra: Des origines l'Islam. Paris, 1985.
. Trois tudes sur l'Arabie romaine et byzantine. Brussels, 1982.
Sauvaget, J. "Les Ghassanides et Sergiopolis." Byzantion 14 (1939): 11530.
. La Mosque omeyyade de Mdine: tude sur
les origines architecturales de la mosque et de la basilique. Paris, 1947.
Savignac, M. R., and M. Abel. "Chronique." RB 2
(1905): 592606.
Schippmann, K. Die iranischen Feuerheiligtmer. Berlin, 1971.

Page 214
Schlumberger, D. La Palmyrne du nord-ouest. Paris,
1951.
Schumacher, G. The Jaulan. London, 1888.
Segal, J. B. Edessa: "The blessed city." Oxford, 1970.
Serjeant, R. B. "
and
, the Sacred Enclosure
in Arabia." In id., Studies in Arabian History and Civilization, vol. 3. London, 1981.
. "Saint Sergius." In id., Studies in Arabian History and Civilization, II. London, 1981.
Serjeant, R. B., and R. Lewcock, eds.
Islamic City. London, 1983.

: An Arabian

Seyrig, H. "Antiquits syriennes: Les Dieux arms et

les Arabes en Syrie." Syria 47 (1970): 77112.
. "Antiquits syriennes: L'Incorporation de
Palmyre l'empire romaine." Syria 13 (1932):
26677.

690/754

. "Inscriptions grecques." In G. Tchalenko, Villages antiques de la Syrie du Nord: Le Massif du

Blus l'poque romaine, 3: 262. Paris, 1958.
Shahd, I. "The Arabs in the Peace Treaty of A.D.
561." In id. Byzantium and the Semitic Orient before
the Rise of Islam, VII. London, 1988.
. Byzantium and the Arabs in the Fifth Century.
Washington, D.C., 1989.
. Byzantium and the Arabs in the Fourth Century. Washington, D.C., 1984.
. Byzantium and the Arabs in the Sixth Century.
Washington, D.C., 1995.
. "
post
." In The Islamic World from
Classical to Modern Times: Essays in Honor of Bernard Lewis, ed. C. E. Bosworth et al., 32336. Princeton, N.J., 1989.
. The Martyrs of Najrn. New Documents.
Brussels, 1971.

691/754

. "The Patriciate of Arethas." In id., Byzantium

and the Semitic Orient before the Rise of Islam. London, 1988.
. "Procopius and Arethas, Part One." In id. Byzantium and the Semitic Orient before the Rise of
Islam. London, 1988.
. "The Restoration of the Ghassanid Dynasty,
A.D. 587: Dionysius of Tellmahre." ARAM [A Festschrift for Dr. Sebastian P. Brock] 5 (1993): 491503.
Sinclair, T. A. Eastern Turkey: An Architectural and
Archaeological Survey. London, 198790.
. "The Site of Tigranocerta.I." REArm 25
(199495): 183253.
Sotiriou, G., and M. Sotiriou.
Athens, 1952.

Spanner, H., and S. Guyer. Rusafa: Die Wallfahrtsstadt des Heiligen Sergios. Berlin, 1926.

692/754

Speidel, M. P. "Late Roman Military Decorations, I:

Neck- and Wristbands." Antiquit tardive 4 (1996):
23543.
. Riding for Caesar: The Roman Emperor's
Horse Guards. London, 1994.
Spier, J. "Medieval Byzantine Amulets and their Tradition." JWI 56 (1993): 2562.
Stein, E. Histoire du bas-empire. 2d ed. Vol. 1. Paris,
1959. Vol. 2. Paris, 1949.
Stern, H. "Sur quelques pavements palo-chrtiens
du Liban." CArch 15 (1965): 2137.
Stetkeyvych, J. The Zephyrs of Najd: The Poetics of
Nostalgia in the Classical Arabic
. Chicago, Il.,
1993.
Strommenger, E. "Ausgrabungen in Tall
MDOG 123 (1991): 734.
. "Die Ausgrabungen in Tall
125 (1993): 511.

1990."

1992." MDOG

693/754

. "Die Ausgrabungen in Tall

126 (1994): 1131.

1993." MDOG

Page 215
Syme, R. Anatolica: Studies in Strabo. Edited by A.
Birley. Oxford, 1995.
. "Tigranocerta: A Problem Misconceived." In
Armies and Frontiers in Roman and Byzantine Anatolia, ed. S. Mitchell, 6170. Oxford, 1983.
Taha, A. "Prospection du site romaine tardif de Juwal
(cuvette d'el Kowm, Syrie)." Cahiers de l'Euphrate 56
(1991): 6166.
Tardieu, M. Les Paysages reliques: Routes et haltes
syriennes d'Isidore Simplicius. Louvain, 1990.
Tate, G. Les Campagnes de la Syrie du Nord du IIe
au VIIe sicle: Un Exemple d'expansion dmographique et conomique la fin de l'antiquit. Paris,
1992.
Tchalenko, G. glises syriennes bma. Paris, 1990.
. Villages antiques de la Syrie du Nord: Le
Massif du Blus l'poque romaine. Paris, 195358.

695/754

Teixidor, J. "Deux inscriptions palmyrniennes du

Muse de Bagdad." Syria 40 (1963): 3346.
. "La Palmyrne orientale: Frontire militaire ou
zone douanire?" In Frontires d'empire: Nature et
signification des frontires romaines, ed. P. Brun et
al., 95103. Nemours, 1993.
Thierry, J. M. Monuments armniens du Vaspurakan.
Paris, 1989.
Tillemont, L. S. Le Nain de. Mmoires pour servir a
l'histoire ecclsiastique des six premiers sicles.
Venice, 1732. Revision of the second Paris edition.
Trimingham, J. S. Christianity among the Arabs in
Pre-Islamic Times. London, 1979.
Trombley, F. R. Hellenic Religion and Christianization,
c. 370529. Leiden, 199394.
. "War and Society in Rural Syria, c. 502613
A.D.: Observations on Epigraphy." BMGS 21 (1997):
154209.

