The Roman Eastern Frontier and the Persian Wars AD 363-628. (Geoffrey Greatrex, Samuel N. C. Lieu)

Формат документа: pdf
Размер документа: 3.88 Мб

Прямая ссылка будет доступна
примерно через: 45 сек.

Теги: Древний Рим. Английский язык
  • Сообщить о нарушении / Abuse
    Все документы на сайте взяты из открытых источников, которые размещаются пользователями. Приносим свои глубочайшие извинения, если Ваш документ был опубликован без Вашего на то согласия.


PART II AD 363–630
A narrative sourcebook
Edited and compiled by Geoffrey Greatrex
and Samuel N.C. Lieu
London and New York

First published 2002by Routledge
11 New Fetter Lane, London EC4P 4EE
Simultaneously published in the USA and Canada by Routledge
29 West 35th Street, New York, NY 10001
Routledge is an imprint of the Taylor & Francis Group
© 2002 Geoffrey Greatrex and Samuel N.C. Lieu
All rights reserved. No part of this book may be reprinted or reproduced
or utilised in any form or by any electronic, mechanical, or other means, n
owknown or hereafter invented, including photocopying and recording, or in
any information storage or retrieval system, without permission in writi
ng from the publishers.
British Library Cataloguing in Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging in Publication DataA catalog record has been requested
ISBN 0–415–14687–9
This edition published in the Taylor and Francis e-Library, 2005.
“To purchase your own copy of this or any of Taylor & Francis or Rout
ledge’s collection of thousands of eBooks please go to
ISBN 0-203-99454-X Master e-book ISBN (Print Edition)


List of mapsix
Preface x
Acknowledgements xiii
Abbreviations xv
Eastern Roman emperors (363–630) xvii
Persian kings (363–630) xviii
Glossary xix
Notes on the sources xxi
Maps xxviii–xxxii
1 The Peace of Jovian and its aftermath in Mesopotamia (363–399) 1
2 The evolution of the north-east frontier (363–399) 20
3 The Mesopotamian frontier in the fifth century 31
4 The north-east frontier in the fifth century 53
5 The Anastasian War and its aftermath (502–525) 62
6 Justinian’s First Persian War and the Eternal Peace (c.525–540) 82
7 Justinian’s Second Persian War: the southern front (540–545) 102
8 Justinian’s Second Persian War: the northern front (540–562) 115
9 Justinian’s Second Persian War: diplomatic relations (545–562) 123
10 The failure of the Peace of 562 (562–573) 135

11 The war under Tiberius (574–582)151
12 The reign of Maurice (582–602) 167
13 The Persian takeover of the Near East (602–622) 182
14 The Roman recovery under Heraclius (622–630) 198
15 The Khuzistan Chronicle (first part) 229
16 The evidence of epigraphy: the eastern frontier (363–630) 238
Notes 247
Bibliography 323
Index of sources 350
General index 355

1 The Near East in late antiquityxxviii
2 The southern theatre of war (Mesopotamia and surrounding regions) xxix
3 The northern theatre of war (Armenia and surrounding regions) xxx
4 The western Transcaucasus (Lazica and surrounding regions) xxxi
5 The Transcaucasus xxxii
6 Provenance of inscriptions discussed in Chapter 16 239

This book differs in several respects from its predecessor. Instead of offering merely a
series of translated excerpts from ancient authors, we have attempted to fill the gaps
between extracts, thus providing a history of the frontier and of Romano–Persian rela-
tions from 363 to 630. We have not sought to address broader issues, such as trading
relations, or to analyse long-term developments (for instance) in diplomacy between
the two powers. Discussion of these matters may be found elsewhere, and we may take
the opportunity to refer the interested reader to a thematic sourcebook in German on
the whole Sasanian period by E. Winter and B. Dignas (2001). We have consciously
departed from the traditional sourcebook system of numbering the extracts in each
chapter. Our intention is to encourage those using the sources here assembled to refer
to the actual passage in the source itself, rather than simply to Greatrex and Lieu 9.2.
We hope that this will lead to greater clarity for both students and researchers, and
perhaps to more of an awareness of the actual sources themselves. We believe that the passages translated here are of great importance to the subject.
It does not follow from this that passages excluded from this book are not. Various
criteria, such as relevance, the availability of an English translation and the reliability
of the source in question, determined whether a given excerpt was included or not.
Hence, just as in volume 1 few passages from Ammianus Marcellinus were presented,
here the reader will not find much Procopius, Joshua the Stylite or Sebeos, because
translations of all three are readily obtainable. We have not included any material by
George of Pisidia, whose poems in honour of Heraclius are an important source for
the early seventh century. Our decision in this was prompted partly by difficulties in
how to interpret the information presented by George, and partly by the fact that an
English translation of many of his poems is being prepared by Mary Whitby. None -
theless we have tried to give references to all relevant sources, even when they are not
quoted. By contrast, there are substantial excerpts from Zachariah of Mytilene and
other lesser-known Syriac sources. Fairly long sections of Theophanes and the
Chronicon Paschale have been included, particularly concerning the reign of
Heraclius. Although excellent English translations of both are available, we consid -
ered them too important to omit. The spelling of names is always problematic in a work of this nature. We have opted
for conventional forms of names commonly referred to, e.g. Khusro, Shahrvaraz,

Heraclius, Bahram and the Tur Abdin. In the case of the name Khusro, used by both
Armenian and Persian rulers, we have used the Armenian form Khosrov for Armenian
rulers of this name. We thus refer to Arsaces rather than Arshak, but keep to the Arme-
nian form of Mushe³ and others who do not feature in Latin or Greek sources. In less
obvious cases we have tended simply to transliterate the name in question, occasion -
ally correcting it where possible (e.g. rendering the Greek name Nakhoragan as the
Persian title nakhveragan). We have avoided diacritics on the whole, and have omitted
the Syriac letter ainin cases where a name is commonly used, e.g. Sabrisho for
Sabrisho‘, Abdin for ‘Abdin. In Chapter 16, upon the recommendation of Irfan
Shahîd, we have tended to keep to the transliteration of Arabic place names adopted
by IGLS. Anachronistic references to Greeks and Parthians have been retained; the
former refers, of course to the Romans, the latter to the Persians. We have referred on
occasion to ‘volume 1’, by which we mean Dodgeon and Lieu 1991. We have also
tended to refer to the ‘Transcaucasus’ rather than ‘Caucasia’, although the latter term
is probably now more common for the region south of the Caucasus mountai
ns. In translations, we have consistently referred to the ‘king of Persia’ and the ‘Roman
emperor’. In fact, the word used for both rulers is almost invariably the same; it is for
the sake of convenience that we have translated them differently, so that references to
‘king’ and ‘emperor’ are instantly clear. We have also generally preferred to translate
phrases such as ‘king of the Romans’ as ‘Roman emperor’ to avoid verbosity. The
translations themselves aim to keep as close as possible to the original text, and may
therefore appear somewhat inelegant; clarity, we hope, has not been sacrificed. We
should point out too that we have considered the eastern frontier as running from the
Transcaucasus in the north to Syria in the south. We have not sought to include mate-
rial from the provinces of Palestine or Arabia on the whole, save when they became the
scene of conflict between Romans and Persians. In Syriac texts, we have translated ‘Arbaye as Arabs, but have left Tayyaye as it
stands: the latter is the more common term for Arabs, but we thought it best to
preserve the distinction. We have distinguished throughout between Jafnids and
Ghassanids: as several scholars have recently pointed out, the former were the descen -
dants of Jafna who ruled (part of) the Ghassan tribe.
1We have latinised the Greek
terms strat¢gos andstrat¢lat¢s where it is possible to give the formal Roman office with
certainty (e.g. magister militum per Orientem); if in doubt, however, we have left the
Greek. We have referred consistently to the catholicoiof the Persian and Armenian
Churches, although aware that these terms are anachronistic to some extent. They are,
however, convenient and conventional.
2We refer throughout to Monophysites and
Nestorians for the sake of convenience, despite the unsatisfactory nature of these
labels. Our system of bracketing also requires explanation:
( ) usually indicates an insertion by the translator to indicate that a word or phrase,
though necessary to the English, is not present in the original text.
< > indicates an insertion by the editor of the text.
[ ] indicates a passage which the editor of the text considers was not present in the

original (and hence should be omitted) or, in the case of inscriptions, words or parts
of words no longer surviving, but the restoration of which has been sugg
The abbreviation
AG, used in connection with chronicle entries, refers to a date
according to the Seleucid era (starting on 1 October 312
BC). Access to the new edition
of Malalas’ Chroniclehas proved difficult for us. We have used it for the passages trans -
lated, but elsewhere have had to content ourselves with giving book and chapter refer -
ence, together with the pages in the Bonn edition. Many individuals have been of assistance to us in the preparation of this volume,
often producing translations and answering questions at short notice. Among them
we should like to thank Sebastian Brock, Richard Burgess, Ian Colvin, Jan van
Ginkel, Marina Greatrex, Tim Greenwood, George Hewitt, Robert Hewsen, James
Howard-Johnston, Robert Hoyland, Peter O’Brien, Stephen Rapp and Irfan Shahîd.
Among them, we are particularly indebted to Jan van Ginkel, Marina Greatrex and
Tim Greenwood, who provided us with numerous translations which appear here.
Uncredited translations and revisions are by Geoffrey Greatrex. Engelbert Winter was
kind enough to send us a draft of his forthcoming sourcebook, written in collabora -
tion with Beate Dignas, on the East Roman frontier (from 224 to 630). We have tried
to incorporate some references to this useful work, which offers a thematic approach
rather than a narrative such as we have constructed. We are grateful also to the various
graduate students who assisted in some of the translations for the project, in particular
to Ken Donovan at Cardiff and John Holton at Dalhousie. John Holton was respon-
sible for preparing the brief guide to the sources, which was then revised by the
editors. We must also thank Corey and Stacey Owen and Marian Jago for the prepara-
tion of the index and Jane Anson for her diligent copy-editing. It is a pleasure to acknowledge the support for the reseach behind this volume
which has come from the Seven Pillars of Wisdom Trust and the Social Sciences and
Humanities Research Council of Canada. The Hugh Last and Donald Atkinson
Funds of the Society for the Promotion of Roman Studies also generously provided a
grant to help cover the cost of preparing the maps. The maps themselves were drawn
up by R. Hewsen, and we are grateful to him for the work involved. There will be a website associated with this work, which will permit additional
references and material to be supplied on a continuing basis.
Geoffrey Greatrex and S.N.C. LieuApril 2001

The editors of this collection are grateful to the following publishers and authors for
permission to republish copyright material:
The journalByzantionfor translations which first appeared there in 1999.
The journal Revue des Études Byzantines for a translation which appeared in REB46
The Calouste Gulbenkian Foundation for an extract from N. Adontz and N.
Garsoïan, Armenia in the Period of Justinian (Lisbon, 1970).
Princeton University Press for three extracts from C. Pharr, The Theodosian Code,
copyright © 1952 by Clyde Pharr (new edition 1980 by Roy Pharr).
The State University of New York Press and C.E. Bosworth for two passages from
C.E. Bosworth, The History of al-Tabari. Volume 5. The S¡s¡nids, the Byzantines, the
Lakhmids, and Yemen (1999).
The journal Phoenixfor one translation by R.C. Blockley from Phoenix39 (1985).
The University of California Press for a translation from O. Maenchen-Helfen, ed.
M. Knight, The World of the Huns (1973).
Francis Cairns (Publications) Ltd and R.C. Blockley for translations by R.C. Blockley
from Eunapius, Priscus, Malchus and Menander, taken from The Fragmentary
Classicising Historians of the Later Roman Empire, vol.2 (1983) and The History of
Menander the Guardsman (1985).
AMS Press for translations by E.W. Brooks and F.J. Hamilton of The Syriac Chronicle
known as that of Zachariah of Mytilene (1899).
Byzantine and Modern Greek Studies and Frank Trombley for three translations by
F.R. Trombley from BMGS21 (1997).
Greek, Roman and Byzantine Studies for one translation by O.J. Schrier from GRBS33

Georgetown University Press for one translation by C. Toumanoff fromStudies in
Christian Caucasian History (1963).
Dumbarton Oaks for one translation from DOP23–4 (1969–70) by Averil Cameron
and one from Nikephoros. Short History, ed. and tr. C. Mango (1990).
The Australian Association of Byzantine studies, Brian Croke, Elizabeth Jeffreys and
Roger Scott for translations from Malalas (1986) and Marcellinus comes(1995).
Michael and Mary Whitby for ten translations from The History of Theophylact
Simocatta (Oxford, 1986).
The University of Pennsylvania Press for an extract from Maurice’s Strat¢gikon:
Handbook of Military Strategy, translated by George T. Dennis, copyright © 1984
University of Pennsylvania Press, reprinted with permission.
The two extracts from S.P. Scott’s translation of The Civil Law are reprinted by
permission of Thomas Jefferson University.
The extract from Libanius, Autobiography and Selected Letters is reprinted by permis-
sion of the publishers and the Trustees of the Loeb Classical Library from Libanius:
vol.479, translated by A.F. Norman, Cambridge, Mass.: Harvard University Press,
1992. The Loeb Classical Library ® is a registered trademark of the President and
Fellows of Harvard College.
The extracts from Procopius are reprinted by permission of the publishers and the
Trustees of the Loeb Classical Library from Procopius: vols. 048, 173, 290, 353,
translated by H.B. Dewing, Cambridge, Mass.: Harvard University Press, 1914,
1928, 1935, 1940. The Loeb Classical Library ® is a registered trademark of the Presi-
dent and Fellows of Harvard College.
The extracts from Theophanes are © Cyril Mango and Roger Scott 1997. Reprinted
from The Chronicle of Theophanes Confessor: Byzantine and Near Eastern History
284–813, translated with an Introduction and Commentary by Cyril Mango and
Roger Scott, with the assistance of Geoffrey Greatrex (1997), by permission of Oxford
University Press.
The extract from Novel 28is © David Braund 1994. Reprinted from Georgia in
Antiquity: A History of Colchis and Transcaucasian Iberia 550
BC–AD 562 by David
Braund (1994) by permission of Oxford University Press.
Some entries in the glossary are based on the entries to be found in The Chronicle of
Theophanes Confessor: Byzantine and Near Eastern History
AD 284–813, translated with
an Introduction and Commentary by Cyril Mango and Roger Scott, with the assis -
tance of Geoffrey Greatrex (1997) and in G. Greatrex, Rome and Persia at War,
502–532(1998). We are grateful to both Oxford University Press and to Francis
Cairns (Publications) Ltd for permission to use this material.