696/754

Ulbert, T. "Beobachtungen im Westhofbereichen der

Groen Basilika von Resafa." DaM (1993): 40316.
. "Bischof und Kathedrale (4.7. Jh): Archologische Zeugnisse in Syrien." In Actes du XIe congrs
international d'archologie chrtienne, ed. N. Duval, 1:
42957. Rome, 1989.
. "Eine neuentdeckte Inschrift aus Resafa (Syrien)." AA 1977: 56369.
. Resafa. Vol. 2: Die Basilika des Heiligen
Kreuzes in Resafa-Sergiupolis. Mainz, 1986.
. Resafa. Vol. 3: Der Kreuzfahrerzeitliche Silberschatz aus Resafa-Sergiupolis. Mainz, 1990.
. "Resafa-Sergiupolis 197683." AAS 33.2
(1983): 6982.
.
"Resafa-Sergiupolis:
Archologische
Forschungen in der Nordsyrischen Pilgerstadt." In
Syrien: Von den Aposteln zu den Khalifen, ed. E. M.
Ruprechtsberger, 11227. Mainz, 1993.

697/754

. "The Silver Treasures of Resafa / Sergiupolis." In The Near East in Antiquity, ed. S. Kerner,
10510. Amman, 1990.
. "Ein umaiyadischer Pavillon in
DaM 7 (1993): 21331.

."

. "Villes et fortifications de l'Euphrate l'poque

palo-chrtienne." In Archologie et histoire de la
Syrie, vol. 2, ed. J.-M. Dentzer and W. Orthmann,
28396. Saarbrcken, 1989.
. "Zwei sigmafrmige Mensaplatten aus Syrien." In Pietas: Festschrift fr Bernhard Ktting, ed. E.
Dassmann and K. Suso Frank, 55965. Mnster,
1980.

Page 216
. See also P.-L. Gatier.
Van Dam, R. Saints and Their Miracles in Late
Antique Gaul. Princeton, N.J., 1993.
Velmans, T. "Observations sur quelques peintures
murales en Syrie et Palestine et leur composante byzantine orientale." CArch 42 (1994): 12338.
Vikan, G. "Art, Medicine and Magic in Early Byzantium." DOP 38 (1984): 6586.
. "Two Unpublished Pilgrim Tokens in the
Benaki Museum and the Group to Which They
Belong." In
, 34146. Athens,
1994.
Villeneuve, F. "L'conomie rurale et la vie des campagnes dans le Hauran antique" (1er sicle avant J.C.VIe sicle aprs J.-C.) In Hauran: Recherches
archologiques sur la Syrie du sud a l'poque hellnistique et romaine, ed. J.-M. Dentzer, 63136. Paris,
1985.

699/754

Vog, M. de, and W. H. Waddington. La Syrie central: Architecture civile et religieuse du Ier au VIIe
sicle. Paris, 1865.
Wagner, E.,
atur der frhen

. Eine Studie zur arabischen Liter. Wiesbaden, 1965.

Walker, J. T. "'Your heroic deeds give us pleasure!':

Culture and Society in the Christian Martyr Legends of
Late Antique Iraq." Ph.D. diss., Princeton, 1998.
Walter, C. "The Intaglio of Solomon in the Benaki Museum and the Origins of the Iconography of Warrior
Saints." DCAE 15 (198990): 3342.
. "The Origins of the Cult of Saint George."
REB 53 (1995): 295326.
. Review of Studies on Constantinople, by C.
Mango. REB 53 (1995): 37476.
. "Some Unpublished Intaglios of Solomon in
the
British
Museum,
London."
In
, 36568. Athens, 1994.

700/754

. "The Thracian Horseman: Ancestor of the

Warrior Saints." ByzF 14 (1989): 1.65773, with 2.pls
24955.
Weber, T. "Karawangtter in der Dekapolis." DaM 8
(1995): 20311.
Weigand, E. "Das sogenannte Praetorium von
Phaena-Mismije." In Wrzburger Festgabe Heinrich
Bulle, 7192. Stuttgart, 1938.
Weiss, H. "Archeology in Syria." AJA 98 (1994):
10158.
Weitzmann, K., ed. Age of Spirituality: Late Antique
and Early Christian Art, Third to Seventh Century.
New York, 1979.
. The Monastery of Saint Catherine at Mount
Sinai: The Icons, vol. 1: From the Sixth to the Tenth
Century. Princeton, N.J., 1976.
Wemhoff, M. "Ein mittelalterlicher Wohnkomplex in
Resafa." DaM 8 (1995): 24768.

701/754

Whitby, M. The Emperor Maurice and His Historian:

Theophylact Simocatta on Persian and Balkan Warfare. Oxford, 1988.
. "Procopius and the Development of Roman
Defences in Upper Mesopotamia." In The Defence of
the Roman and Byzantine East, ed. P. Freeman and
D. Kennedy, 71735. Oxford, 1986.
Whittaker, C. R. Frontiers of the Roman Empire: A
Social and Economic Study. Baltimore, 1994.
Whittow, M. The Making of Orthodox Byzantium,
6001025. London, 1996.
Wiessner, G. "Christlicher Heiligenkult im Umkreis
eines sassanidischen Groknigs." In Festgabe
deutscher Iranisten zur 2500 Jahrfeier Irans, ed. W.
Eilers, 14155. Stuttgart, 1971.

Page 217
. Christliche Kultbauten im
198283.

. Wiesbaden,

. Zur Mrtyrerberlieferung aus der Christenverfolgung Schapurs II. Gttingen, 1967.