Journal abbreviations not noted here follow the system used byODB.
AAASH Acta Antiqua Academiae Scientiarum Hungaricae
AASS Acta Sanctorum (Paris 1863–1940)
AAES Publications of an American Archaeological Expedition to Syria in 1899– 1900(New York 1903–30)
AnTard Antiquité Tardive
BICS Bulletin of the Institute of Classical Studies
CHI The Cambridge History of Iran, vol. 3, ed. E. Yarshater (Cambridge 1983)
CIL Corpus Inscriptionum Latinarum (Berlin 1863–)
DRBE The Defence of the Roman and Byzantine East, ed. P. Freeman and D. Kennedy, BAR International Series 297 (Oxford 1986)
EFRE The Eastern Frontier of the Roman Empire, ed. D.H. French and C.S. Lightfoot, BAR International Series 553 (Oxford 1989)
EIr Encyclopaedia Iranica, ed. E. Yarshater (London 1985–)
FCH The Fragmentary Classicising Historians of the Later Roman Empire, 2 vols, R.C. Blockley (Liverpool 1981–3)
FHG Fragmenta Historicorum Graecorum, ed. C. Müller, vols 4–5 (Paris 1851–70)
IGLS Inscriptions grecques et latines de la Syrie, edd. L. Jalabert, P. Mouterde et
al. (Paris 1929–)
LA Late Antiquity. A Guide to the Postclassical World, ed. G.W. Bowersock, P. Brown and Oleg Grabar (Cambridge, Mass., 1999)
LRE A.H.M. Jones The Later Roman Empire, 284–602, 3 vols (Oxford)
MGHAA Monumenta Germaniae Historica, Auctores Antiquissimi(Berlin 1877–
3 The Oxford Classical Dictionary, third edition, edd. S. Hornblower and A.
Spawforth (Oxford 1996)
ODB The Oxford Dictionary of Byzantium, ed. A. Kazhdan, 3 vols (Oxford 1991)
PG Patrologia Graeca (Paris 1857– )
PL Patrologia Latina (Paris 1884– )

PLRE Prosopography of the Later Roman Empire, ed. J. Martindaleet al., 3 vols
(Cambridge 1971–92)
PO Patrologia Orientalis (Paris 1903– )
PUAES Publications of the Princeton University Archaeological Expedition to Syria in 1904–5 and 1909 (Leiden 1907–49)
RBNE The Roman and Byzantine Near East , ed. J.H. Humphrey, 2 vols
(Portsmouth, RI, 1995–9)
SRA The Byzantine and Early Islamic Near East. States, Resources, Armies, Studies in Late Antiquity and Early Islam, vol. 3, ed. A. Cameron
(Princeton 1995)

Julian (361–3)
Jovian (363–4)
Valens (364–78)
Theodosius I (379–95)
Arcadius (395–408)
Theodosius II (408–50)
Marcian (450–7)
Leo I (457–74)
Leo II (474)
Zeno (474–5)
Basiliscus (475–6)
Zeno again (476–91)
Anastasius I (491–518)
Justin I (518–27)
Justinian I (527–65)
Justin II (565–78)
Tiberius II (578–82)
Maurice (582–602)
Phocas (602–10)
Heraclius (610–41)xvii

Shapur II (309–79)
Ardashir II (379–83)
Shapur III (383–8)
Bahram IV (388–99)
Yazdgerd I (399–420)
Bahram V (420–38)
Yazdgerd II (438–57)
Hormizd III (457–9)
Peroz (459–84)
Balash (484–8)
Kavadh (488–96)
Zamasp (496–8)
Kavadh again (498–531)
Khusro I (531–79)
Hormizd IV (579–90)
Khusro II (590–628)
Kavadh II (628)
Ardashir III (628–30)
Khusro III (630)
Boran (630–1)xviii

azatsAn Armenian term literally meaning ‘free’, i.e. the nobility, and thus ‘cavalry’,
supplied in the main by the nobles. Garsoïan 1989: 512–13.
a secretis A high-ranking imperial secretary. ODB204.
catholicos The term in Greek for the head of the Persian, Armenian or other
Transcaucasian churches. ODB1116.
centenarion, pl. centenariaA unit of weight (100 Roman lbs), often applied to gold
coins. ODB1121.
comes, pl. comitesA count, a term given to various imperial officials. LRE104–5,
ODB 484–5.
comes domesticorum The official in charge of the corps of officer cadets, the
domestici et protectores. LRE 372
comes excubitorum The official in charge of the excubitors (on whom see below).
comes foederatorum The official in charge of the ‘federate’ (allied) forces. LRE665.
comes (sacrarum) largitionum The count of the Sacred Largess, a high-ranking offi-
cial in charge of finances. LRE427–38, ODB486.
comes Orientis The count of the East. LRE373–4.
comes rei militaris A military count.LRE105.
comitatenses The Roman field army, more mobile than the limitanei(on whom see
below). LRE608–10, ODB487.
dux, pl. ducesa duke or military commander attached to a particular province. LRE
609–10, ODB659.
erist‘avi, pl. erist‘avebi(oldererist‘avni) In feudal times in Georgia (Iberia) it signi -
fied the ruler/governor of a regional province or community. Also used to signify
‘military leader, general’ (equivalent to the Persian sparapet).
excubitor A member of an elite corps of 300 Roman soldiers. LRE658–9.
foederati, sing. foederatusTechnically, soldiers serving in the Roman army by the
terms of a treaty (foedus), but in fact scarcely different from comitatensesby the
sixth century. LRE663–4, ODB794.
hazarapet Commander of a thousand/chancellor. An Armenian office, originally
military in nature, but by late antiquity civilian. Garsoïan 1989: 531–2.
kanarang The

Persian military commander in charge of Abharshahr (the north-
eastern frontier of the Persian empire). Christensen 1944: 107 n.3.

limitaneiThe ‘frontier forces’ of the Roman empire. LRE608–10, ODB1230.
magister militum Master of soldiery, high-ranking military official. LRE608–10,
ODB 1266–7.
magister militum per Armeniam The master of soldiery stationed on the north-
eastern frontier. LRE271.
magister militum per Orientem The master of soldiery stationed on the eastern frontier.
magister militum praesentalis A master of soldiery stationed in the capital. LRE124–5.
magister officiorum The master of offices, the head of the central civil administra -
tion of the empire. LRE368–9, ODB1267.
malkhaz An Armenian title meaning prince or lord. The exact nature of the office is
uncertain. Garsoïan 1989: 542.
mihran The name of a Persian noble family, often misinterpreted by Roman
sources as an office. Christensen 1944: 105 n.3.
marzban The Persian title of a governor of a province. Christensen 1944: 136–
nakharar The Armenian term for a noble. Garsoïan 1989: 549.
nakhveragan A Persian title, often mistaken for a name in Greek sources (Nakhoragan),
derived from the same root as nakharar. Christensen 1944: 21 and 452.
nomisma pl.nomismata The Greek term for the Latin solidus(see below).
numerus A unit of (Roman) soldiers. LRE659.
phylarch A commander of auxiliaries (often Arab) allied to the empire. LRE611,
ODB 1672.
praetorian prefect An important regional civil functionary. LRE370–2, ODB
quaestor (sacri palatii) High-ranking imperial official concerned with legal matters.
LRE 387, ODB 1765–6.
Royal Gate (or Gate) The headquarters of the Persian king, wherever he happens to
be. See Peeters 1951a: 143 n.2.
scholae (palatinae), sing. scholaThe corps of guards of the imperial palace. LRE
647–8, ODB1851–2.
scriniarius Member of an imperial department (or scrinium).LRE449.
silentiarius A member of the corps of Roman palace attendants. LRE571–2, ODB
solidus, pl. solidiA Roman unit of currency. There were 72 gold solidito
(Roman) pound of gold. ODB1924.
spahbadh A Persian term for a general. Christensen 1944: 131.
sparapet The Armenian term for commander-in-chief (which can refer to the
magister militum). Garsoïan 1989: 560–1.
stade A classical unit of measurement of distance. ODB1373.
gos The traditional Greek word for a general, often a magister militum.ODB1964.
lat s Another Greek word for a general, often a magister militum.
tagma, pl. tagmata A unit of soldiers (esp. a mobile unit), the Greek equivalent of a
numerus. ODB2007.
turmarch Commander of a turmaor military detachment, a term which came into
use in the eighth century. ODB2100.

(Sources which are quoted only once or twice are introduced in the notes.)
Agapius(died after 941) Also known as Mahbub ibn Qustantin, bishop of Hierapolis in
Euphratesia, author of a universal history in Arabic called The Book of the Title, from the
Creation to his own day; the surviving portion records events down to 776. His work preserves
fragments of the Greek Chronicle of Theophilus of Edessa (died 785) and uses earlier Syriac
Agathias (c.532–c.580) Greek historian, lawyer and poet. A native of Myrina in Asia Minor,
he was the author of a continuation of Procopius’ history, covering the years 552–8.
Ammianus Marcellinus (c.330–c.395) Latin historian but of eastern origin. A native of Tyre
or Sidon, he saw service on the eastern frontier on the staff of Ursicinus. The surviving portion
of his history which covers the years 353–78 is an essential source of eyewitness information on
the later Persian wars of Constantius II and the expedition of Julian.
(Anon.) Life of John the Almsgiver (V. Ioh. Eleem.) An epitome of earlier biographies of John,
the patriarch of Alexandria (610–619/20), by his contemporaries John Moschus and
Antiochus Strategius A monk in Palestine and author of an eyewitness account of the Persian
siege of Jerusalem in 614. The original Greek version of the work is lost, but Georgian and
Arabic versions survive.
Augustine (354–430) Bishop of Hippo Regia in Africa, Latin theologian. Author of the
Confessions and theCity of God, among many other works.
Barhebraeus (1225–86) Western name of Gregory Abu’l-Faraj, a Syriac scholar and
Monophysite bishop, who composed many works, including his ChronicleorChronography,a
universal history based on the Chronicleof Michael the Syrian; its secular history covers the
period from Adam to the Mongol invasions, while its ecclesiastical history covers the period
from Moses’ brother Aaron to his own day.
Cedrenus, Georgius (twelfth century) A Byzantine chronicler whose work, based on earlier
chronicles, begins with the Creation and goes down to the year 1057.
Chronicon anonymum ad a.d. 819 A short annotated Syriac chronicle of important events
and people from Christ to 819. Its author was a Monophysite, probably a monk from a monas -
tery near Mardin. It was later drawn on extensively by the Chronicon ad
AD 846 pertinens.
Chronicon anonymum ad a.d. 1234 pertinens A universal Syriac history composed c.1240 by

an anonymous Edessan author. It covers much of the same material as theChronicleof Michael
the Syrian while diverging in many details. It also preserves important passages of otherwise lost
authors including John of Ephesus and Dionysius of Tel-Mahre.
Chronicle of Arbela A Syriac work giving biographies of the bishops of Arbela (in Adiabene)
up to the sixth century
AD . The reliability of its first sections has been called into question.
Chronicle of Edessa (sixth century) A Syriac chronicle (entitled Histories of Events in Brief),
consisting of a list of important events and Church figures connected with the city of Edessa,
covering the period from the second to the sixth centuries.
Chronicon Paschale (630s) A Byzantine universal chronicle which covers the period from
Adam to 629/30 from a Constantinopolitan perspective. The very last section of the work is
lost. The chronicle preserves important information on Heraclius’ final campaign against the
Persians, based on the emperor’s own despatches.
Claudian (c.370–404) A Latin poet from Egypt, whose panegyrics and epigrams are a major
source for the military and political history of the period 395–404.
Codex Justinianus A predominantly Latin collection of imperial constitutions from the
period of Hadrian to Justinian I, forming a part of the Corpus Juris Civilis. It was comissioned
by Justinian as a categorised and internally consistent replacement of earlier codes including the
Codex Theodosianus. It was first publicly introduced in 529, with an amended edition estab -
lished in 534.
Codex Theodosianus An official collection of imperial constitutions from
AD 312 until 438
when the Code was published by Theodosius II.
Constantine VII Porphyrogenitus (905–59) Emperor of the Macedonian dynasty (945–59),
who compiled an encyclopaedic treatise on court ceremony entitled De ceremoniis aulae
byzantinae (On the ceremonies of the Byzantine court ).
Corippus (Flavius Cresconius Corippus) (died after 567) A Latin poet and native of Africa,
who celebrated the successes of the Roman general John Troglita in an epic poem, the Iohannis,
and who later at Constantinople commemorated the accession of Justin II in four hexameter
books in c.566.
Cyril of Scythopolis (c.525–59) A monk and hagiographer who composed a number of
biographies of Palestinian monks including Sabas, Abraham, Kyriakos, Theodosius, and
Theognius; the biographies incorporate useful historical details.
Pseudo-Dionysius (of Tel-Mahre) Unknown author of a late eighth-century universal
chronicle in Syriac, also known as the Zuqnin chronicle, which records events from biblical
times to 775 and incorporates important information from lost sources (including the entire
work conventionally ascribed to Joshua the Stylite).
E³ishe (? sixth century) Author of an Armenian History which describes the unsuccessful
revolt of Vardan Mamikonian against Sasanian overlordship in 450/1.
Ephrem Syrus (c.306–73) Syrian poet and theologian. Born in or near Nisibis, he was
compelled to leave his native city because of the treaty of 363 which surrendered it to the
Persians. He later settled in Edessa. His Carmina NisibenaandHymni contra Julianum both
include eyewitness material on the events they describe.
Epic Histories An anonymous work written in Armenian in the 470s and based on local oral
traditions. It relates in epic style the history of the later Arsacid dynasty as well as that of the
Mamikonians and that of the house of Gregory the Illuminator.

Epitome de CaesaribusTitle given to an anonymous collection of biographies (in Latin) of
emperors from Augustus to Theodosius. The work is sometimes wrongly associated with
Aurelius Victor.
Eunapius (c.345–after 414) Greek sophist, born near Sardis in Lydia, author of a continua -
tion of the history of Dexippus from a pagan viewpoint covering the period
AD 270–404.
Eustathius of Epiphania (died c.505) Historian and author of a chronicle which now survives
only in fragments. It probably covered the period from the fall of Troy to the Roman wars
against Persia in 502–5, and was used as a source by Malalas and Evagrius.
Eustratius (died after 602) A hagiographer and priest of Hagia Sophia in Constantinople. He
wrote a biography of the Persian saint Golinduch, based on the work of Stephen of Hierapolis,
as well as a panegyric of his teacher Eutychius.
Eutropius (late fourth century) Latin historian who accompanied Julian on his Persian expe -
dition. His epitome of Roman history, published in 369/70, begins with Romulus and finishes
with the reign of Jovian.
Evagrius (c.536–c.594) Greek lawyer and ecclesiastical historian. A native of Epiphania in
Syria, he wrote a Church history, incorporating a considerable amount of information on
secular matters, covering the period 431–594.
Festus (died 380) Latin historian. He was magister memoriaewhen he published his brief
summary (Breviarium) of Roman history in 369/70.
Firdausi (Abu ‘l-Kasim) (940/1–1020) Persian poet and author of the Sh¡hn¡ma, a vast epic
of Persian history from the Creation to Sasanian times, which makes use of lost sources, some of
which derive ultimately from the Sasanian royal annals.
George of Pisidia (died 631/4) Church official at Constantinople and panegyrist of the
Emperor Heraclius. Many of his epic poems recount the successes of Heraclius, which are
attributed to divine favour. His style is often allusive and is difficult to use for historical
Gregory of Nazianzus (329/30–c.390) Theologian and bishop of Constantinople (380–381)
and of Nazianzus (382–84); author of numerous homilies and other works.
Gregory of Tours (c.540–593/4) Bishop of Tours and author of the Historiesin ten books.
The work starts with Creation, but concentrates on the rising power of the Franks from the fifth
century down to his own day.
Isaac of Antioch (fifth century) Syriac writer and author of many poetical works, which are
mostly mixed together with those of Isaac of Amida.
Jacob of Edessa (died 708) Syriac scholar and bishop of Edessa. Author of a Chronicle, which
continued that of Eusebius, taking it up to 692. Only fragments of this now survive, which can
be supplemented by material from later works (derived from Jacob).
Jerome (Eusebius Hieronymus) (c.342–420) Biblical scholar and theologian. Born near
Aquileia, he spent much of his life in Bethlehem. He translated Eusebius’ Chronicleinto Latin
and extended it up to 378.
John Chrysostom (c.340–407) Bishop of Constantinople (398–404) and famous orator.
Expelled from Constantinople for his criticism of the Empress Eudoxia, he died in exile at
Cucusus in Armenia.

John Diakrinomenos(late fifth century) Author of an ecclesiastical history from Theodosius
I to Zeno, only fragments of which survive.
John of Biclar (died 621) Bishop of Gerona in Spain. John was brought up in Constanti -
nople, returning to Spain in c.576. His chronicle in Latin covers the pe
riod 567–90.
John of Ephesus (c.507–586/8) Syriac historian and Monophysite leader in Constantinople
during the reign of Justinian I, ordained bishop in 558. He wrote the Lives of the Eastern Saints,
a collection of stories about 58 holy men and women of the Syriac-speaking world of his day, as
well as a Church History with a Monophysite perspective, of which only about one-third
survives, covering the period 571–86.
John of Epiphania (sixth/seventh century) A relative of the Church historian Evagrius, a
lawyer, and the author of a history which continued where Agathias had left off. Only one frag -
ment survives.
John Lydus (the Lydian) (490–c.565) Scholar, bureaucrat and writer who composed a trea -
tise, De Magistratibus (On the Magistracies), a history and description of the Roman offices of
Jordanes (? died 552) Former secretary of a Gothic leader. Author of a three-part history in
Latin, one section of which, the Romana, offers a brief synopsis of Roman history from
Romulus to 550/1.
Pseudo-Joshua the Stylite Edessan priest and monk of unknown date, to whom a sixth-century
Chronicle of the Persian War (entitledThe History of the Time of Troubles in Edessa, Amida, and all
Mesopotamia) is often attributed. The work is remarkably detailed and informative.
K‘art‘lis C‘xovreba (Georgian Chronicles) An official collection of some Georgian historical
works written between the eighth and fourteenth centuries, commissioned in the early eigh-
teenth century by King Vakhtang VI. It includes a History of the Kings of Iberia,aHistory of King
Vakhtang Gorgasali (HVG) and other works.
Khuzistan Chronicle (also known as the Guidi chronicle) An anonymous work compiled in
the 660s by a Nestorian author, probably in Khuzistan. It is valuable as a source written within
the Persian kingdom not long after the events related.
´azar (´azar P‘arpec‘i) (second half of fifth century) Author of a History of Armenia,
containing an account of the Armenian revolt against Persia in 451, and of the career of Vahan
Mamikonian in the aftermath of the Armenian defeat.
Leontius of Neapolis (seventh century) Bishop and hagiographer who wrote a Lifeof St John
Eleemon (John the Almsgiver), an important source on Egypt in the early sevent
h century.
Libanius (314–c.393) Greek sophist and rhetor of Antioch. One of the most influential
pagans of the fourth century; his speeches and letters provide a wealth of information on events
in the East.
Malalas, Ioannes (c.490–c.575) Rhetor and official at Antioch, who probably moved to
Constantinople in the 540s. He wrote the first Byzantine universal chronicle, covering the
period from Adam to
AD 565. His work preserves many important notices on Antioch and the
province of Syria.
Malchus of Philadelphia (fl. late fifth century) A sophist at Constantinople who wrote a
history called Byzantiaka, whose surviving fragments cover the period 473–480.
Marcellinus Comes(sixth century) Latin author who wrote a chronicle at Constantinople
which originally covered the period 379–518 (designed as a continuation of that of Jerome) to

which he later added a sequel extending up to 534. A second supplement covering the period
until 548 (where the manuscript breaks off) was added by a later, unkn
own author.
Maurice(c.539–602) Byzantine emperor (582–602) to whom is often attributed a work
entitled the Strat¢gikon, a Greek military treatise on cavalry warfare which includes a survey of
various foreign peoples.
Menander Protector (Menander the Guardsman) (fl. late sixth century) Constantinopolitan
historian and palace guardsman who wrote a history covering the period 558–582, which was
designed as a continuation of Agathias, and survives today in seventy fragments.
Michael the Syrian (1126–99) Monophysite patriarch of Antioch and compiler of an impor -
tant chronicle in Syriac from the Creation to 1194/5. His work makes use of many earlier
sources, now lost (cf. Chr. 1234, above).
Moses Khorenats‘i Name given to the author of a work of great importance on Armenian
history, covering the period to 439, but of uncertain date.
Movs¢s Daskhurants‘i (or Ka³ankatuats‘i) (? tenth century) Identified as the compiler of a
tenth-century historical text focused upon successive kings, princes and Church leaders of
Caucasian Albania. The work incorporates much earlier material from unidentified sources, in
particular on the 620s.
Narratio de rebus Armeniae (c.700) A short Armenian text which survives only in Greek.
Composed c.700, it concentrates primarily upon Armenian ecclesiastical history. Unusually for
a surviving Armenian text, it is favourable towards the Council of Chalcedon.
Nicephorus I (c.750–828) Patriarch of Constantinople and historian. Author of a Short
History (theHistoria Syntomos) describing the events of 602–769 from an anti-Iconoclastic
Oracle of Baalbek An apocalyptic work, one of a series of supposed pronouncements of the
Sibylline oracle. This ‘oracle’ was written in the opening years of the sixth century and provides
an overview of earlier events, cast as a prophecy.
Orosius (fl. early fifth century) Spanish priest. In 417 he published, at the invitation of
Augustine, a Historia adversus Paganos – i.e. an apologetic work designed to show how much
better off the empire was since the birth of Christ.
Pacatus (late fourth century) Gallic orator and author of a panegyric of the Emperor
Theodosius I which was delivered in the emperor’s presence at Rome in
summer 389.
Parastaseis Syntomoi Chronikai Anonymous work describing the monuments of Constanti -
nople, possibly dating to the beginning of the eighth century.
Philostorgius (c.368–c.439) Greek ecclesiastical historian. His principal work, a history of
the Church from the Arian viewpoint, has come down to us mainly in an epitome by Photius
and in the historical sections of the Artemii Passio.
Priscus (c.410–c.474) Rhetorician and author of a history of the eastern Roman empire
covering the period 434–74, which survives only in fragments.
Procopius (fl. sixth century) Greek historian of the reign of Justinian. A native of Caesarea
(Palestine), he was military secretary to Belisarius and eyewitness to many of his campaigns.
Quodvultdeus (died c.453) Bishop of Carthage, correspondent of Augustine and probable
author of the Liber promissionum et praedictorum Dei (theBook of the promises and predictions of
God), completed in the mid-fifth century.

Rufinus(of Aquileia, c.345–410) Latin author and translator who translated and abridged
Eusebius’ Church History, supplemented by two books detailing the period 324–395.
Sebeos (seventh century) Traditionally identified as the author of an Armenian history which
records Near Eastern history in the period 590–655, with brief updating notices down to 661.
The work is a compilation put together by an unknown author interested in international
affairs – Byzantine, Sasanian and Islamic.
Socrates (Scholasticus) (c.380–c.439) Greek lawyer and ecclesiastical historian. He was the
author of a continuation of the Church Historyof Eusebius which extends from 305 to 439.
Sozomen (fifth century) Greek ecclesiastical historian and another continuator of Eusebius.
His work, which covers the period 323–425, is heavily dependent on that of Socrates but
contains valuable information on the history of Christianity in Armenia and the Sasanian
Stephen of Taron, also known as Aso³ik, ‘singer’ (eleventh century) Armenian historian who
wrote a Universal History around the year 1000.
Suda A Greek encyclopaedic lexicon compiled about the end of the tenth centur
AD .
Synodicon Orientale A work which contains the acts of the East Syrian Church (in late antiq -
uity, the Church in Persia) from 410 to 775. The date at which the work was compiled is uncer -
tain, but it preserves much useful documentation, including records of c
ouncils and letters.
Tabari, Abu Ja‘far Muhammad b. Jarir al- (c.839–923). Arab historian, whose principal
work contains an important monograph on the history of the Sasanian dynasty in Persia, for
which he used earlier, now lost, sources.
Themistius (c.317–c.388) Greek philosopher and rhetor whose panegyrics found favour
with every emperor from Constantius II to Theodosius I. His speeches are also important as
sources for the history of his time.
Theodore Lector (Theodore Anagnostes) (died after 527) Ecclesiastical historian from
Constantinople, who wrote a Church Historydown to 527. It now survives in fragmentary
Theodoret (c.393–c.466) Greek ecclesiastical historian. A native of Antioch, he later became
the bishop of Cyrrhus. His Church Historycontinues that of Eusebius down to 425 in a style
rather different from that of Socrates and Sozomen.
Theophanes (Confessor) A Byzantine monk who compiled a chronicle (in Greek) covering
the years 284–814. His account of Heraclius’ campaigns, often drawing on official reports and
poems of George of Pisidia, is especially valuable.
Theophanes (of Byzantium) (late sixth century) Author of a Historika, an account of the
period 566–81, whose work survives in only a few fragments.
Theophylact Simocat(t)a (fl. first half of seventh century) Greek rhetor, born in Egypt. He
served in Constantinople as a court official during the reign of Heraclius. His Historycontinues
the work of Menander, covering the period 582–602 (and some earlier events), for which he
drew on earlier sources (such as John of Epiphania).
V. Theod. Syk. TheLife of Theodore of Sykeon was composed by his disciple George, who went
on to become hegoumenos(head) of the monastery at Sykeon.
Zachariah of Mytilene (c.465/6–after 536) The Church Historyattributed to Zachariah was

compiled in Amida c.569 and contains important information on secular events in the sixth
century (as well as ecclesiastical history of the fifth and sixth centu
Zonaras, Ioannes(twelfth century) Byzantine historian and canonist, author of a universal
history (in Greek) down to 1118 which preserves fragments of many lost
Zosimus (late fifth/early sixth century) Greek pagan historian and imperial bureaucrat,
whose only surviving work, the Historia Nova, covers Roman history 180–410. The work
contains much material drawn from the now lost Historiesof Eunapius and Olympiodorus.

Map 1The Near East in late antiquity

Map 2The southern theatre of war (Mesopotamia and surrounding regions)

Map 3The northern theatre of war (Armenia and surrounding regions)


‘‘ ‘

Map 4
The western Transcaucasus (Lazica and surrounding regions)

Map 5The Transcaucasus

Peace terms agreed by Jovian during the return from Persia (363)
The Emperor Julian died from the wound he sustained in battle on 26 June 363. On the follow -
ing day a relatively unknown commander by the name of Jovian was elevated to the throne. He
continued the withdrawal from Persian territory, but, unable to cross the river Tigris, the army
halted for four days at a place called Dura. There the Persians entered into negotiations with
Jovian, who agreed to almost everything they demanded: the more time passed and supplies ran
short, the less room for manoeuvre was available to him. Thus the new emperor surrendered
several territories east of the Tigris to the Persians, as well as valuable frontier fortresses in east-
ern Mesopotamia, such as Nisibis; he also agreed not to lend assistance to the Armenian king
Arsaces. Probably on 11 July the march back to Roman territory resumed; after enduring terri-
ble privations, the remnants of the expeditionary force arrived back at Nisibis. See Chrysos
1976: 25–32, Blockley 1984: 31, 34–7, Gutmann 1991: 162–4, Winter and Dignas 2001:
The attitude of the sources to Jovian’s peace ranges from utter condemnation to an accep-
tance that he accomplished all that was possible under the circumstances; Christian writers
naturally shift the blame for the surrender of Roman territory onto Julian.
(a) Sources hostile to the peace
Agath. IV.26.6–7: see volume 1.
Ammianus XXV.7.1–8.7: While these vain attempts were going on, king Shapur,
both while at a distance, and also when he had drawn near, received from his scouts
and from our deserters a true account of the gallant exploits of our men, of the
disgraceful massacres of his own troops, and of the killing of his elephants, in numbers
he did not recall (being lost before) during his reign. (And he heard) also that the
Roman army, being hardened by its continual labours since the death of its glorious
chief, did not now think so much, as they said,
3of safety as of revenge, and (was
resolved) to put an end to the distress of their current situation either by a complete
victory or by a glorious death. (2) These many fearsome matters made him reflect, for
he was aware by experience that our troops who were scattered in large numbers over

these provinces could be assembled by a simple password, and knowing also that his
own men after their heavy losses were in a state of the greatest alarm; at the same time
he also heard that we had left behind in Mesopotamia an army little inferior in
numbers to that before him. (3) Above all (else), the thing which reduced his anxious
mind to inactivity was the fact that 500 men had safely at one moment crossed that
swollen river
4by swimming, and having slain his guards, had emboldened the rest of
their comrades to similar boldness. (4) In the meantime, as the violence of the stream
prevented any bridges from being constructed, and as everything which could be
eaten was consumed, we passed two days in great misery; and so the soldiery was
raging, roused by starvation and anger, and was in a hurry to perish by the sword
rather than by hunger, the most degrading death. (5) But the eternal power of the
heavenly god was on our side and beyond our hopes the Persians made the first over -
tures, sending Surena
5and another noble as ambassadors to establish peace; they
themselves were in a state of despondency, as the Romans, having proved superior in
almost every battle, weakened them daily. (6) But the conditions which they proposed
were difficult and intricate, since they pretended that, out of regard for humanity,
their most merciful monarch was willing to permit the remains of our army to return
home, provided the Caesar, with his officers, would satisfy his demands.
6(7) In reply,
Arintheus was sent with the prefect Salutius; and while the proper terms were being
discussed with great deliberation, four more days were spent in great suffering from
want of provisions, more painful than any kind of torture. (8) And in this space of
time, if before the ambassadors were sent, the emperor, making full use (of the oppor-
tunity), had retired slowly from the territories of the enemy, he would certainly have
reached the forts of Corduene, a rich region belonging to us, only 100 miles from the
spot where these transactions were being carried on.
7(9) But Shapur obstinately
demanded (to use his own language) the restoration of those territories which had
been taken from him a short while ago by Maximian;
8but as was seen in the process of
the negotiation, he in reality required, as the price of our redemption, five provinces
on the other side of the Tigris – Arzanene, Moxoene, Zabdicene, as well as Rehimene
and Corduene, with fifteen fortresses and Nisibis and Singara, and Castra
9an extremely well placed fort. (10) And though it would have been better
to fight ten battles than to give up any of them, still a set of flatterers harassed our
timid emperor with harping on the dreaded name of Procopius, and affirmed that if
he (Procopius) returned with the fresh troops under his command when he heard of
the death of Julian, he would easily bring about a revolution which no one would
10(11) Jovian, excessively 11inflamed by the pernicious reiteration of these evil
counsels, without further delay gave up everything that was demanded, (with this
proviso), which he obtained with difficulty, that Nisibis and Singara should go over to
Persian control without their inhabitants, and that the Roman garrisons in the forts to
be surrendered should be permitted to retire to fortresses of our own. (12) To this
another deadly and unfair condition was added, that after this treaty was concluded
aid was not to be brought to Arsaces against the Persians, if he implored our aid,
though he had always been our friend and trusty ally. And this was insisted on by
Shapur for two reasons, in order that the man might be punished who had laid waste

Chiliocomum 12at the emperor’s command, and also that the opportunity might
henceforth be given for invading Armenia without a check. In consequence of this it
fell out subsequently that Arsaces was captured alive, and that amid different dissen -
sions and disturbances, the Parthians laid violent hands on the greater portion of
Armenia, where it borders on Media, and on the town of Artaxata. (13) Once this
ignoble treaty had been made, in order that nothing might be done during the armi -
stice, in contravention of its terms, some men of rank were given as hostages on each
side: on ours, Nemota,
13Victor and Bellovaedius, tribunes of distinguished legions:
and on that of the enemy, one of their chief nobles named Bineses, and three other
satraps of note. (14) So peace was made for thirty years, and ratified by solemn oaths;
and we, returning by another line of march, avoiding the parts near the river because
they were rugged and difficult, suffered severely for want of water and
(8.1) The peace which had been granted on pretence of humanity was turned to the
ruin of many who were so exhausted by want of food as to be at the last gasp, and who
in consequence had set off in secret, and were carried away by the current of the river
from not being able to swim; or, if able to overcome the force of the waves so far as to
reach the bank, they were caught by the Saracens or the Persians, whom (as we said a
little earlier) the Germans had driven out, and slain like cattle or sent off a distance to
be sold as slaves. (2) But when the resounding trumpets openly gave the signal for
crossing the river, it was with a terrible ardour that every individual hastened to rush
into all kinds of danger, preferring himself to all his comrades, in the desire to avoid
many dangers. Some tried to guide the beasts which were swimming about at random
with paddles hurriedly put together; some, seated on bladders, and others, being
driven by necessity to all kinds of expedients, sought to pass through the opposing
waves by crossing them obliquely. (3) The emperor himself with a few others crossed
over in the small boats, which we said were saved when the fleet was burnt, and then
ordered the same vessels to go backwards and forwards until we were all brought
across. And at length all of us, except such as were drowned, reached the opposite
bank of the river, being saved amid our difficulties by the favour of the supreme deity. (4) While fear of impending disasters oppressed us, we learnt from information
brought in by our scouts that the Persians were throwing a bridge over the river
beyond our sight, so that after peace was established and treaties (concluded), and the
turmoils of war had been stilled, they might come upon our invalids as they proceeded
carelessly onwards, and on the animals long exhausted with fatigue. But when they
found their purpose discovered, they relinquished their base design. (5) Being now
relieved from this suspicion, we hastened on by rapid marches, and approached Hatra,
an ancient town in the middle of a desert, which had been long since abandoned,
though at different times the warlike emperors, Trajan and Severus, had attacked it
with a view to its destruction, but had been almost destroyed with their armies, which
we have also related in our history of their exploits.
15(6) And as we now learnt that
over the vast plain before us for seventy miles in that arid region no water could be
found but such as was brackish and fetid, and no kind of food but southernwood,
wormwood, dracontium, and other bitter herbs, we filled the vessels which we had
with sweet water, and after the camels and the rest of the beasts of burden had been

slain, other types of food, even harmful ones, were sought. 16(7) For six days the army
marched, until at last even grass, the last comfort of extreme necessity, could not be
found; then Cassianus, duxof Mesopotamia, and the tribune Mauricius, who had
been sent forward some time ago with this object, came to a Persian fort called Ur,
and brought some food from the supplies which the army under Procopius and
Sebastian, by living sparingly, had managed to preserve. (tr. Yonge, revised)
Eunapius frg.29.1.3–6, 10–15: ( … ) Coming to Nisibis, a populous city, he (Jovian)
stayed there for only two days, lavishly consuming all its resources and having neither
a kindly word nor a good deed for the inhabitants.(…)When, as has been said, he
became emperor of the Romans after Julian, ignoring everything else in his eagerness
to enjoy the rank that had devolved upon him, he fled from Persia, hurried to reach
the Roman provinces to display his (good) fortune, and handed over to the Persians
the city of Nisibis, which had long been subject to the Romans.
18 (tr. Blockley,
Festus, Breviarium 29:Jovian took over an army (which had proved) superior in bat -
tles, but which had been thrown into confusion by the sudden death of the emperor it
had lost. When supplies were running short and the way back threatened to be rather
long, the Persians held up the passage of the columns with frequent attacks, at one
moment from the front, at another from the rear, as well as attacking the flanks of
those in the middle; after some days had been spent (thus) such was the respect for the
Roman name that it was from the Persians that the first mention of peace came; and
the army, exhausted by starvation, was allowed to be led back. The conditions
imposed – (something) which had never previously occurred – were costly to the
Roman state, as Nisibis and part of Mesopotamia were handed over; Jovian, new to
the throne, and greedier for power than glory, assented to them.
Libanius, Or.1.134 (I, 147.25–148.5): (…)Wewere sure that by these great afflic-
19heaven (p.148) gave us a sign of some great disaster, and, as we prayed that our
guess should not be right, the bitter news reached our ears that our great Julian was
being carried in his coffin, that some nonentity held the throne, and that Armenia,
and as much of the rest of the empire as they liked, was in Persian hands.
Libanius, Or.18.277–8 (II, 357.8–358.4): (Libanius is complaining of the change in
Roman policy once Jovian was on the throne) In the first place, they did not stand up
to those whom previously they used to put to flight; secondly, ensnared by this word
‘peace’, for the enemy applied the same technique again, they all demanded its accep -
tance without demur, and the new emperor was the first to be taken in by it. The
Mede found them hankering after peace and dillied and dallied with question and
answer, accepting this point, deferring that, and exhausting their supplies with a string
of parleys. (278) When they were in want of food and everything else, and they were
begging for everything, and dire necessity encompassed them, at that moment (p.358)
he (the Persian) presented his minimum terms, the cities, territories and provinces –
the (very) walls (which guaranteed) the safety of the Romans. Our new emperor
agreed and evacuated them all, and made no bones about it. (tr. Norman,