Wifstrand, A. "Autokrator, Kaisar, Basileus. Bemerkungen zu den griechischen Benennungen der
rmischen Kaiser." In PAMA Martino P. Nilsson
A.D. IV Id. Iul. anno MCM XXXIX dedicatum, 52939.
N.p., 1939.
Wilkinson, J. "Column capitals in the
." In
Bayt al-Maqdis:
al-Malik's Jerusalem, ed. J. Raby
and J. Johns, 12539. Oxford, 1992.
. Jerusalem Pilgrims before the Crusades.
Warminster, Wilts., 1977.
Will, E. "Damas antique." Syria 71 (1994): 143.
Wirth, E. Syrien: Eine geographische Landskunde.
Darmstadt, 1971.

703/754

Woods, D. "The Emperor Julian and the Passion of

Sergius and Bacchus." JECS 5 (1997): 33567.
Wuthnow, H. Die semitischen Menschennamen in
griechishen Inschriften und Papyri des vorderen Orients. Leipzig, 1930.
Zacos, G., and A. Veglery. Byzantine Lead Seals.
Basel, 1972.
Ziegler, R, "Aigeai, der Asklepioskult, das Kaiserhaus
der Decier und das Christentum." Tyche 9 (1994):
187212.

Page 218
This page intentionally left blank.

Page 219

INDEX
dynasty, 183
Abbibas, 100
al-Malik, Umayyad caliph, 17778, 181
al-Rahman I, amir of al-Andalus, 175
ibn

, poet, 126

al-Abjaz, Jabal, 7273

Abraham, bishop of Rusafa, 83, 139n29; tomb of (?),
91
Abu Jafna

ibn al-Mundhir, 135, 173, 173n152

Abu Nuwas, 127n143

Abu Rujmayn, Jabal, 2, 7071, 72n66, 7374, 98
al-Basri, poet, 127n143
Acacius of Amida, bishop, 53

706/754

Adam, 154, 16869

Adarmaanes, Iranian commander, 63
ibn Zayd, poet and diplomat, 126, 165
Adiabene, 40
Ahudemmeh, bishop of Beth
149, 162, 165

, 12128, 141, 144,

, 114n76
Aleppo, 182, 184. See Beroea
Alexander, bishop of Hierapolis / Manbij, 7, 2628, 78,
80, 93, 95, 118, 159
Alexander, governor of Edessa, 167, 169
Alexander Akoimetes, saint, 19, 74
, 110
Ambar, 63
ambo, 88, 92n146, 120n112

707/754

Ambrus, 134
Ameras, 109, 109n41
Amida, 97n177, 101, 156
Ammianus Marcellinus, historian, 97
, 101, 107
ibn

, poet, 174

ibn Matta, ecclesiastical historian, on Marutha of

Mayperqat, 5152
amulet with Sergius image, 38, 42
, 127n143
Anasartha / Khunasira, 73, 112n63, 16263
Anastasius, patriarch of Antioch, 140
Anastasius, Roman emperor: frontier politics of,
6365, 65n25, 132; and Rusafa, 65, 77, 9294,
12829, 182; and S. Sergius, 65, 112, 132

708/754

Anastasius the Persian, saint, 67, 96, 121

Andarin. See Androna
Andrew, apostle, 45
Androna / Andarin, 88n130, 153
animals: and holy sites, 1011, 9899; as offerings,
8687; as threat, 101, 121
Anti-Lebanon mountains, 2, 70, 104, 114, 189
Antioch, 1213, 1516, 18, 47n10, 63, 70, 73, 102,
114, 136, 139
Antiochus, dux of Augusta Euphratensis, 10, 15, 157
Antonius, martyr, 96
Apamea, 63, 73, 89, 114
, 70
, 120, 125

Page 220
Arab Jacobite synaxarion, 8586, 90n138
Arabs: in frontier politics, 50, 60, 60n3, 6168, 128,
13436, 14173, 181; at Mamre, 97; and martyrium of
John, 163; and martyrium of Thomas, 163; as pastoralists, 23, 43, 60, 60n3, 61, 66, 72, 76, 14449, 163,
165, 169, 170; and Sergius cult, 96, 98, 11719,
12129, 133, 169, 17882, 18991; and S. Symeon,
100, 148. See also Ghassanids; Jafnid dynasty; Lakhmids; pastoralism
Arak, 70, 77n86, 78
Arbela, 120
Arethas, 64. See al-Harith
Armenia, between Rome and Iran, 49, 49n17, 50,
5556
Arsamosata, 97n177
Artaxata, 68
Asorestan, Armenian name for Beth

, 55

710/754

Athanasius of Clysma, saint, 1718

Augusta Euphratensis, 10, 13, 1618
, 125
, 124, 128
Azarethes, Iranian commander, 63
, god, 4042, 99n183
. See Zorava /
Babai the Great, hagiographer, 140
Babisqa, dedication
11617n91, 179n19

of

Solomonidas,

11617,

Bacchus, saint: companion of Sergius 24, 2829,

31n80, 43, 129, 132; dedications to, 90, 104, 1089,
11112, 11416, 120, 131, 131n4, 132n6; feast day
of, 22n54; images of, as rider, 188, 190; images of, as
standing soldier, 29; relics of, 90n140, 91, 189; tomb
of, 60, 90
badiya, 68, 72. See also steppe

711/754

Bahram II Chobin, 14n18, 13637

Balad, 12122, 124
Balbion, clarissimus, 14748
Balikh, river, 70
Balis. See Barbalissus / Balis
baptism, nomad, 96, 122, 148, 148n69
Barbalissus / Balis, 10, 13, 19, 21, 60, 69, 90, 112n63
Barbarian Plain, 1, 5, 37, 65, 65nn2628, 66, 76, 120,
137, 191
Bartholomew, saint, relics of, 47, 65
Basil, bishop of Caesarea, 97
Batnae, 97
bema, in basilica A at Rusafa, 8182, 86n119,
92n146
Beroea / Aleppo, 72n66, 75n79, 117, 118