(b) Sources which are balanced in their attitude to the peace
Artemii Passio70:see volume 1.
Cedrenus I.539.16–21: see volume 1.
Chr. 724, 133.25–134.9: see volume 1.
Epic Histories IV.21:(describing the treaty of 363)(…)Andhewrote in the treaty in
the following way: ‘I have granted you,’ he (Jovian) said: ‘the city of Ncbin (Nisibis),
which is in Aruestan and Syrian Mesopotamia. And I am withdrawing from the
middle country of Armenia. If you will be able to overcome and subject them, I shall
not support them.’ The king of the Greeks was then in a difficult situation, and
(being) in an uncomfortable position, he in this manner sealed a deed of the words;
and he gave (the agreement) to the king of Persia and so freed himself from hi m(…).
(tr. Greenwood)
Eutropius, Breviarium X.17.1:After this, Jovian, who was then serving as a
domesticus, was chosen to take over the empire by the general consensus of the army;
he was known to the soldiers more by the reputation of his father than by his own
merits. Since matters were already in a troubled state and the army suffering from a
lack of supplies, and after he had been defeated in one battle, then another,
21he made
a peace with Shapur which was necessary indeed, but shameful; he was punished in
regard to the frontier, and several parts of the Roman empire were hande
d over.
Josh. Styl. 7 (242.2–8): After the death of Julian in Persia, which took place in the
year 674 (362/3), Jovinian (i.e. Jovian), who reigned over the Romans after him,
valued peace more than anything; and for the sake of this he allowed the Persians to
have authority over Nisibis for 120 years, after which they were to restore it to its mas-
22( … ) (tr. Wright, rev. M. Greatrex)
Orosius, Hist.VII.31.1–2: In the 1117th year after the foundation of Rome, Jovian
was made the 37th emperor at a critical juncture of events. When, trapped by unfa-
vourable terrain and surrounded by the enemy, he could find no opportunity for
escape, he agreed a treaty with the Persian king Shapur, which, although some con-
sider it unworthy, was nevertheless quite necessary; (2) and so, in order that he might
rescue the Roman army safe and unharmed not only from the attack of the enemy, but
also from the danger of the territory, he ceded the town of Nisibis and part of Upper
Mesopotamia to the Persians.
Ps.-Dion. I, 180.4–8: He (Jovian) made peace between the two empires and ceded
Nisibis to the Persians. The persecution came to an end in Persia because of the peace
he made and all the churches were (re)opened. All the inhabitants of Nisibis went into
exile to Amida in Mesopotamia and he constructed walls for them west of the city. (tr.
Lieu, rev. Brock)
Rufinus, HEXI.1 (1001.10–1002.6): ( … ) In fact divine clemency immediately
attended him (Jovian) against all (p.1002) hope, since they (the soldiers) were held (in
a position) closed in on all sides by the enemy, and no means of escape was available.
Suddenly they beheld spokesmen sent by the barbarians to seek peace, promising vict -
uals and other necessities to the army, (which was) stricken by starvation, and

repaying the boldness of our men with every kindness. But when peace was made for
29 years, he returned to Roman soil ( … ).
Socr.HEIII.22.5–7 (218.11–18): The cry of all in unison therefore acknowledged
that they were Christians; he (Jovian) accepted the throne, (6) suddenly taking over in
a crisis, in Persian territory, with the soldiers perishing from hunger, and he brought
the war to an end by a treaty. The treaty was shameful to Roman glory, but necessary
in the situation. (7) For he received a fine in regard to the boundaries of the empire
and handed over to the Persians Nisibis in Mesopotamia, city>; he then departed from there.
Soz. HEVI.3.2 (239.17–20): Since matters were in a dangerous and confused state as
a result of Julian’s expedition
23and the army suffering from a lack of provisions, he
(Jovian) saw that it was necessary to come to terms and handed over to the Persians
one of the (territories) previously tributary to the Romans.
Zonaras XIII.14.4–6 (217.3–13): He (Jovian) thus accepted the title of emperor, and
made a treaty with the Persians which was not befitting to the Romans, but was made
under compulsion. (5) He conceded two famous cities to them, Nisibis and Singara,
and transferred the inhabitants of them elsewhere, by whom, in their grief, he was vio -
lently abused. (6) He abandoned to them (the Persians) many provinces and rights
which had belonged to the Romans for a long time. When the hostages had been
handed over by each side, the treaty was confirmed. The Romans then moved off, but
endured a shortage of provisions, for they did not even have enough water. (tr.
Dodgeon, revised)
Zos. III.31.1–2: Although the (Roman) army was in these circumstances, the Persians
nonetheless entered negotiations about friendship and sent out Surena and others
from those in power among them. Jovian received their words about peace and des-
patched Salutius the praetorian prefect and with him Arintheus. After talks between
the two sides had taken place concerning this matter a treaty for thirty years was con-
cluded. It was agreed that the Romans should cede the province of Zabdic
ene to the
Persians, as well as Corduene, Rehimene and Zalene;
24and in addition to these and in
all these (regions, they were to hand over) the fortresses in them, fifteen in number,
with their inhabitants, possessions, animals and all their equipment, while Nisibis
should be handed over without its inhabitants. For it was determined that they would
be resettled where the Romans decided. (2) The Persians also took most of Armenia,
allowing the Romans to hold a small part of it. Once the treaty had been concluded on
these terms and had been confirmed with contracts by both sides (the Persians) gave
the Romans a free path for their return home; (for it was understood that the Romans)
would not harm the lands of the Persians, and that they (in return) would not be the
victims of Persian ambushes.
(c) Sources favourable to the peace
Augustine, De civ. deiIV.29 (123.37–43): The army, deprived of it (food), and after
even (Julian) himself had swiftly been slain by the wound (inflicted) by an enemy, was
reduced to great scarcity. And no one would have escaped, for from all sides the enemy

was attacking the soldiery, which was distressed by the death of the emperor, had not
peace been agreed and the frontiers of the empire fixed in that place where they remain
today; they were settled without such a great loss as Hadrian had conceded,
rather by a compromise agreement.
Gregory Nazianzenus, Or.5.15:The one who immediately succeded to the throne
after that man, and after that man was proclaimed in the camp itself – and the great
danger absolutely required a leader – was a man outstanding, among other things, for
his piety, and in appearance was truly worthy of supreme power. He was utterly
unable to engage the Persians or to move forward, and yet he in no way lacked courage
or zeal: the army had thrown up its hands and thrown away its hopes, but he sought to
break up camp and he looked for any means by which he could safely accomplish this,
since he had become the inheritor not of the empire, but of a defeat. Had not therefore
the Persians, moderate in victory – for they have a law to know how to measure success
– or through fear of one of the other things talked of,
27come to terms, which were so
unexpected and humane, nothing would have prevented not even a firebearer being
left to the army, as they say.
28To such an extent did the Persians have them (the
Romans) in their grip, fighting on home territory and encouraged yet more by the
events that had taken place; for it is sufficient to experience a piece of good luck to give
hope for the future. But now, as I said, it was the task of one man to rescue the army
and to leave the Romans their sinews: for they (the soldiers) were the sinews, even if
they had fared badly more by the thoughtlessness of their general than by their own
lack of courage. They (the Persians) came to an agreement on these terms, so disgrace-
ful and unworthy of Roman might, to put it very shortly. If anyone would absolve that
man (Julian) of responsibility for them, and blame this man (Jovian), I would say he is
a senseless reckoner of the events which took place then. ( … )
Jerome, Chron.a.364:see volume 1.
Joh. Chrys., De sancto Babylo 123.7–26:( … ) Therefore when he (Julian) had fallen
so shamefully, and the soldiers saw themselves in dire straits, they prostrated them-
selves at the feet of their enemies and gave oaths that they would withdraw from the
safest place of all, which was like an impregnable wall for our inhabited world.
(Because) they happened (to be dealing with) humane barbarians, they escaped in this
way and returned – few out of many, and even they, after physical sufferings, in shame
(at the terms) according to which they had been obliged by their oaths to withdraw
from their ancestral possessions. It was possible to see a sight more pitiful than any
capture in war. For those who lived in that city (Nisibis) were suffering at the hands of
those from whom they expected to receive thanks because they had kept (safe) all
those within as in a harbour, just like a breakwater, and because always, in place of
everyone, they had been exposed to every danger. (Now) they were being transferred
to another’s country, giving up their houses and fields, dragged from their ancestral
possessions – and they endured these things at the hands of their own people. Such
were (the benefits) we had enjoyed from the noble emperor.
Mal. 13.27 (258.73–259.93/335.1–336.5): 30When Jovian went out with his army
from the desert to the fertile Persian land, he considered anxiously how he might go
out from Persian territory. Now the Persian king Sabburarsakios (Shapur) had not yet

learnt of the emperor Julian’s death and was filled with terror. Pleading and begging
for peace, he sent from Persarmenia one of his nobles, named Surraeina (Surena), as
an ambassador to the Roman emperor. The most sacred emperor Jovian welcomed
him cordially and consented to receive (p.259) the peace embassy, saying that he too
would send an ambassador to the Persian king. ambassador,> asked the emperor Jovian to agree on a peace treaty there and then.
Selecting one of his senators, the patrician Arintheus, he entrusted the whole affair to
him. He agreed to abide by the terms approved or agreed by him, for the emperor was
too proud to make a peace treaty with the Persian senator or ambassador; and he pro-
vided a truce in the war for three days during the peace negotiations. It was agreed
(p.336) between the Roman patrician Arintheus and Surena, senator and ambassador
of the Persians, that the Romans should cede to the Persians the entire province
known as Mygdonia and its metropolis known as Nisibis, just the city with its walls
alone, without the people who inhabited it. (tr. Jeffreys and Scott, re
Philostorgius, HEVIII.1 (104.2–7): On the next day after the apostate’s (Julian’s)
death, the army appointed Jovian emperor. He – for there was no other way to safety,
since the whole army had been reduced to one tenth (its orginal size) – made a treaty
with the Persian for thirty years, giving up to him all claim to Nisibis and the fortresses
which had been erected by the Romans as a barrier against the Persians as far as Arme-
nia. (tr. Walford, revised)
Themistius, Or.5.66a (96.15–19): That the Persians voted for you (Jovian) no less
than the Romans they showed by throwing (down) their weapons when they heard
about the public proclamation; and they who had felt confident against them (the
Romans) soon afterwards grew more respectful.
Thdrt. HEIV.2.2–3 (211.18–24): ( … ) Immediately the God of all showed him
(Jovian) his consideration and solved the problem which had arisen. For the Persian
king, learning of his accession to the throne, sent envoys on an embassy about peace.
He then despatched provisions to the soldiers and ordered that a market-place be set
up for them in the desert. (3) When he had made a thirty-year treaty, he let the army
go in good health.
31( … )
Perhaps the most favourable version of all comes in the so-called Julian Romance, a work com-
posed in Edessa probably in the late fourth century. 32
Julian Romance, 220.17–24: They (the leading Romans and Persians) received ideas
from each other, and came from disputation to a decision. There was great dispute
between them concerning what had been done illegally in the territory of the two
sides. They considered the countries which Shapur had laid waste in the Roman terri -
tory in the days of the sons of Constantine, and the places which Julian had destroyed
in the land of the Persians. There was less advantage on the side of the Romans than
on that of the Persians;
33and Nisibis was given to the Persians for a hundred years
with all its land on its eastern side. That was done voluntarily and without compul -
34(tr. Gollancz, rev. M. Greatrex)
Julian Romance, 224.22–225.2: They (Jovian and Shapur) prepared a document on

the frontier regarding Nisibis and the peace of the churches in Persia (to last for) a
hundred years according to the contract which came from the kings. And Nisibis was
surrendered to the Persians, devoid of its inhabitants, in the month of August in the
year 674 (363), according to the chronology of the Greeks. Jovian was satisfied in his
mind that, by making peace, he would be able by his skill to rescue his troops (from)
the imprisonment in which they found themselves. He had not yet (p.225) set foot
upon the territory of his kingdom when the news of the salvation of the churches was
sped throughout all the lands of the Romans. (tr. Gollancz, rev. M. Gre
Jovian himself took pains to cast his return from Persia in the best possible light. He despatched
supporters to the west as soon as he could, in order to consolidate his position partly by repre-
senting the outcome of the expedition as a success (Ammianus XXV.8.8–12). This propaganda
effort is reflected in the coins issued by Jovian, which proclaimed ‘Victory of the Romans’ and
‘Victory of Augustus’; solidiissued at Thessalonica and Sirmium referred to the ‘Safety of the
State’ and depicted the emperor holding the Christian labarumover a Persian prisoner.
(d) The Persian tradition
Tabari, I, 843/62–3 (Nöldeke 62–3): While Jovinianus (Jovian) was quite willing to go
to Shapur, none of his commanders favoured this intention. He, however, insisted and
came to Shapur, with eighty of the most respected men in the camp and in the whole
army, and wearing the crown. When Shapur heard that he was coming he went out to
meet him. They both fell on the ground (in obeisance) before each other; then Shapur
embraced him in gratitude for what he had done for him. He ate a meal with Shapur
and was of good cheer. Shapur meanwhile, however, gave the Roman leaders and gover-
nors to understand that if they had made any other than Jovinianus to be emperor, they
would all have perished in Persia; it was only because of his nomination that he did not
let them feel his power. So through his efforts Jovinianus’ position became very strong.
Then he said: ‘The Romans attacked our land, killed many people, chopped down date-
palms and other trees in Sawad and devastated agriculture; now, they shall either pay us
the full price for what they destroyed and laid waste, or give up the town Nisibis with its
territory as compensation.’ Jovinianus and his military leaders agreed to give Shapur
compensation and yielded Nisibis to him. When the inhabitants of this town heard of
it, they emigrated to other places in the Roman empire because they feared the reign of a
king who was of a different religion.
36When Shapur heard this, he had 12,000 people of
good family from Istakhr, Ispahan and other regions of his lands sent to Nisibis and
settled them there.
37(tr. A. al-Issa and D. Dance)
Tabari’s evidence may be supplemented by a rock relief at Taq-i Bustan which, it is argued,
depicts the investiture of Shapur II, who triumphs over the Emperor Julian. See Trümpelmann
1975, Azarpay 1982: 184–7 (emphasising the role of Mithras, as god of boundary-lines, in pun -
ishing Julian), Nicholson 1983 and Schippmann 1990: 35. The famous Paris cameo, depicting
a Sasanian king grasping the left arm of a Roman emperor, may also be relevant. Although the
cameo is usually associated with Shapur I and Valerian (so, e.g. Fowden 1993: 23 plate 1),
others have attributed it rather to Shapur II’s defeat of Jovian. See
von Gall 1990: 57–9.