712/754

Bertramn, bishop of Bordeaux, 129

Beth, in compound place names, 61, 61n6
Beth

, 5556, 12122, 124, 127

Beth Batin, synod of, 92

Beth Garmai, 102
Beth Lashpar, 140
Beth Rsapha, 124
Bethsaloe, 121, 121n115
Betproclis / Furqlus, 73, 74n74
, Jabal, 70, 74
al-Biruni, 167
Bishri, Jabal, 2, 69, 7172, 72n66, 98
Bithrapsa, 61
Black Sea, 70

713/754

Bordeaux, finger of S. Sergius at, 129

Bostra / Busra, 73, 75, 90, 104, 10612, 14546, 163
brothers, 9n3
burial, ad sanctos, 158
Burj: near Dumayr, 4; near Raphanaea, 4, 114
Busra. See Bostra / Busra
Busr al-Hariri, 104, 111
, 143
Calama / Guelma, inscription of Solomon of Dara, 47
Callinicum, 62, 68, 70, 128
camelarius, 36, 37, 76
camel meat, 123, 143, 143n44
Candidus, bishop of Rusafa, 13334

714/754

Chalcedonians, 3, 92, 112, 136, 13941, 143, 171,

187; cohabitation at holy sites of and non-Chalcedonians, 15657
Chalcis / Qinnasrin, 4, 2223, 26, 29, 60, 63, 73, 75,
1045, 112, 163
chiton of SS. Sergius and Bacchus, 29, 32, 38
chlamys of SS. Sergius and Bacchus, 29, 3132, 91,
189
Choricius of Gaza, 9697
Christian and Muslim symbiosis, 17783, 179n17,
18083, 18991
Circesium, 63, 65, 70, 128, 133
Cisseron, Mount, 132
Constantinople: Chrysotriclinos at, 169; church of S.
Polyeuctus at, 48; church of

Page 221
S. Sergius at, 130, 13233, 168, 189; decoration in
Studite monastery at, 155n84; defended by relics,
4546, 48; images of S. Sergius originating from, 29,
43; al-Mundhir at, 172; translation of S. Sergius's
thumb to, 65, 92; view of Syria-Mesopotamia from, 49,
141
Constantia /

, 103

Conversion of church to mosque: near Aleppo, 182; at

Damascus, 17779; at Umm al-Surab, 109
Conversion of temple to church: at Damascus,
177n13; at Mismiya, 155
Cross, Holy: Arab devotion to, 180; cult of, 102; at
Rusafa, 8283, 139n29; as votive offering, 132, 134,
137, 141
cross-dressing, forcible, as insult, 9, 14, 14nn1820,
18, 2324, 24n60
Ctesiphon, 70, 140
Cyprus, silver bowl from, 32, 43

716/754

Cyril of Scythopolis, hagiographer, 164

Cyrus and John, saints, 157
Damascus: Christian and Muslim architecture at,
17778, 181; cult of saints at, 1056, 111, 143,
164n120; geopolitical position of, 66, 70, 7375,
14647; and medieval cult of S. Sergius, 187, 189,
190; persecution of Christians at, 15
Damian, patriarch of Alexandria, 144
Dara, 47, 6465, 94n155, 1012, 119, 135, 16566
, 75, 10910
Dar Qita, 116
Dastagird, 67, 121
Dayr

, 75

Dayr al-Nasrani, 108

Dayr al-Qadi, 108
Dayr al-Zafaran, 78

717/754

Dayr al-Zawr, 6970

Demetrius, saint, 89n132
desert, 2, 3, 72, 98
Diocletian, Roman emperor, 13, 18, 19n42, 20,
20n48, 21, 76
Dionysias. See Suwayda
dolines, 77, 181
Domitianus, bishop of Melitene and Iranian martyrs at
Mayperqat, 58, 136, 16566, 171
Dumayr, 4
Dur, 111
eagle, 154, 154n81, 155, 168
Edessa / Urfa, 27, 104, 118n99, 119n104, 135,
16667, 171; architectural decoration at, 78; cults of
god
and of S. Sergius at, 41
effeminacy and male saints, 3233, 35

718/754

Ehnesh, 11819
Eitha / Hit, Sergius inscription redated to sixth century,
1057, 107n31
, 108
Emesa / Homs, 4, 70, 7374, 102, 11214, 60n3
Ephrem, patriarch of Antioch, 143
Ephrem the Syrian, saint, 47n10, 80n97
Epiphania / Hama, 73, 113, 115
, 108, 113
Euphrates, river, 1, 7, 10, 13, 13n10, 17, 63, 6566,
6970, 73, 78, 97, 118, 12425, 127n145, 133, 158,
160, 175, 184
Euthymius, saint, 16465, 169n143, 171
Evagrius, historian, 13436, 139
feasts of SS. Sergius and Bacchus, 16, 2223, 25, 28,
9697

719/754

feet, pierced, 10, 23, 25, 110

fire temple, 167
Flavius

, illustrius ordinarius, 147

Forty Martyrs of Sebaste, relics of, 48

Furqlus. See Betproclis / Furqlus
Gaboulon, 63
Gadara / Umm Qays, 146
Galerius Maximianus, Roman emperor, 1213
Gaza, church of S. Sergius, 32n83, 97n173
George, saint, 4, 101, 1078, 108n38, 110, 114,
118nn99100, 179 (Jarjis), 179n19, 188, 190. See
also Giwargis
George of Izla, 121
Gerasa / Jarash, 108
Ghasm, 109