The aftermath of the peace treaty
The surrender of Nisibis in July 363 was a dramatic event, as emerges from the following
accounts, several by eyewitnesses. Many of the displaced inhabitants settled in Amida, which
therefore grew considerably in size and required extensive refortificati
Ammianus XXV.8.13–17; 9.1–6, 12:(Jovian sends out messengers to announce his
accession and to report his peace treaty in a favourable light) (13) But Rumour (being
always the most rapid bearer of bad news), outstripping these couriers, flew through
the different provinces and nations, and above all others struck the citizens of Nisibis
with bitter sorrow when they heard that their city was surrendered to Shapur, whose
anger and enmity they dreaded from recollecting the losses he had regularly endured
in his frequent attempts to take the place. (14) For it was clear that the whole eastern
empire could have fallen under the power of Persia, if it had not been for the resistance
which this city, with its admirable position and its mighty walls, had offered. But
miserable as they now were, and although they were filled with a still greater fear of
what was to come, they were supported by this slender hope, that, either from his own
inclination or from being won over by their prayers, the emperor might consent to
keep their city in its existing state, as the strongest bulwark of the East. (15) While
different reports were flying about of what had taken place in the army, the scanty
supplies which I have spoken of as having been brought were consumed, and necessity
might have driven the men to (eat) the bodies of their own (dead), if the flesh of the
animals slain had not lasted them a little longer; but the consequence of our destitute
condition was that most of the arms and baggage were thrown away; for we were so
worn out with this terrible famine that whenever a single bushel of corn was found –
which seldom happened – it was sold for ten pieces of gold at the least. (16) Marching
on from there, we came to Thilsaphata,
39where Sebastian and Procopius, with the
tribunes and chief officers of the legions which had been placed under their command
for the protection of Mesopotamia, came to meet the emperor as the solemn occasion
required, and being kindly received, accompanied us on our march. (17) After this,
proceeding with all possible speed, we rejoiced when we saw Nisibis, where the
emperor pitched a standing camp outside the walls; and being most earnestly
entreated by the whole population to come to lodge in the palace according to the
custom of his predecessors, he obstinately refused, being ashamed that an impregnable
city should be surrendered to an enraged enemy while he was within its w
alls. (9.1) The next day Bineses, one of the Persians of whom we have spoken as the
most distinguished among them, hastening to execute the commission of his king,
demanded the immediate performance of what had been promised; and by the per -
mission of the Roman emperor he entered the city and raised the standard of his
nation on the citadel, announcing to the citizens a miserable emigration from their
native place. (2) Immediately they were all commanded to leave the place, weeping
and stretching forth their hands in entreaty not to be compelled to depart, affirming
that they by themselves, without either state provisions or soldiers, were sufficient to
defend their own home in full confidence that Justice would be on their side while

fighting for the place of their birth, as they had often found her to be before. Both
nobles and common people joined in this supplication; but they spoke in vain as to
the winds, the emperor fearing the crime of perjury, as he pretended, though fearing
other things. (3) Then Sabinus, (a man) eminent among his fellow-citizens both for
his fortune and birth, replied with great fluency that Constantius too was at one time
defeated by the Persians in the terrible strife of fierce war, that in his flight he was led at
last with a small body of comrades to the ungarded station of Hibita,
40where he lived
on a scanty and uncertain supply of bread which was brought him by an old woman
from the country; and yet that to his last day he lost nothing; while Jovian, at the very
beginning of his reign, was yielding up the wall of his provinces, by the protection of
which barrier they had hitherto remained safe from the earliest ages. (4) But as noth -
ing was accomplished, for the emperor persisted stubbornly in alleging the obligation
of his oath, presently, when Jovian, who had for some time refused the crown which
was offered to him, accepted it under a show of compulsion, an advocate named
Silvanus, exclaimed boldly, ‘May you, o emperor, be so crowned in the rest of your
cities.’ But Jovian was offended at his words, and ordered the whole body of citizens to
leave the city within three days, complaining as they were at the existing state of
affairs. (5) Accordingly, men were appointed to compel obedience to this order, with
threats of death to everyone who delayed his departure; the walls were disturbed by
mourning and lamentation, and in every quarter of the city the common sound from
all was of wailing, matrons tearing their hair when about to be driven from their
homes, in which they had been born and brought up, the mother who had lost her
children, or the wife her husband, about to be driven far from the place of the spirits of
their dead; the piteous crowds wept, embracing their doorways or thresholds. (6)
Every road was crowded, each person escaping as he could. Many, too, took off with as
much of their property as they thought they could carry, while casting aside abundant
and costly furniture, for this they could not remove for lack of beasts of burden .(…)
(12) Accordingly, when the citizens had been withdrawn, the city surrendered, and
the tribune Constantius had been sent to deliver up to the Persian nobles the garri -
soned forts and districts agreed upon, Procopius was sent forward with the remains of
Julian, to bury them in the suburbs of Tarsus, according to his directions while alive.
(tr. Yonge, revised)
Ephrem Syrus, Hymni contra Iul. II.22, 25 (80.9–14, 80.27–81.4): The Magian
(Shapur II) who entered our place, kept it holy, to our shame, he neglected his temple
of fire and honoured the sanctuary, he cast down the altars which were built through
our laxity; he abolished the enclosures, to our shame, for he knew that from that one
temple alone had gone out the mercy which had saved us from him three ti
(25) That city (Nisibis) which was the head of the area between the rivers preserved
the sack-cloth of the blessed one and was exalted. The tyrant by his blasphemy had
abased it and it was humbled. Who has weighed its shame, how great it was! For the
city which was the head of all that West they have made the last heels of all that East.
(tr. J. Lieu)
Ephrem Syrus, Hymni contra Iul. III.1–2 (81.19–82.3): A fortuitous wonder! There
met me near the city the corpse of that accursed one which passed by the wall;

banner which was sent from the East wind the Magian took and fastened on the tower
so that a flag might point out for spectators that the city was the slave of the lords of
that banner. Response: Praise to him who clothed his corpse in shame. (2) I was
amazed as to how it was that there met and were present the body and the standard,
both at the same time. And I knew that it was a wonderful preparation of justice
(p.82) that while the corpse of the fallen one was passing, there went up and was
placed that fearsome banner so that it might proclaim that the injustice of his diviners
had delivered that city. (tr. J. Lieu)
Mal. 13.27 (259.93–15/336.5–337.2):When this had been confirmed and the peace
treaty committed to writing, the emperor Jovian took with him one of the satraps, a
Persian named Junius, who was with the ambassador, to conduct him and his army
safely from Persian territory, and to take over the province and its met
ropolis. On reaching the city of Nisibis, the emperor Jovian would not enter it, but
encamped outside the walls. But Junius, the Persian satrap, entered the city at the
emperor’s command, and set up a Persian standard on one of the towers, since the
Roman emperor had ordered that all the citizens to the last man were to depart with
all their possessions. Silvanus, a comesin rank and a magistrate of the city, came out
and threw himself down before the emperor, beseeching him not to surrender the city
to the Persians, but did not persuade him. For he said that he had taken an oath, and
added that he did not wish to have a reputation among all people as a perjurer. Then,
having built a walled city outside the wall of the city of Amida, to the wall of the city of Amida and> called it the town of Nisibis, and made all the
people from the (p.337) area of Mygdonia live there, including the magistrate
43(tr. Jeffreys and Scott, revised)
Zos. III.33.1–34.2: Once peace had been concluded with the Persians in the way I
have described, the Emperor Jovian returned with his army in safety; he encountered
many rugged places and waterless regions and lost many of his forces in his journey
through enemy country. He ordered Maurice, one of his tribunes, to bring food for
the army from Nisibis and to meet him with this as far off (from Nisibis) as he could;
others he sent off to Italy to announce the death of Julian and his own accession. (2)
With difficulty and much distress he drew near to Nisibis, but not wishing to be in the
city which had been handed over to the enemy, he set up camp in an open enclosure in
front of the gate. On the next day he received crowns, along with supplications; and all
those in the city entreated him not to abandon them and not to subject them to the
experience of barbarian practices – they who had been brought up willingly under
Roman laws for so many years. (3) (They pointed out that) it was disgraceful besides
that while Constantius had taken on three Persian wars and had been beaten in all of
them, he had retained hold of Nisibis and with much effort kept it safe, even when it
was under siege and came into the gravest danger. Yet he, when no such exigency
existed, (proposed) to hand over the city to the enemy and to show the Romans a day
which they had never witnessed, (when they would be) obliged to allow such a great
city and (its) territory to be bestowed on an enemy. (4) The emperor, hearing this,
pointed to what had been agreed; (but) Sabinus, the leader of the city council, added
another argument to those put forward by the people who had come in supplication.

(He asserted that) they had no need of any expenses for their war against the Persians,
nor of any help from outside; they would be able to sustain the war which would
engulf them with their own bodies and funds. And when they won, the city would
again be subject to the Romans, fulfilling their requests in the same way as before. (5)
But once the emperor had said that it was not possible to breach what had been
agreed, men from the city frequently besought him not to deprive the Roman empire
of such a frontier bastion. (34.1) Since they did not achieve anything else – the
emperor withdrew in anger, while the Persians wanted to take possession of the prov-
inces, the fortresses and the city according to the treaty – the inhabitants of the prov -
inces and the fortresses, if they had been unable to flee secretly, gave themselves up to
the Persians to treat them as they wished. The Nisibenes obtained a truce for their
move and many, (in fact) almost all, emigrated to Amida, while a few settled in other
44(2) Everywhere was filled with wailing and lamentation, since it was believed
that with Nisibis handed over to the Persians, every city lay exposed to their attacks.
So great was the grief of the residents of Carrhae when news of Julian’s death was
brought that they stoned the one who reported the news, and erected a huge pile of
stones on top of him. So great a change did the death of one man bring to public
Roman inability to upset the status quo (March 368) 45
From the moment of his accession in 364, Valens was forced to deal with matters in western
Asia Minor and the Balkans; hence Themistius, in a speech for the emperor Valens, plays up the
undesirability of wars of revenge. See Turcan 1966: 882–3, Blockley 1992: 30, Vanderspoel
1995: 170–1, Greatrex 2000a: 36.
Themistius, Or.8.114c (172.11–18): So that I may pass over other matters, I declare
that even this (point) alone is sufficient to warrant one prayer (of thanks), and its ben-
efit reaches everyone: this is that Mesopotamia has not been recovered, the Scythians
beyond (the river Danube) have not been punished, and the Germans have not rebuilt
the cities which they sacked. For even if we gained the first (region), only the Syrians
would notice; and if the other one, (only) the Thracians (would notice), or (if) the
other one (again), (only) the Gauls, and (likewise only) the country adjacent to each
victory (would notice).
Valens’ arrival in Antioch (370)
Mal. 13.29–30 (261.39–48/338.8–19): He (Valentinian I) marched against the Per -
sians, sending his brother Valens, whom he had made Caesar on 1st April; Valens did
not fight but marched out and made a peace treaty, for he had full power as the repre -
sentative of his brother. The Persians had come and sued for peace. (30) Thus, having
arrived at Antioch in Syria with the greater part of the military forces on 10 November
in the 14th indiction,
46Valens lingered there to conclude the peace treaty with the

Persians. He negotiated a treaty for seven years, with the Persians suing for peace and
ceding half of Nisibis. 47(tr. Jeffreys and Scott, revised)
Zos. IV.13.1–2: While matters were in this state in the West, the Emperor Valens set
out for the East against the Persians, something he had been intending from the outset
(of his reign).
48He advanced slowly, providing the help that was necessary for the
cities which sent ambassadors and arranged everything else appropriately, readily
bestowing on those with just requests what they sought. (2) Having arrived in
Antioch, he made ready the (tools) of war with all caution and spent the winter in the
palace there. In spring he departed for Hierapolis and from there led forward the
troops against the Persians; and when winter arrived, he went back to Antioch. The
war against the Persians was then prolonged and while the emperor was at Antioch, a
case of a surprising nature arose, for this reason.
Tensions in Mesopotamia (373) 50
Only ten years after Julian’s disastrous defeat – and despite the thirty years’ peace signed by
Jovian – the Romans were apparently making military preparations in t
he East. 51
Themistius, Or.11.148d (224.11–16): We see the emperor (Valens) displaying
judgement in these matters not only in private affairs, but also in public. For why did
he agree to peace with the Scythians and withhold it from the Persians? For both bar-
barian tribes are hostile to the Roman empire, but the one is spirited and unintelli-
gent, while the other is treacherous and deceitful.
The refortification of Amida (367/75)
CIL III.213 (= 6730): Under the emperors Valentinian, V[alens and] Gratian, pre-
served (by God), perpetual [leaders?] and triumphators, eternally Augusti, the city was
built from its foundations by the direction of their piety.
We may note here the likely construction of a fort at this time at Pagnik Oreni, usually identi -
fied with Roman Dascusa, to the north-east of Melitene, guarding a crossing of the Euphrates
from Armenia. Whether it was built to defend against the Persians, as they sought to take over
Armenia, or against Hunnic raids, is unclear. See Gregory 1997: I, 214,
II, 46–8.
The rising of Mavia (c.376)
By the late fourth century the Arabs, or Saracens as they were now often known, were emerging
as a significant force on the eastern frontier. 53One instance of this importance is provided by
the revolt of Mavia late in Valens’ reign; it followed the death of the king of the tribe, proba -
bly that of the Tanukhids, and was only brought to a halt by diplomatic means. Queen Mavia
was conciliated by the appointment of a bishop not tainted by the emperor’s Arianism, and
her daughter was married to a high-ranking commander in the East, Victor. But the rebellion
had inflicted considerable damage on the eastern provinces, and the Roman inability to deal

with the situation was an ominous precedent for future relations with Arab tribesmen. See
Bowersock 1980: 477–95 (378), Sartre 1982a: 140–4 (c.376), Shahîd 1984: 140–69 (dating
the revolt to 375, 183–4), Graf 1989: 348–9, Gutmann 1991: 185–7 and Woods 1998:
RufinusHEXI.6 (1010.12–1011.2): While Lucius was behaving with great arro -
gance and savagery, 55Mavia, queen of the race of the Saracens, began to disturb the
towns and cities of the Palestinian and Arabian frontier in a fierce war, and at the same
time to lay waste the neighbouring provinces. And when she had worn down the
Roman army with frequent battles, and after she had slain many and turned the rest
(p.1011) to flight, she was asked for peace; she promised that she would not embrace
peace unless a certain monk, Moses by name, should be ordained bishop for her tribe.
Socr. HEIV.36.1–2 (270.25–271.3): When the emperor had left Antioch, the Sara -
cens, who had previously been bound by treaty, then revolted, under the command of
Mavia, a woman, (p.271) whose husband had died. (2) Everything in the East was
therefore laid waste by the Saracens simultaneously. But some divine providence
restrained the (attacks) of the Saracens by the following means.
Soz. HEVI.38.1–5 (297.12–298.4): Around this same time, following the death of
the Saracen king, the treaty with the Romans was dissolved; Mavia, the wife of this
man, gained control of the nation and laid waste the cities of Phoenice and Palestine,
as far even as Egypt, attacking those who inhabited the left side of the Nile as one sails
upstream, (in) the region called Arabia.
57(2) It was not possible for the war to be
considered easy because it was undertaken by a woman, and they say that this fight
was (so) tough and hard to win for the Romans that the commander of the soldiers in
Phoenice and Palestine summoned the general (in charge) of all the cavalry and
infantry forces of the East. He laughed at the appeal and excluded from battle the man
who had summoned him.
58(3) But when he had drawn up his forces in battle order to
face Mavia, who was leading her forces against (him), he was routed, and with diffi-
culty rescued by the commander of the Palestinian and Phoenician soldiers. For when
he saw him (the general) in danger, he reckoned it foolish to remain outside the fray in
accordance with his order. Running forward, he engaged the barbarians and gave him
(the general) the opportunity of a safer flight, (4) while he himself conducted a gradual
withdrawal, firing off missiles in his flight, and he beat off the advancing enemy with
arrows. Many of those living there still now commemorate these events, and among
the Saracens they are (recounted) in odes.
59(5) Because the war was becoming a
burden, it seemed necessary (to the Romans) to send an embassy to Mavia concerning
peace. ( … )
Sozomen (HE VI.38.14–16) goes on to recount the baptism of another Arab leader, Zocomus,
‘not long before the present reign’. He therefore, like Mavia, became a useful ally of the Romans.
Neither the date of his conversion nor the location of his tribe can be fixed with any certainty,
however. See Sartre 1982a: 144–6, Shahîd 1984: 188–9 on Zocomus, Isaac 1998a: 449.
For further important developments in the East (in Armenia) under Valens, see Chapter 2

The promotion of a frontier post
Resaina, a Roman veteran colony set up probably under Septimius Severus, and since the peace
of Jovian not far from the frontier, was evidently upgraded by Theodosius I early in his reign.
Theodosius also constructed walls for smaller places in Syria, such as Gindarus, 40 km north-
east of Antioch, and Lytargon, also known as Litarba, nearly 50 km east of Antioch (Mal. 13.39
Chr. Ede.35:In the year 692 (380/1) Theodosius the Great built the city of Resaina
in Osrhoene.
Mal. 13.38 (267.83–6/345.21–346.2): Theodosius made the village formerly known
as Rophaeina (Resaina) into a city, which was renamed Theodosiopolis; from that
point the village (p.346) received the status of a city during the consulship of
Merobaudes and Saturninus (
AD 383) up to the present day. (tr. Jeffreys and Scott,
Roman–Persian negotiations (381–7)
By the start of 383, a marked improvement in relations between Rome and Persia had occurred,
no doubt in part occasioned by the change of rulers in both states. See Greatrex 2000a: 41–4. 62
Orosius, Hist.VII.34.8: In these days 63the Persians, having slain the persecutor Julian
and often defeated other emperors, and having just now put Valens to flight, spewed
forth with coarse insults their satisfaction at their latest victory. They sent envoys back
and forth to Constantinople to Theodosius, and asked for peace as suppliants; a treaty
was then made, which the whole East has enjoyed in great calm up to now.
Themistius, Or.16.212d–213a (304.4–10):
64If we have accomplished these things,
then we may do more; just as we defeated the Scythians (Goths) without blood or
tears, so we shall be reconciled with the Persians shortly, and thus we shall regain
Armenia; so too we shall recover as much of Mesopotamia as belonged to others
so we shall name many consuls for their good deeds and their good servic
es. 66
In 384 a Persian embassy arrived in Constantinople, its principal aim no doubt being to
announce the accession of Shapur III (383–388). But it is clear that it was just part of a contin -
uing process of negotiation, which was to result three years later in the partitioning of Armenia
(see Chapter 2). From a Sasanian seal gem discovered in Pakistan it is supposed that the Persian
ambassador of Shapur III was one Yazd¡n-Friy-Sh¡b¥hr. See Curiel and Gignoux 1975: 41–4,
Stock 1978: 171–6 with Greatrex 2000a: 42.
Marc. com. a.384.1: Persian ambassadors came to Constantinople requesting peace
from the Emperor Theodosius. 67(tr. Croke)
Pacatus, Pan. Lat. II (XII) 22.4–5: ( … ) Persia itself, formerly a rival of our state and
notorious for the deaths of many Roman leaders, seeks to excuse by obedience what -
ever harsh (actions) she has committed against our rulers. (5) At last that king (of
Persia), having previously disdained to admit that he is a man, now admits fear and

pays reverence to you in those temples in which he (himself) is revered, (and) then by
sending an embassy and by offering gems and silk, and in addition by providing tri-
umphal animals for your chariots. Although in name (he is) still a federate, he is now,
however, a tributary in his devotions.
Peace with Persia agreed (387)
Epitome de Caesaribus 48.5 (174.30–1):He made peace also with the Persians, who
had asked for it. 69
The Hunnic invasion of the eastern provinces (395)
In 395, and again in 397 or 398, substantial numbers of Huns from the steppes crossed the
Caucasus mountains and penetrated deep into the Roman east (and the western portions of the
Sasanian empire). The Hunnic invaders, although eventually defeated by Eutropius, were able
to wreak great havoc throughout the eastern provinces: the citizens there were unaccustomed to
such lightning strikes, while many of the troops usually stationed in the east had been with -
drawn by Theodosius to combat the western usurper Eugenius, who was defeated only in 394.
See Maenchen-Helfen 1973: 48–50, Thompson 1996: 31–2, Heather 1998: 501–2, Greatrex
and Greatrex 1999: 65–73. Their most devastating incursion took place in 395; another was
feared in 396, but failed to materialise. The impression made by the Huns is very clear from the
following extracts; even if some are clearly derivative of one another, it is worthy of note that
Joshua the Stylite, writing over one hundred years later, makes specific allusion to the episode.
Chr. 724, 136.20–137.9: And in this year, the accursed people of the Huns came into the
lands of the Romans and passed over Sophene, Armenia, Mesopotamia, Syria,
Cappadocia, as far as Galatia; they took a great number of captives and turned back so that
they might return to their (own) country. But they went down to the banks of the
Euphrates and Tigris, in the province of the Persians, and reached the Persians’ royal city;
they did no damage there, but laid waste many villages by the Euphrates and Tigris, and
killed (many) and took a great number of captives. But when the Huns heard that the
Persians were marching against them, they prepared to flee, and (the Persians) pursued
them and killed one of their detachments, and took back (p.137) from them all the spoils
they had seized; and they freed male captives from them 18,000 in number, and led them
to their cities Selok and Kaukaba, which are called Ardashir and Ctesiphon, where they
were for many years. The Persian king assigned rations for them – bread, wine, date wine
and oil. From these 18,000 there remained but a fe w(…)thefirst thousand; and the
Persians let them go so that they might return to their (own) land. (tr. M. Greatrex)
Chr. Ede. 40:In the month of July that year (395), the Huns crossed over into the ter -
ritory of the Romans. (tr. M. Greatrex)
Claudian, In Rufinum II.28–35:Others, brought by an unlooked-for pass across the
Caspian Gates and the Armenian snows, fall upon the wealth of the East.
72Now the
pastures of Cappadocia and Argaeus, father of swift horses, are in smoke; now the deep
Halys is reddened and the Cilician does not defend himself on his steep mountain. The

pleasant districts of Syria are laid waste and the enemy horse tramples the unwarlike
(banks of) the Orontes, (more) accustomed to dances and a people rejoicing in songs.
Claudian,In Eutropium I.245–51:Across the Phasis are led Cappadocian mothers,
and captive herds, led away from their ancestral quarters, drink the frosts of the
Caucasus, exchanging the pastures of Argaeus for Scythian forests. Beyond the
Cimmerian marshes, the Gates of the Tauri, the flower of Syrian (youth) is enslaved.
The terrible barbarians cannot cope with their spoils; sated with booty, they turn to
Cyrillonas, ed. Bickell, 586–7 (verses 243–84): Every day stirrings, every day news,
every hour torments, every (day) conflicts. Your will has drawn the East into captivity,
and the wasted cities are uninhabited. The West is being struck, and behold, its cities
hold peoples who do not know you. Dead are the merchants, the prayers have ceased,
the women are widowed, the offerings have stopped. The North is distressed and full
of wars. If you delay, my Lord, they will destroy me. If the Huns conquer me, my
Lord, why have we taken refuge with the martyrs? If their swords kill my sons, why did
we adopt your great cross? (p.587) If you will yield to them the cities, where will be the
glory of your holy church? Not a year has passed since they came to destroy me and
took my children prisoners, and lo, they are threatening to humiliate our land a
second time. (tr. Maenchen-Helfen, rev. M. Greatrex)
Euphemia, §4, 46.2–8 (Syriac), cf. 150 (Greek): In the year 707 (395/6) by the reck-
oning of the Greeks the Huns had come forth, and they took prisoners and laid waste
the land and came as far as Edessa. And Addai, the strat¢lat¢sat that time, did not give
permission to go out against them because of treason in their midst,
and for this cause the armies of the Romans came down and lived in Edessa for a time.
(tr. Burkitt, rev. M. Greatrex)
Jerome, Ep.60.16: The East seemed to be free from these evils and merely dismayed at
the reports; but lo, (I tell) you, in the year gone by,
74the wolves not of Arabia but of
the north, were unleashed from the furthest crags of the Caucasus, and overran so
many provinces in a short time. How many monasteries were captured, how many
rivers were stained with human blood! Antioch was besieged and other cities past
which flow the Halys, Cydnus, Orontes and Euphrates. Hordes of captives were led
off; Arabia, Phoenicia, Palestine and Egypt were seized by fear.
75( … )
Philostorgius, HEXI.8 (138.3–8): But the eastern Huns crossed the river Tanais
(Don), and pouring into the provinces of the East, broke through Greater Armenia
into the district called Melitene. From there they attacked Euphratesia and penetrated
as far as Coele Syria; and, overrunning Cilicia, they engaged in an incredible slaughter
of people. (tr. Walford, revised)
Ps.-Dion. I, 187.17–188.14: In this year (706 [394/5]), the Huns entered the terri -
tory of the Romans and subdued all the lands of Syria
76which are in the vicinity of the
Cahja mountain, 77that is Arzanene, and Martyropolis, and Amida and Anzitene and
Samosata; 78and after crossing the Euphrates, they cut the bridge behind them. And
on all sides the Roman forces gathered against them and destroyed them; not one of
them escaped. And at that time the people in the fortress of Ziatha were besieged.
When the terrible scourge of the Hunnic army reached the region of Amida, all the

people of the country fled and entered the fortresses (p.188) which are by the river
Tigris and by the Deba; 80and they are called the fortress of Ziatha the great and the
fortress of Ziatha the lesser, and the fortress of Eghil of Sennacherib the king of
Assyria. That great fortress of Ziatha was situated between the Tigris and the Deba.
The Deba flows past the wall from the west and the Tigris from the east, and they mix
together to the south of the wall. (The place) is very rugged and inaccessible because it
lies at a great height and has only one gate. But the Huns seized the gate of the wall and
also the aqueducts which go down to the Tigris and the Deba; they stood on these and
held them until the men who lay on the Cahja (mountain) perished; and at last those
who were left handed over the fort. But the Huns, who are men without pity, slaugh -
tered the whole populace with the edge of the sword and made the rest captive; and
they set fire to the whole fort and it was never again inhabited.
81(tr. M. Greatrex)
Socr. HEVI.1.5–7 (311.23–8): When therefore the Emperor Arcadius met the army
outside the gates (of Constantinople) as was customary, the soldiers at that moment
killed Rufinus, the emperor’s prefect. (6) For Rufinus was suspected of aspiring to the
sovereignty and had the reputation of having invited the Huns – the barbarian nation
– into the Roman empire. (7) For at that time they were overrunning Armenia and
certain parts of the East.
Soz. HEVIII.1.2 (347.10–13): Meanwhile the barbarian Huns were overrunning the
Armenias and certain parts of the eastern provinces. It was said that Rufinus, the pre-
fect of the East, secretly invited them to set the empire in confusion, since he was sus-
pected in any case of wishing to seize the sovereignty.
Josh. Styl. 9 (243.2–8): What made these words
82find credence was the devastation
and depopulation which the Huns wrought in the Greek territory in the year 707
(395/6), in the days of the emperors Honorius and Arcadius, the sons of Theodosius
the Great, when all Syria was delivered into their hands by the deception of the prefect
Rufinus and the negligence
83of the strat¢lat¢s Addai. (tr. Wright, rev. M. Greatrex)
The Roman counter-attack on the Huns
Despite the lack of evidence, it is clear that the eunuch Eutropius, for all the criticisms of
Claudian, managed to defeat some Huns in 397 or 398. See Cameron 1970: 125 (placing the
campaign in summer 398), Blockley 1992: 47 (preferring 397).
Claudian, In Eutropium–6:You (Eutropius) will not now terrify the Arme -
nians with javelins and bow, nor will you drive on your swift horse acro
ss the fields.
Claudian, In Eutropium I.252–4:He (Eutropius), however (for what will shame a
weak slave? Or what could blush in his red face?), returns as a victor
( … ).

Roman–Persian negotiations concerning the Caspian Gates from 363
This excerpt is of great importance to our understanding of Roman–Persian relations in the
fifth century. While some regard it as proof of the existence of a treaty between the two sides, by
which the Romans made annual payments to the Persians, others draw attention to how little it
actually tells us. Of the two positions, the minimalist is preferable. As Blockley concludes, ‘all
that the passages of Lydus allow us to conclude is that the issue of the defence of the Caucasus
might have emerged first during the third quarter of the fourth century; that it was discussed
inconclusively until the reign of Anastasius; and that there was no agreement.’ See Blockley
1985: 63–6, quotation from 66, Blockley 1992: 50–1, Luther 1997: 104–6.
Joh. Lyd. De Mag.III.52–53 (212.14–214.7): (…)Aslong as the Romans con-
trolled Artaxata and the regions even beyond, they were able, since they were on the
spot, to resist them (the barbarians who might come through the Caspian Gates). But
when they evacuated these and other regions under Jovian, the Persians were unable
to defend both their own and the former Roman territory, and unbearable turmoil
constantly gripped the Armenias subject to each state. Therefore, after the failure
under Julian, talks were held between Salutius, who was then prefect, and the Persian
grandees and later with Yazdgerd,
2in order that, sharing the cost, both states should
build a fortress on the aforementioned pass and bring help to the area in checking the
barbarians who were overrunning it.
3But since the Romans were embroiled in wars in
the west and north, the Persians, insofar as they were more exposed to the barbarians’
incursions, were compelled to build the fortress against them there, naming it
4in their own language and establishing a garrison there. And the enemy
was unable to effect an entrance. (53) For this reason the Persians attacked the
Romans and spread little by little over Syria and Cappadocia, alleging that they had
been wronged and had been deprived of the money for common (p.214) projects to
the amount of the Roman contribution.
5As a result, the first Sporacius was sent by
Theodosius I to negotiate with the Persians. He, through the power of his money and
his ability at speaking, almost persuaded the Persians to leave the Romans in peace and
be their friends, since the Romans were so generous towards them.
6And this affair

dragged on until the reign of our Emperor Anastasius, being talked over, decreed
about and, in short, having been prevaricated over. ( … ) (tr. Blockley, revised)
Wars in Armenia (363–87)
Once Jovian had agreed not to help the Armenian king Arsaces, Shapur took the opportunity to
try to seize control of Armenia. The precise terms of the agreement of 363 concerning Armenia
are unknown, and were clearly interpreted differently by Shapur and the Romans. The war
soon took its toll on the kingdom, and Arsaces surrendered to Shapur in 368/9 (Epic Histories,
IV.21–50). See Baynes 1955: 198–9.
Ammianus XXVII.12.1–3:The king of Persia, the aged Shapur, who from the very
commencement of his reign had been addicted to the love of plunder, after the death
of the emperor Julian, and the disgraceful treaty of peace subsequently made, for a
short time seemed with his people to be friendly to us; but presently he trampled
underfoot the agreement which had been made under Jovian, and poured a body of
troops into Armenia to annex that (country) to his own authority, as if the validity of
the agreements had expired. (2) At first by various types of tricks he harassed (this)
nation, rich in manpower, through slight expenditure, seeking to influence some of
the nobles and satraps, and making sudden inroads into (the districts belonging to)
others. (3) Then through careful enticements combined with perjury, he captured
their king Arsaces, having invited him to a banquet, and ordered him to be conducted
to a hidden back door. His eyes were put out, and he was loaded with silver chains,
which among them is looked upon as a solace under punishment for men of rank, tri-
fling though it be; then he removed him to a fortress called Agabana, where, after
being tortured, he perished by the executioner’s blade.
8(tr. Yonge, revised)
Shapur was active on a broad front, intervening also in the Transcaucasus, where he ousted the
Roman appointed king of Iberia, Sauromaces. Since the peace of 363 made no mention of Iberia,
Shapur’s actions were particularly provocative, as Ammianus makes clear. See Toumanoff 1963:
150 n.5, 460, Chrysos 1976: 45–8, Blockley 1984: 36–7, 48 n.61, Braund 1994: 260.
Ammianus XXVII.12.4: After this (the capture of Arsaces), in order that his
(Shapur’s) perfidy might leave nothing unpolluted, having expelled Sauromaces,
whom the authority of the Romans had made governor of Iberia, he conferred the
government of that district on a man of the name of Aspacures, giving him a diadem
to mark the insult offered to the decision of our emperors.
10(tr. Yonge, revised)
With Arsaces disposed of, it remained for Shapur to eliminate the rest of the royal family, who
had sought refuge in the fortress of Artogerassa. The Armenian tradition differs somewhat from
the Roman: there is no reference to Cylaces and Arrabannus, nor to an initial defeat of the Per -
sians. See Garsoïan 1989: 303–5.
Ammianus XXVII.12.5–9: Having accomplished these (feats) with malicious enthu -
siasm, he committed the charge of Armenia to a eunuch named Cylaces and to