720/754

Ghassanids: and frontier defence, 64, 64nn2122, 66,

133, 14149; legacy of, 17475, 175n4; and alMundhir building outside Rusafa, 15152, 154n81,
15657, 16164, 169, 171
al-Ghina, 154
Ghur, 4, 114, 114n79
Ghuta, oasis, 105, 146
Giwargis, Iranian martyr, 140

Page 222
God of Sergius, 108, 129
Golan. See Jawlan
Gregory, bishop of Nyssa, 96
Gregory, bishop of Tours: on chickens at Rusafa,
8687; on martyr cult, 85n118, 129; on public readings of martyr acts, 44n115; on saints as defenders,
46
Gregory, patriarch of Antioch, 136, 139, 141, 166
Gulanducht, Iranian martyr, 65, 90, 120, 135, 135n19,
136, 156
Haditha, 61
Hama, 184, 188. See also Epiphania / Hama
haram, 99, 17071
al-Harith, Ghassanid leader, 133, 14344
Harran, in Laja, 163

722/754

Harran, in Mesopotamia, 60, 92

al-Harawi,

ascetic, 182

Hassan ibn Thabit, 14546

Hatra, 67n36
Hawran: association with Ghassanids, 14446, 148,
157, 171; cults of god
and of S. Sergius in, 41;
definition of, 74n78; routes relating to, 73, 7576; Sergius cult in, 75, 102, 102n5, 103, 10412; S. Sergius
as rider in 39, 11011; toponymy of, 7475
Hawran, Jabal, 74n78, 75, 106, 108, 169
Hayat, 107n32
healing, 11, 38, 4243, 68, 8586, 89, 190
Hebron, 97
Heraclius, Roman emperor, 102
Hiba, bishop of Edessa, 119
Hierapolis / Manbij, 7, 96, 135

723/754

Hind, Lakhmid princess, 126

Hira, Lakhmid capital, 62, 11922, 124n130, 126,
162, 171;
Banu Mazin at, 162
Hisham, Umayyad caliph, 72, 126n140, 175, 178,
18183
Hit: on Euphrates, 125; in Hawran. See Eitha / Hit
"holy rider," 4243
Homs, 184, 187. See Emesa / Homs
Hujrid dynasty, 64, 64nn2122
Hulwan, 140
Huwirta (Huarte), 89n131, 89n135, 15455, 16869
Huwwarin, 172
Ibn al-Azraq, historian, on Marutha of Mayperqat, 51,
55n46
Ibn Butlan, physician, 68, 183

724/754

Ibn Ishaq, biographer of the Prophet Muhammad,

103, 180
Ibn Shaddad, historian, 184
al-Idrisi, geographer, 18384
, 108
, 118
al-Qays in Zabad inscription, 117
Imtan, 106
Invention of the Relics of Saint Bartholomew on
Marutha of Mayperqat's relic-collecting, 55n46
Iranian martyrs: acta of, 57; hymns on, 56n47, 58
Ishoyahb of Arzun, catholicos, 140
, monastery of, 164
Isriya. See Seriane / Isriya
Jabal. See under proper names

725/754

Jabiya, 14344, 146, 169, 174

Jacob Baradaeus, 121n116, 133
Jacob of Sarug, vi, 2526, 112, 128
Jafna, phylarch at Jabiya in 587, 144
Jafnid dynasty, 64, 64n22, 142, 14445. See also
Ghassanids
jahiliya, 103, 174
Jarir, poet, 180, 188
Jarjis. See George
Jasim, 14547
Jawlan (Golan), 14344, 14648, 154n81
Jazira, 121, 127n145
Jerusalem: capitals in al-Aqsa mosque, 154n81;
Church of the Resurrection (Holy Sepulcher) at,
89n136, 17778; Dome of the Rock at, 17778; Eudocia and, 26; monastery of Anastasius the Persian

726/754

near, 67; on pilgrim itineraries, 113, 120, 135, 156;

Pulcheria, Theodosius and, 4748, 132; and the True
Cross, 141
Jesus, 177; Sergius as double for, 180, 180n23
Jews, 3, 24, 97, 108n36, 129, 143n44, 180, 183
Jilliq, 143
Jiza, 110
John, apostle, 98
John, monk of Judaea, 98
John, patriarch of Antioch, 2627, 93

Page 223
John the Baptist, 14748, 163; church of, at Damascus, 17778
John the Baptist and Addai the Apostle, martyrium of,
at Edessa, 167
John of Ephesus, historian, 156
John of Epiphania, historian, 63, 136
John Moschus, monastic writer, 98, 100
John of Rusafa, vir illustris, 102, 135, 166
Jordan, river, 74n78
Joseph, bishop of Rasiphta, 80n97
Julia, saint, 90n138
Julian, archibishop of Bostra, 11112
Julian, bishop of Salaminias, 113
Julian, Roman emperor, 1215, 147

728/754

Julian, saint, 4
Justin, Roman emperor, 83, 87, 94, 132, 156
Justinian, Roman emperor: and S. Leontius at
Damascus, 111; and Mayperqat 58; and Rusafa,
9394, 141, 182; and S. Sergius, 12933, 137, 139,
168
Kafr, 1078
Kafr Antin, 118, 148n66
Kaper Koraon, 29, 43, 114, 114n81; lamp-stand dedicated to S. Sergius alone, 11516
Kasr-i Shirin, 140
Kawad I, Sasanian king, frontier politics of, 51, 57,
6264, 66, 119
al-Kawm, 70, 77n86
Khabur, river, 70, 121
Khidr, 179, 190

729/754

Khirbat al-Mafjar, 181

Khulla, 20
Khunasira. See Anasartha / Khunasira
Khusrau I, Sasanian king, 63, 95n165, 102, 114, 128,
133, 141
Khusrau II, Sasanian king: 121, 165; at Dara, 16567,
171; at Edessa, 102; at Mayperqat, 5859; at Rusafa,
13439, 139n29, 14041, 182; and S. Sergius, 121,
12829
Khirbat al-Bayda, 175n4
Kitab al-aghani, xii, 175
Kufa, 181
Labid ibn