Arrabannus, 12a couple of deserters whom he had received some time before (one of
them, it was said, had been a prefect of that nation, and the other commander-in-
chief); and he enjoined them to use every exertion to destroy Artogerassa, a town
(made) strong by (its) walls and (its) garrison, in which were the treasures and the wife
and son of Arsaces. (6) These generals commenced the siege as they were ordered. And
as it is a fortress placed on a very rugged mountain height, it was inaccessible at that
time, while the weather was cold and the ground covered with snow and frost; and so
Cylaces being a eunuch, and, as such, suited to feminine flatteries, took Arrabannus
with him and quickly approached the walls; having received a promise of safety, he
and his companion were admitted into the city as he had asked. He advised the
defenders and the queen in a threatening fashion to calm the passion of Shapur, the
most unmerciful of all men, by a speedy surrender. (7) And after many arguments had
been urged on both sides, the woman bewailing the cruel fortune of her husband, the
men who had been most active in wishing to compel her to treachery, pitying her dis -
tress, changed their views; and conceiving a hope of higher preferment, they in secret
conferences arranged that at an appointed hour of the night the gates should be sud -
denly thrown open, and a strong detachment should sally forth and fall upon the ram -
parts of the enemy’s camp, (surprising it) with sudden slaughter; they promised that
they would see to it that those undertaking these things would pass unnoticed. (8)
Having ratified these matters with an oath, they left the town, and led the besiegers to
acquiesce in inaction by representing that the besieged had required two days to delib-
erate on what course they ought to pursue. Then in the middle of the night, which was
spent in sleep all the sounder on account of the greater freedom from danger, the gates
of the city were thrown open, and a strong body of young men poured forth with great
speed, creeping on with noiseless steps and drawn swords, and suddenly, when they
entered the camp of the enemy, who feared nothing, they slew many men as they lay
(there) without meeting any resistance. (9) This sudden treachery and the unexpected
loss of the Persians gave grounds for a fierce quarrel between us and Shapur; and
another cause for his anger was added, in that the Emperor Valens received Pap, the
son of Arsaces, who at his mother’s instigation had left the fortress with a small escort,
and bade him stay at Neocaesarea, a most celebrated city of Pontus Polemoniacus,
where he was treated with great liberality and high respect. Cylaces and Arrabannus,
attracted by this humanity of Valens, sent envoys to him to ask for assistance, and to
request that Pap might be given to them for their king.
13(tr. Yonge, revised)
Epic Histories IV.55:Then Shapur king of Persia sent a certain two of his princes, one
of whom was called Zik and the other Kar¢n,
14with five million men against the
country of Armenia, to go, undermine and destroy the country of Armenia. And they
came and arrived at the land of Armenia. Then, when the queen of the country of
Armenia Paranjem, the wife of Arsaces (Arshak) king of Armenia, saw that the forces
of the Persian king had come and filled the countries of Armenia, she took with her
some eleven thousand select armed, warlike azats, and together with these she rushed
from the Persian forces and entered the fortress of Artagerk in the land of Arsharunik.
Then the entire Persian army came up, installed itself around the fortress, kept watch,
surrounded, and besieged it. Now those who were fortified inside trusted in the

strength of the place, while the others entrenched and settled themselves outside
around the valleys. And they were installed around the fortress for thirteen months
but were unable to take the fortress because the site was extremely strong. They
ravaged and devastated the whole land. They went out and plundered the whole land,
and from the surrounding districts and countries led captive men and animals back to
their fortified camp. They brought in supplies from elsewhere, consumed them, and
besieged and guarded the fortress. ( … ) (tr. Greenwood)
Shapur’s initial failure at Artogerassa and Roman support for Pap led to an escalation of hostili-
ties, with the Romans drawn into playing an increasingly active role. Initially, however, they
seem to have tried to keep their distance.
Ammianus XXVII.12.10–13: However, for the moment assistance was refused them
(Cylaces and Arrabannus); but Pap was conducted by the general Terentius back to
Armenia, where he was to rule (that) nation in the meantime without any of the insig -
nia (of royalty), which was a very wise regulation, in order that we might not be
accused of breaking the treaty of peace. (11) When this arrangement became known,
Shapur was enraged beyond all bounds, and collecting a vast army, entered Armenia
and ravaged it by open pillaging.
16Pap was terrified at his approach, as were also
Cylaces and Arrabannus, and, as they saw no reinforcements (coming), they fled into
the recesses of the lofty mountains which separate our frontiers from Lazica; there they
hid in the depths of the woods and among the defiles of the hills for five months, elud-
ing the various attempts of the king (to discover them). (12) And Shapur, when he saw
that his efforts were being spent in vain while the winter stars were shining, burnt all
the fruit trees, and all the fortified castles and camps of which he had become master
by force or treachery. He also surrounded Artogerassa with the whole weight of his
numerous forces, and, after battles of varying outcomes, when the defenders grew
slack, he set fire to it (as it lay) exposed; and he carried off from there the captured wife
of Arsaces and all his treasures.
17(13) For these reasons the comesArintheus was sent
into these districts with an army, to aid the Armenians in case the Persians should
attempt to harass them by (another) similar campaign.
18(tr. Yonge, revised)
Epic Histories V.1:Then after all this Mushe³, the son of Vasak, 19collected the whole
free contingent of the remaining men who had survived and travelled together with
them to the king of the Greeks. He presented the request of the country of Armenia
and all the instances of tribulation which had befallen them, and requested from the
emperor Pap, son of Arsaces, as king over the country of Armenia. The great king of
the Greeks made Pap the son of Arsaces king over the country of Armenia as he
(Mushe³) had sought from him, and the king of the Greeks gave them great support;
they sent to the land of Armenia together with king Pap the stratelatnamed T¢rent
and a certain comesAd¢with six million men.
20( … ) (tr. Greenwood)
The Roman intervention, even if not on the scale claimed by the Epic Histories, sufficed topersuade
Shapur to continue the struggle by diplomatic, rather than military, means. The events narrated
by Ammianus here fall probably in 370.

Ammianus XXVII.12.14–18:At the same time, Shapur, who was extraordinarily
cunning, being either humble or arrogant as best suited him, under pretence of an
intended alliance sent secret messengers to Pap to reproach him for being neglectful of
his own interests, since, under the guise of royal majesty, he was really the slave of
Cylaces and Arrabannus. These two Pap hastily killed, having lured them (to him)
with flattery; (and he then) sent the heads of the dead men to Shapur in proof of his
22(15) When this calamity became generally known, Armenia would have
been ruined without a blow, if the Persians had not been so alarmed at the approach of
Arintheus that they forbore from invading it again. They contented themselves just
with sending ambassadors to the emperor, demanding of him not to defend this same
nation, in accordance with the agreement made between them and Jovian. (16) Their
ambassadors were rejected, and Sauromaces, who, as we have said before, had been
expelled from the kingdom of Iberia, was sent back with twelve legions under the
command of Terentius; and when he was already close to the river Cyrus,
entreated him that they might both reign as partners, being cousins, alleging that he
could not withdraw nor cross over to the side of the Romans, because his son Ultra
was still in the hands of the Persians as a hostage. (17) Once the emperor learnt this, in
order to put an end to the difficulties arising out of this affair too through sensible
planning, he assented to the division of Iberia, so that the Cyrus should divide it in the
middle; Sauromaces was to retain the portion next to the Armenians and Lazi, and
Aspacures to have the districts which border on Albania and Persia.
24(18) Shapur,
indignant at this, exclaimed that he was unworthily treated, because we had assisted
Armenia contrary to the terms of the treaty, and (because) the embassy which he had
sent to procure redress had been forgotten, and because it had been decided that the
kingdom of Iberia be divided without his consent or knowledge; and so, as if the gates
of friendship had been shut, he sought assistance among the neighbouring nations,
and prepared his own army in order, with the return of fine weather, to overturn all
the arrangements which the Romans had made with a view to their own interests.
Yonge, revised)
In the following year, 371, Roman troops became directly involved in the war. See Gutmann
1991: 176 and n.81, Seager 1996: 280.
Ammianus XXIX.1.1–4: At the conclusion of the winter, Shapur, king of the Persian
peoples, being full of cruelty and arrogance from the confidence (engendered by) his
former battles, having completed his army to its full number and greatly strengthened
it, sent out a force of heavy cavalry, archers and mercenary troops to make an invasion
of our territories. (2) Against this force the comesTrajan and Vadomarius,
26the former
king of the Alamanni, advanced with a mighty army, having been placed under orders
from the emperor to see to it that they contain rather than harass the Persians. (3)
When they arrived at Vagabanta,
27a place well suited for the legions, they stood up to
a fierce and rapid charge made upon them by the squadrons of the enemy, and reluc -
tantly retreated with deliberate care so as not to be the first to slay any of the enemy
soldiers, and not to be looked upon as guilty of having broken the treaty. At last, under

the pressure of extreme necessity, they came to an engagement with the barbarians,
and after having slain a great number of them, emerged the victors. (4) During the ces-
sation of regular operations (which ensued) several light skirmishes were undertaken
by both sides, which ended with different results; and after a truce was agreed by
common consent, and the summer had been used up, the generals of the two sides
departed, although they were still at odds (with one another). The king of Parthia,
intending to pass the winter at Ctesiphon,
28returned to his own home, and the
Roman emperor went to Antioch; and while he tarried there, secure from foreign ene -
mies, he very nearly perished through domestic treachery, as the series of events
described (here) will show.
29(tr. Yonge, revised)
Shapur’s diplomacy continued nonetheless, and finally bore fruit. Pap had the patriarch Nerses
executed, and then made unrealistic demands of the Romans. They in turn brought about the
king’s downfall and death, probably in 375 (Epic Histories V.24,Vit. Ners. 12 [p.36], Mos.
Khoren. III.39). See Garsoïan 1967: 313 and n.56, Garsoïan 1989:
Ammianus XXX.1.1–5, 18–21: Amidst all these difficulties and disturbances which
the treachery of the dux(Marcellinus) had provoked by the treacherous murder of the
king of the Quadi, a terrible crime was committed in the East, where Pap, king of
Armenia, was also murdered by secret plots. The original cause of this matter, brought
about by wicked design, we have ascertained to be this. (2) Some clever men who
often fed on public misfortunes heaped a number of accusations against this prince for
acts which they imputed to him even though he had just become an adult, and mali-
ciously exaggerated them to Valens. Among these men was the duxTerentius, a man
who walked about dejectedly and always (appeared) somewhat sorrowful, and
throughout his life was an unwearied sower of discord. (3) He, having formed an asso-
ciation with a few people of Pap’s nation
31whom a consciousness of their own crimes
had filled with fear, was continually harping in his letters to the court on the deaths of
Cylaces and Arrabannus, adding also that this same young king was full of haughtiness
in all his conduct, and that he behaved with excessive cruelty to his subjects. (4)
Therefore Pap was invited to court as if he were about to take part in a treaty, which,
on account of existing circumstances, had to be made then; (he was treated) with the
ceremony befitting a king, and then was detained at Tarsus in Cilicia, with a show of
honour, without being able to procure permission to approach the emperor’s camp or
to learn why his arrival had been so urgent, since on this point everyone preserved a
rigid silence. At last, however, by means of private information, he learnt that
Terentius was endeavouring by letter to persuade the Roman sovereign to send at once
another king to Armenia, in case, out of hatred to Pap, and in the expectation that he
would return, his nation, which at present was friendly to us, should revolt to the
Persians, who were most eager to reduce them under their power by either violence,
fear, or flattery. (5) He (Pap), reflecting on this warning, foresaw that the threat of a
grievous death hung over him. Now aware of the plot, he could not perceive any
means of safety, save by a speedy departure, (and) by the advice of his most trusty
friends he collected a body of 300 persons who had accompanied him from his own

country. With horses selected for special speed, acting as men are wont to do (when
faced by) great and uncertain fears, that is to say, with more boldness than prudence,
late one afternoon he started boldly forth at the head of his escort, moving in wedge
formation.(Pap successfully escapes from Roman territory.) (18) In this manner (through further
accusations against Pap) the baffling hatred which Valens had conceived against him
(Pap) was increased; and plan after plan was laid to take his life, either by force or by
stratagem; and orders to that effect were transmitted by secret letters to Trajan, who at
that time was in Armenia, in command of the forces (there). (19) Trajan accordingly
began to surround Pap with treacherous blandishments – at one time showing him
some letters of Valens, showing his calm state of mind
32– at another, partaking cheer -
fully of his entertainments. At last, when his deception was prepared, he invited him
with great respect to supper. Pap, fearing no hostility, came, and was seated in the
place of honour at the feast. (20) Exquisite delicacies were set before him, and the
splendid palace resounded with the music of string and wind instruments. Presently,
flushed with wine, the master of the feast himself left it for a moment, under pretence
of some natural want, and immediately a ferocious and violent barbarian of the troop
they call Scurrae was sent in, brandishing a drawn sword and glaring fiercely, to
murder the youth, against whose escape ample precautions had been taken. (21) As
soon as he saw him, the young king, who as it happened was leaning forwards beyond
(his) couch, jumped up and drew his dagger to defend his life by every means in his
power, but was stabbed in the breast, and fell like a disfigured victim (at the altar),
being shamefully cut to pieces with repeated blows. ( … ) (tr. Yon
ge, revised)
Epic Histories V.32:And the king, Pap, changed his mind and turned his heart away
from the king of the Greeks, and he wished to unite in love and make an agreement
with the king of Persia. And so he began to make an alliance with the king of Persia
and to send him envoys for the sake of an agreement. He also sent envoys to the king
of the Greeks (to say): ‘Caesarea, together with ten cities, belongs to me; now give
them back. My ancestors also built Urha (Edessa). Now, if you do not wish to cause
conflict, give it back; otherwise we will fight a great war.’ But Mushe ³and all the
Armenian princes urgently proposed to the king that he should not break the cove -
nant with the kingdom of the Greeks. He, however, did not listen to them and openly
manifested the hostility which he had towards the king of the Greeks.(…)(tr.
In the wake of Pap’s assassination, Valens became more willing to negotiate with Shapur, and a
sequence of embassies passed back and forth between the two rulers. We follow the chronology
of Blockley 1987: 227 n.22.
Ammianus XXX.2.1–5: These are the events (which took place) in Armenia that are
worthy of note (i.e. the assassination of Pap); but Shapur, after the previous defeat
which his troops had experienced, and having heard of the death of Pap, whom he had
been earnestly labouring to win to his own alliance, was struck by great grief; and, as
the activity of our army increased his apprehension, he paved the way for still greater