, 170

Laja, 75, 104, 106, 110, 111, 155, 163

Lakhmids, 61, 6364, 64n22, 66, 124n130, 142
Laterculus Veronensis, 1617

730/754

Leontius, saint, 90, 96, 104, 111

"Letter of the Archimandrites," 110n51, 143n45, 144,
146, 164
"Life of Justinian," 132, 132n6
Longinus, centurion and saint, 4, 114
Mabrakta, 121
, 108
, 18990
Mamluk dynasty, 184
Mamre, Oak of, 97
Manbij. See Hierapolis
maniakion, 9, 29, 31, 31n81, 32, 43. See also torque
Mar

, 121

Mar Musa al-Habashi, 187

731/754

Marcellinus, martyr, 96
Marcianus, MVM per orientem, 172
Mardin, 78
Mari ibn Sulayman, ecclesiastical historian, on
Marutha of Mayperqat, 51, 54
Maronius, chorepiscopus, 84, 87
Marousa, 100
martyr cult: architecture of, 8486, 88, 88n130,
8991, 96, 14748; early attestation of at Eitha / Hit,
revised, 1057
, 108
Martyropolis. See Mayperqat (Martyropolis) / Silvan
Maruta, metropolitan of Takrit, 12528, 144, 149, 165
Marutha, bishop of Mayperqat: cultivation of martyr
cult on the frontier by, 44, 48, 5459, 141; diplomatic
career of, 5253; vitae of, 5052, 5556

732/754

Marwan II, Umayyad caliph, 40n100

mashhad of S. Sergius at Rusafa, 182
, historian, 89n131
al-Matyaha, plain of, 118
Maurice, Roman emperor, 77, 128, 134, 136, 14041,
17273
Mauricius of Apamea, martyr, 96
Mavia, 163
Mawsil (Mosul), 120
Maximianopolis / Shaqqa, 75, 101, 1057
Maximinus Daia, Roman emperor, 12, 1516
Mayperqat (Martyropolis) / Silvan: establishment of
martyr cult at, 44, 5557; geopolitical setting of,
4850, 5759

Page 224
Mecca, 99n183, 127n143, 177
Medina, 168, 177
Megas, honorary consul, 115
Melitene, 189
, 108
Menuthis, 90n136, 157
metaton, 4, 11314
Michael, archangel, 4, 114
Michael of Crete, icon painter, 190
Michael the Syrian, historian, 90, 143
miracles, 8, 1011, 23, 27, 38, 68, 86, 188
Mismiya, 155
monasteries, 61n4, 105, 108, 110, 121, 123,
123n125, 12427, 127n143, 128, 136, 14041, 144,

734/754

146, 146n57, 14849, 156, 16465, 171, 177, 179,

181, 18485, 187, 189, 190. See also monks
Mongols, 18485
monks, 1, 3, 9n3, 10, 19, 19n45, 20, 20n49, 28, 49,
67, 74, 80n97, 18485, 127. See also monasteries
"monophysites." See non-Chalcedonians
Chalcedonians / "monophysites"

anti-

mosque: at Damascus, 181; in Hawran, 109; at Jerusalem, 154n81; at Khirbat al-Mafjar, 181; at Kufa,
62n114, 181; as place of judgement, 168. See also
Rusafa / Sergiopolis, architecture of
Mouchasos, cameleer in Hawran, 76
, Umayyad caliph, 168
Muhammad, prophet, 103, 145, 170, 188; companions of, 182
al-Mundhir, Ghassanid leader, relations with Rome,
17173. See also al-Mundhir, building of, at Rusafa

735/754

al-Mundhir, building of, at Rusafa, 14973; in context

of tribal churches, 16268; crosses, 151; eagle decoration, 151, 154, 154n81, 155; inscription, 151, 155,
15961, 164, 172; location in cemetery, 15759; marine motif, 151, 154; plan, 15253; relation to north
gate, 158; and site of Sergius's tomb, 15859, 171;
surrounded by encampment, 16970; and uses / abuses of churches and holy sites, 16268
al-Mundhir III, Lakhmid leader, 6263, 64n22, 87n127
Muqaddasi, geographer, 178
Mushannaf, 75, 107
name: of saint, as protection, 47, 101; of Sergius, reverence for, 19n42, 39, 124, 12829. See also Sergius,
use of name
Nabak, 187
, 108, 110, 105, 111
Nawa, 113
Nestorian, 3, 12021, 123, 128, 140, 157, 189

736/754

Nicomedia, 13, 15
Nisibis, 64, 68, 47n10, 119, 119n104, 12021,
128n148, 135, 137
non-Chalcedonians / anti-Chalcedonians / "monophysites," 3, 23, 80n97, 92, 110, 112, 12022,
123n126, 125, 127, 13233, 136, 14041, 14344,
164, 171. See also Chalcedonians
Notitia dignitatum, 21, 73, 74n74
II, Lakhmid leader, 6162, 64n22
III, Lakhmid leader, 126, 136
ibn al-Harith, 95n166
ibn al-Mundhir, Ghassanid leader, 173
Occariba /

, 73, 74n74

, 108, 111, 132

oil, blessed, 85, 85n118, 86, 86n119, 89, 89n31, 90
Orthosias, 114, 114n77

737/754

Palmyra, 17, 6667, 69n44, 70, 7477, 77n86, 78,

98n180, 184
Passio of SS. Sergius and Bacchus: anachronism in,
1820; animals in, 1011, 9899; and Constantinople
inscription, 131; emperor in, 1117; miracles of S.
Sergius in, 11, 25, 188; and al-Mundhir building,
15758; narrative of, 811; redactions of, 8n1; related
texts, 1718, 2226; and Zorava inscription, 11011
pastoralism: and buildings, 146, 146n56, 149, 163,
163n118, 16970, 179; and Christian cult, 23, 86,
9899, 1045, 14649, 163, 165, 169, 181; in frontier
zone, 2, 3, 60, 60n3, 61, 66, 67, 7273, 7576,
12128, 14249, 163, 163n118. See also Arabs;
badiya; Rusafa / Sergiopolis, topography and relation
to routes; steppe
Paulinus, bishop of Nola, 45

Page 225
Persian Gulf, 69
Peter of Callinicum, patriarch of Antioch, 144
Peter the Iberian, bishop of Maiuma, 114
Philippopolis / Shahba, 106
phoenix, 154n83
phrourarch, 11314
Piacenza Pilgrim, 60, 90
pilgrimage: to Rusafa, 27, 8586, 89, 94100; to Sergius churches in Mesopotamia, 12329
pilgrim tokens, 38, 38n91, 4243, 86
polytheism, 9, 11, 1516, 25, 9597, 108, 110, 121,
131n4
primicerius scholae gentilium, 8, 1920, 32
processions, 9, 28

739/754

Procopius, historian, 3, 623, 65, 9294, 95n165,

127, 13032, 134
Provincia Arabia, calendar of, 1056
Prudentius, poet, 1822
Ptolemais, 132
Pulcheria, sister of Theodosius II, and holy relics,
4748, 132
puns, 19n42
Qalamun ridges, 2
, 186
, 148
Qara, 187
Qaraqosh, 120
Qardagh, Iranian saint, 40, 120
Qaryatayn, 70

740/754

Qasr al-Hayr al-Gharbi, 175

Qasr ibn Wardan, 113
Qasr Sarij, 124, 124n131, 127
qibla, 177, 181
Qinnasrin. See Chalcis / Qinnasrin
, 67, 177, 183n35
Rabban bar

, 68

Rabbula, bishop of Edessa, 119

Rafniya. See Raphanaea / Rafniya
Raham, 104
al-Ramthaniya, 146, 146n57, 14749
Raphanaea / Rafniya, 4, 114
Raqqa, 184
Rawaq, Jabal, 70

741/754

relics, 2122, 177; as defenders, 3, 2526, 45, 5459,

101; in martyr cult, 8485, 8990; robbery of, 11, 21,
25, 90n140, 93. See also Sergius, relics of
religious accomodation in steppe, 122, 122n125,
12325
reliquaries, 55nn4546, 85, 8990, 134, 141
re-use of images, 35, 35n85
Rhesaina, 102
rider gods / saints, 4042, 11011, 139
Rome, city of, 102
routes, 6176, 9899, 1045, 114
Rozafat, 132n6
Rusafa / Sergiopolis: in frontier politics, 6567, 76,
13241, 143, 15152, 16973, 18184, 19091;
meaning of name, 71, 71n58; processions at, 28, 85;
raids on, 61, 93, 134; site of martyrdom of S. Sergius
at, 10, 11, 15859; topography and relation to routes,

742/754

13, 7, 2021, 6077, 9699. See also animals; pilgrimage; relics; Sergius
Rusafa / Sergiopolis, architecture of: baptistry, 92,
153, 159; basilica A, 78, 8087, 91, 9395, 96n170,
124, 159, 175, 17779, 181, 182n31, 184; basilica B,
28, 7778, 8085, 8791, 9495, 96n170, 104, 154,
159, 184; basilica C, 78n91; brick martyrium, 28, 78,
80, 84, 87; cisterns, 71, 9395; episcopal quarters,
92n145; hostel, 94; khan, 9495, 183; monastery,
126, 80n97, 184; mosque, 84, 91, 17582, 184; alMundhir building, 14973; shops, 68, 80; streets, 78,
78n89; tetraconch, 78, 82, 91, 154; Umayyad pavilion,
175, 175n7; walls, 7778, 9394. See also alMundhir, building of, at Rusafa
Rusafa, near Cordova, 175, 175n10
, 113
Sabrisho, catholicos of Ctesiphon, 14041
Sadad, 18789
saints. See under proper names

743/754

Salaminias / Salamiya, 73, 75, 98n178, 102, 105,

11213, 153, 163
Salamiya, 184, 188. See Salaminias / Salamiya
Salkhad, 108
Salmis, 100
Sanamayn, 155
Sargahan, 120
Sarjis. See Sergius
Schola gentilium, 8, 8n1, 1920, 39
Seleucia, in Cilicia, 85n117, 87n124, 127

Page 226
Seleucia-Ctesiphon, 121, 135; synod of, 53
Sergia, 103n15
Sergiopolis, 92, 92n150. See Rusafa / Sergiopolis
Sergisiyah, monastery of, 189
Sergius, saint: dedications to (see under individual
sites); feast day of, 16, 22n54, 9697, 97n173, 156;
finger of, 129; images of, 96; Muslim reverence for,
17880; relics of, 77, 8445, 180, 18889; reliquary
of, 134, 188n56; as rider saint, 3544, 68, 11011,
139, 18790; sarcophagus of, 8485, 90n140; as Sarjis, 103, 179, 179n19, 180, 182, 182n35; skull of, 189;
as standing soldier, 2935, 4344; thumb of, 65, 92,
129, 132, 180; tomb of, 60, 15859, 171, 182; use of
name, 1014
Sergius II, bishop of Rusafa, 84, 87
Sergius and Martyrius, saints, 188, 188n55, 190n69
Seriane / Isriya, 7374, 112

745/754

Severus, patriarch of Antioch: on Arabs and Rusafa,

96; consecration of, 113; on the Iranian martyrs, 56;
on martyrs as defenders, 46; opposed by Julian of
Bostra, 112; and S. Leontius, 111n57; and SS. Sergius and Bacchus, 2224, 29, 104, 156
Shahba. See Philippopolis / Shahba
Shapur I, Sasanian king, 62
Shapur II, Sasanian king, 62, 120
Shaqqa. See Maximianopolis / Shaqqa
Sharahil, phylarch, 163
Shbayt, Jabal, 118
Shirin, Christian wife of Khusrau II, 13641
Shkodra, monastery of SS. Sergius and Bacchus at,
132n6
al-Sila, Wadi, near Rusafa, 71
silver assemblage from Rusafa, 86, 182, 18587

746/754

Simonopetra, monastery of, 189, 189n63

Sinai, monastery of S. Catherine, 43, 89n135; encaustic icon from, 29, 31, 31n80
Sinjar, 61, 128, 124n133
Sinjar, Jabal, 122, 124
Slayb, tribe, 188n58
Slim, 155
Socrates, historian, on Marutha of Mayperqat, 50, 52
, 27, 104, 118
South Arabia, 64n21, 142, 170n144, 179, 179n17,
180n22, 182n35
Sozomen, historian, 55, 97, 122n123
spolia, 8485, 91
Stephen, saint, relics of, 4748
steppe, 2, 3. See also badiya

747/754

Strata Diocletiana, 20, 22, 70, 74n77, 157

Strata dispute, 66
Sujin, 117
Sukna, 70, 77n86
Sur, 111
Sura, 9n3, 1011, 20, 60, 63, 66, 70, 76, 128, 133,
15758
Suwayda, 108
Symeon, metropolitan of Rusafa, 184
Symeon, saint, 98n180, 99n183, 100, 119, 148
Symeon the Younger, saint, 38, 136
Symeon Beth Arsham, 124n130
Synodicon orientale, on Marutha of Mayperqat, 53,
54n41
synthronon, 91

748/754

Syria-Mesopotamia, fortification of, 76; holy sites in

45, 48, 5759
Tabula Peutingeriana, 21
Taghlib tribe, 72n66, 122n125, 180, 180n22, 188
Takrit, 70, 80n97, 102, 12122, 125
Tal

, 19n45, 21n49

Tal al-Dahab, 113

Tal Musayfna, 124n131
Tanukh, tribe, 64n22, 163, 163n118, 169n143
Tarim, 179, 179n19, 182n35
Taurus, mountain range, 2, 49, 62, 104
Tayyiba: in Hawran, 110; near Rusafa, 70, 72n66,
77n86
templon, 83, 83n108
Terebon, phylarch, 16465, 167, 171

749/754

Tetrapyrgium / Qasr al-Sayla, 10, 2021, 60, 157, 175

Tha'labids, 64
Thecla, saint, 4, 46n5, 85n117, 86n119, 87n124,
89n132, 98, 99n183, 117, 127, 179n17, 190
Theodora, Roman empress, 80n97; and S. Sergius,
13033, 137, 13940, 168
Theodore, metropolitan of Bostra, 133
Theodore, saint, 4, 110, 114, 118

Page 227
Theodore Lector, historian, on Marutha of Mayperqat,
50
Theodoret, bishop of Cyrrhus, 27, 46, 95; on the Iranian martyrs, 56
Theodore of Euchaita, saint, 96, 98
Theodosius II, Roman emperor, 122n123; and holy
relics, 4748, 132; relations with Marutha of Mayperqat, 5355, 55n46, 56
Theophanes, historian: on Arethas, 64, 64n21; on
Marutha of Mayperqat, 50, 55
Theophylact, historian: on the "barbarikon," 65,
65n28; on Khusrau II at Dara, 16566; on link
between Mayperqat and Rusafa, 58; on route through
the north Syrian steppe, 63; on Sergius, 120, 13436
Theotokos, 113, 139, 187
Thessalonica, 32, 32n3, 43, 89n132
Thomas, apostle and martyr, 96, 161

751/754

Tiberius Constantinus I, Roman emperor, 17273

Tigris, river, 121, 125, 127n145
Timothy, apostle, 45
toponymy, 22, 6667, 7475
torque, 18. See also maniakion
tribal church, 122, 16264, 169
tribal mosque, 162n114
Tripolis / Tripoli, 111, 111n57, 114
al-Tuba, 153, 155
Tur

, 19
I, caliph, 174

Umayyad dynasty, 168, 174, 17880; at Rusafa, 84,

91, 104, 109, 126, 175, 184
Umm al-Jimal, 108

752/754

Umm Qays. See Gadara / Umm Qays

Umm al-Surab, 104, 1089
, 164n120
Urfa. See Edessa
Urmiya, Lake, 189
. See Occariba
Vahan, Armenian martyr at Rusafa, 183
Vatopedi, monastery of, 189
Victor and Corona, saints, 18, 24n60, 90n138
votive offerings, 8687, 13233, 13539, 18587
wadis, 7172, 99. See also under proper names
al-Walid I, Umayyad caliph, 17778
Willibald, English pilgrim, 113

753/754

Yaqut, geographer: on Marutha of Mayperqat, 51; on

monastery at Rusafa, 184; on Muslim tomb in Christian monastery, 182
Yarmuk, river, 74n78, 145, 174
Yazdgard I, Sasanian king: relations with Arabs of
Hira, 62; relations with Marutha of Mayperqat, 48, 50,
5254, 55n46, 56, 59, 141
Zabad, 104, 11718, 163
Zagros, mountain range, 2, 7, 120, 140
al-Zaydi, Wadi, 110
Zenobia / Halabiya, 63, 92, 128, 133, 160
Zenobia, Queen, 74
Zorava /
, inscription naming S. Sergius as horseman, 11011
Zoroastrians, 3, 121, 135, 167

@Created by PDF to ePub