(disasters) for himself. (2) He therefore sent Arraces as his ambassador to the emperor,
to advise him that Armenia, as a perpetual cause of trouble, should be utterly
34or, if that plan was unsatisfactory, asking that an end might be put to the
division of Iberia, that the garrison of the Roman sector might be withdrawn, and that
Aspacures, whom he himself had made the sovereign of that nation, might be permit -
ted to reign alone.
35(3) To this proposal Valens replied with the view that he could not
modify the resolutions which had been agreed to by common consent, but rather that
he would maintain them by zealous care.
36Towards the end of winter, a letter was
received from the king in reply to this noble determination, (in which) he offered vain
and presumptuous excuses. For in it Shapur affirmed that it was impossible for the
seeds of discord to be radically extirpated unless those who had been witnesses of the
peace which had been made with Julian were to become involved, some of whom he
had heard were already dead.
37(4) Afterwards the emperor gave greater attention to
the matter, being now in a position to select his plans rather than respond, 38(and)
thinking that it would be beneficial to the state, ordered Victor, the magister equitum,
and Urbicius, the duxof Mesopotamia, to march with all speed to the Persians, bear -
ing a complete and plain answer (to the proposals of Shapur): namely that he, who
boasted of being a just man, and one contented with his own, was acting wickedly in
coveting Armenia, after a promise had been made to its inhabitants that
they should
be allowed to live according to their own laws.
39And unless the garrisons of soldiers
who had been assigned to Sauromaces as auxiliaries returned without hindrance at the
beginning of the ensuing year, as had been agreed, Shapur would (have to) perform
unwillingly the things which he had failed to do of his own accord. (5) And this
embassy would in all respects have been a just and honourable one, if the ambassadors
had not, contrary to their instructions, accepted some small districts in this same
Armenia, which were offered to them.
40After the ambassadors returned, Surena
(second in power after the king) arrived, offering the same districts to the emperor
which our ambassadors had rashly taken.
41(6) He was received with liberality and
magnificence, but dismissed without obtaining what he requested. And then great
preparations were made for war, in order that, as soon as the severity of the winter was
over, the emperor might invade Persia with three armies; and with this object he
began with all speed to bargain for the services of some Scythian mercenaries.
Then Shapur, having not obtained what his vain hopes had led him to reckon on, and
being exasperated more than usual because he had learnt that our emperor was prepar -
ing an expedition, stifled his wrath, and gave Surena a commission to endeavour to
recover by force of arms, if anyone should resist him, the territories which the comes
Victor and Urbicius had accepted, and to do the most harm possible to those soldiers
who had been detailed to aid Sauromaces. (8) His orders were at once carried out. Nor
was it possible to prevent or resist their execution, because a new cause of alarm sud -
denly came upon the Roman state; as the entire nation of the Goths was boldly burst -
ing into Thrace.
43( … ) (tr. Yonge, revised).
Before Valens’ attention was forced back to the Balkans, he had appointed a new king for

Armenia, Varazdat. The king’s chief general, Mushe³, was keen to cooperate with his allies in
the defence of the kingdom in the coming war. 44
Epic HistoriesV.34:He (Mushe³) also consulted with the Greek officers and through
them with the emperor, (asserting) that they should build one city in every district of
the land of Armenia: firstly (they should construct) cities, secondly they should estab -
lish strong fortifications and garrisons throughout the whole land of Armenia as far as
the border of the land at Ganzak, which was on the Persian side and marked the fron -
tier of Armenia. (They should also) enlist all the Armenian azatsby means of imperial
wages, and likewise the forces of the land of Armenia, so that every precaution should
thus be taken against their enemies, the Persian forces. And the king of the Greeks
gladly agreed to do this, so that the country might thus become strengthened in every
way and unshakeable (in its loyalty) on his side, and that the king of Persia might not
be able to usurp the country of Armenia.
45(tr. Greenwood)
However, Varazdat subsequently executed Mushe³, and was ousted from his throne by Manuel
Mamikonean probably in 377; he therefore took refuge with the Romans. Manuel then turned
to Shapur for help (Epic Histories V.37). See Gutmann 1991: 190, Redgate 1998: 137.
Epic HistoriesV.38:Then after this, the queen of Armenia, Zarmandukht, and the
sparapet Manuel sent Garjoyl the malkhazwith many of the Armenian nakharars
together with official letters, gifts, and presents to the king of Persia (to say) that they
would assist him, serve him, submit to him faithfully, and give this country of Arme-
nia to him. Garjoyl set out with his companions, arrived at the court of the Persian
king, presented to him the official letters from the queen and the sparapetof Armenia,
and transmitted to him the message of submission. When the king of Persia saw them,
he received them with great joy, honoured them with great esteem, and lavished gifts
upon Garjoyl. (tr. Greenwood)
The Persian king therefore despatched Surena with a force of 10,000 men to Armenia; he also
sent crowns for Arsaces and Valarshak, the two sons of Zarmandukht and Pap. But a revolt soon
broke out, led by Meruzhan. Manuel also turned against the Persians, and the Armenians suc -
ceeded in asserting their independence for a period of seven years.
46Valens, before he left for
Thrace early in 378, sought to come to an agreement with Shapur over Armenia; the outcome
of his embassy, undertaken by Victor, is unknown.
47Before his death, Manuel crowned Arsaces
IV, son of Pap and Zarmandukht, as king, but he was unable to command the loyalty of all his
countrymen. When Shapur’s successor to the throne in 379, Ardashir II, nominated an alterna -
tive king for Armenia, the way lay open to the partitioning of the kingd
The partition of Armenia (387)
General consensus now favours the date of 387 for the division of Armenia between the two
great powers, but some continue to prefer 384. Direct contemporary references to the partition
are non-existent, and so it is necessary to rely on later Armenian sources, supplemented by a dif -
ferent tradition preserved in the Narratio de rebus Armeniaeand Procopius. See Blockley 1987:

230–4, viewing the partition as a loose arrangement (rather than a rigid division), Gutmann
1991: 229–32, Blockley 1992: 42–4, Garsoïan 1999: 45–6 and G
reatrex 2000a. 49
Epic HistoriesVI.1:Then, after the death of the commander of Armenia Manuel,
nothing could sustain the kingship of Arsaces over the country. But many of the
Armenian nakhararsseparated (from him) and went to the king of Persia; they deliv -
ered to him the country of Armenia, and requested from him an Arshakuni (Arsacid)
king. And he agreed with great joy on his part to give (them) by his order (a king) from
the same Armenian royal house of the Arsacids, and to appropriate for himself
through him the country of Armenia. He then found a youth from the same house
named Khosrov, placed the crown on his head, gave him his own sister Zruandukht as
a wife, and placed under him all the forces in his power. He also gave his deputy Zik as
a tutor to King Khosrov. And they went and arrived in the country of Armenia. And
when King Arsaces saw them, he withdrew, abandoned the place, and went to the dis -
tricts of the Greeks. And so (the king of the Greeks) was assisting Arsaces king of
Armenia and the king of Persia, Khosrov. Then the forces of the king of the Greeks came in assistance; King Arsaces (was)
around the district of Eke³ eats‘ (Acilisene), and the Persian army and King Khosrov
were in the district of Ayrarat. Then envoys and messengers of the two kings – of the
Greeks and of the Persians – went back and forth from one to the other. Consequently
the king of the Greeks and the king of the Persians decided upon a mutual alliance with
one another, and they reckoned that it would be good to divide the country of Armenia
into two between themselves. ‘For,’ they said, ‘this powerful and wealthy kingdom is
situated between us. It will be good now that we shall be able to destroy and ruin this
kingdom. First (we shall) divide it in two through these two Arsacid kings whom we
have installed; then we shall attempt to eat away at them and impoverish them (and)
drive them into submission so that they shall not be able to raise their head between us.’ They confirmed this plan and divided the country in two. The portion on the Per-
sian side belonged to King Khosrov, and the portion on the Greek side belonged to
King Arsaces. But many districts were eaten away and cut off from them here and
there, and only a small part from both countries was left to the two kings.
50But never -
theless, the two Armenian kings, Arsaces and Khosrov, (divided) between them the
core districts of the Armenian kingdom
51which remained to them on both sides; and
the two Arsacid kings, having fixed boundaries between the two portions, established
peace. And so the land of Armenia (was) in two parts (and) obeyed two kings, each
part (being subject) to its own king. However, the portion of Khosrov was larger than
that of Arsaces. And many districts were cut off from both. And the kingdom of
Armenia was diminished, divided, and scattered.
52And it declined from its greatness
at that time and thereafter. (tr. Greenwood)
Narratio de rebus Armeniae, 10–12: (10) And Arsaces the lord of Great Armenia was
subject to him (Theodosius II). (11) But Khosrov the king of Armenia, who ruled in
the land of the Armenians, was under the Persian king Shapur. (12) After four years
Shapur expelled him from his kingship.

Proc.Aed.III.1.8–15 offers a rather garbled account of events, based on a ‘History of the Arme
nians’ (1.6). The two kings between whom Armenia is divided are named Arsaces (in the west)
and Tigranes (in the east); as in the Narratio, the division is said to have taken place during the
reign of Theodosius II, and Procopius reports that the western portion of Armenia was only one
fifth of the size of the eastern.
The partition appears to have been stable, and over the years the boundary line became more
defined. See Blockley 1992: 44–5. At some point after the split, the Romans installed a comesto
succeed Arsaces, but considerable autonomy was left to the Armenians, both in Armenia Inte -
rior and in the Armenian satrapies (under five satraps) until the sixth century.
55The Persians did
not permanently remove the Armenian royal family from the throne until 428.56It is likely that
Iberia too was partitioned, as it had been earlier, with the city of Tukharisi and Cholarzene in
the west remaining under Roman control, and the rest of the kingdom under the Persian-nomi -
nated king. See Chrysos 1976: 48, following Toumanoff 1963: 461, Zuckerman 1991: 536–8
(noting contradictions in the sources), Braund 1994: 261, Redgate 1998: 137–8.
On the Hunnic invasion of 395, which affected not only Armenia and Cappadocia, but also
Syria and Mesopotamia, see Chapter 1.

Roman anxieties over a new Persian king (399)
Yazdgerd I turned out to be one of the most amicable Sasanian rulers with whom the Romans
had to contend. But at the opening of his reign in 399, there were fears that he might prove to
be another Shapur II.
Claudian, In Eutropium II.474–84:Among these rumblings another more sinister
messenger flies in. (It announces that) Babylon, armed again and under a new king, is
threatening; that the (hitherto) quiet Parthians, detesting slothful inactivity, are now
seeking an end to the Roman peace. Rare among the Medes is the murder of kings, for
a single retribution remains for the whole nation; and yet they calmly obey cruel mas-
ters. But what would the year of Eutropius not dare? It has overcome our faithful ally
2and moved the Persians to kill their king, bringing them to the point of break-
ing their agreement. And in order that no quarter may be free from death, it has driven
the torches of the Furies across the waters of the Tigris.
Marutha and the restoration of good relations between the two powers (399)
That relations between the two powers flourished during Yazdgerd’s reign was due to a great
extent to the efforts of Marutha, who later became bishop of Martyropolis. He first visited the
Persian court in 399, where he made a great impression on the Sasanian king. He was therefore
permitted to take with him back to Roman territory the relics of the martyrs who had suffered
during the persecutions of Shapur II’s reign. With the backing of the emperor, the relics were
placed in what was to become a new frontier city; the place was named ‘The City of the Mar -
tyrs’, i.e. Martyropolis. See Noret 1973: 86 (the Greek account), Marcus 1932: 63–4 (the ear -
lier Armenian version), Fiey 1976: 41–3 (a later Arabic account) with Tisserant 1928: 144, Key
Fowden 1999: 49–56. As is clear from the passage below, apparently going back to a contempo -
rary account, the good relations fostered by Marutha soon bore fruit. See Greatrex and Greatrex
1999: 67–8.
Chr. 724, 137.9–22: But when the Persian king Yazdgerd was reigning, he again sent
back 1330 of these captives to their (own) land; but around 800 captives stayed in

Persia, (while) all the others had died through a plague of dysentery 3on account of the
anxiety and distress which they had suffered from the cursed Huns. All these things
the captives told us. Christians too and ascetics have related (these things); and junior
clerics themselves have reported about the good deeds which the captives said were
performed for them and about their gratitude towards the good and clement king
Yazdgerd, a Christian and blessed man among kings. May his memory be blessed and
his last days nobler than his first; (for) throughout his days he did good things for the
needy and wretched. (tr. M. Greatrex)
Yazdgerd I’s guardianship of Theodosius II (c.402) 4
Such was the goodwill between the two courts at this point that the Emperor Arcadius was able
to ensure the succession of his infant son Theodosius (II) in 408 by appointing the Persian king
as some sort of guardian for him. Because war broke out once again twelve years later, contem -
porary historians preferred to omit the story; our first source for the episode is Procopius, and
later writers add a few details. To Agathias, writing in the 570s, the episode seemed almost
beyond belief, to such a point had Roman–Persian relations deteriorated. See Bardill and
Greatrex 1996: 171–80 for an analysis of all the sources concerned.
Agath. IV.26.6–7 (157.5–13): In my opinion anyone who admires this is judging its
good sense 5from later events, not from the first impulse of the plan. How could it be a
good thing to hand over one’s dearest possessions to a stranger, a barbarian, the ruler
of one’s bitterest enemy, one whose good faith and sense of justice were untried, and,
what is more, one who belonged to an alien and heathen faith? (7) And if we are to
grant that no harm was done to the child, but that Theodosius’ kingdom was most
carefully safeguarded by his guardian, even while he was still a babe at the breast, we
ought rather to praise Yazdgerd for his decency than Arcadius for the venture. ( … )
(tr. Cameron, revised)
Proc. WarsI.2.1–10 (7.17–9.7): When the Roman Emperor Arcadius was at the
point of death in Byzantium,
6having a male child, Theodosius, who was still
unweaned, he felt grave fears both for him and for the government, not knowing how
he should provide wisely for both. (2) For the consideration occurred to him that, if he
provided a partner in government for Theodosius, he would in fact be destroying his
own son by bringing forward against him a foe clothed in the regal power; (3) while if
he set him alone over the empire, many would try to mount the throne, taking advan -
tage, as they might be expected to do, of the isolation of the child. (p.8) These men
would rise against the government, and, after destroying Theodosius, would make
themselves tyrants without difficulty since the boy had no relative in Byzantium to be
his guardian. (4) For Arcadius had no hope that the boy’s uncle, Honorius, would suc -
cour him, since the situation in Italy was already troublesome.
7(5) And the attitude of
the Medes disturbed him no less, and he feared lest these barbarians should trample
down the youthful emperor and do the Romans intolerable harm. (6) When Arcadius
was confronted with this difficult situation, though he had not shown himself saga -
cious in other matters, he devised a plan which managed without trouble to preserve

both his child and his throne, either as a result of conversation with certain of the
learned men, such as are usually found in numbers with a sovereign, or from some
divine inspiration which came to him. (7) For in drawing up the writings of his will,
he proclaimed his son as successor to his sovereignty but designated as guardian over
him Yazdgerd, the Persian king, enjoining upon him earnestly in his will to preserve
the empire for Theodosius by all his power and foresight. (8) So Arcadius died, having
thus arranged his private affairs as well as those of the empire. But Yazdgerd, the Per-
sian king, when he saw this writing which was delivered to him, being even before a
sovereign whose nobility of character had won for him the greatest renown, did then
display a virtue at once amazing and remarkable. (9) For, loyally observing the behests
of Arcadius, (p.9) he adopted and continued a policy of profound peace with the
Romans the whole time, and thus preserved the empire for Theodosius. (10) Indeed,
he straightaway despatched a letter to the Roman Senate, not declining the office of
guardian of the Emperor Theodosius, and threatening war against any who should
attempt to enter into a conspiracy against him. (tr. Dewing, revised)
Cedrenus I.586.3–7: And Arcadius died in his thirty-first year, having reigned for 26
years. And he ordained that the Persian king Yazdgerd look after his son Theodosius
and accept an embassy (bearing) 1000 lbs of gold.
Theoph. A.M. 5900 (80.8–24): Arcadius, perceiving that his son, the young
Theodosius, was still very small and unprotected and fearing that someone would plot
against him, proclaimed him emperor and by his will appointed the Persian king
Yazdgerd his guardian. Yazdgerd, the Persian king, after accepting Arcadius’ will,
behaved with ungrudging peacefulness towards the Romans and preserved the empire
for Theodosius. After despatching Antiochus,
9a most remarkable and highly
educated adviser and instructor, he wrote to the Roman senate as follows: ‘Since
Arcadius has died and has appointed me as his child’s guardian,
10I have sent a man
who will take my place. Therefore let no one attempt a plot against the child so that I
need not renew an implacable war against the Romans.’ After Antiochus had come, he
stayed with the emperor. Theodosius was educated wisely in Christian matters by his
uncle Honorius and his sister Pulcheria. And there was peace between the Romans
and the Persians, especially since Antiochus wrote many things on behalf of the Chris -
tians; and thus Christianity was spread in Persia, with the bishop of Mesopotamia,
Marutha, acting as mediator. (tr. Mango and Scott, revised)
Soz. HEIX.4.1 (395.12–13): Then at any rate the Persians, having been stirred up for
battle, concluded a one-hundred-year truce with the Romans.
11( … )
Cooperation in regulating cross-border traffic (408/9)
The spirit of détente between the two sides is well illustrated by a law of 408/9, by which the
two sides sought to confine cross-border trading to designated cities. See Synelli 1986: 89–94,
Isaac 1992: 407, Lee 1993a: 62–4, Winter and Dignas 2001: 213–15.
C.J. IV.63.4: Merchants subject to our government, as well as those (subject) to the

king of the Persians, must not hold markets beyond the places agreed upon at the time
of the treaty concluded with the above mentioned nation, 13in order to prevent the
secrets of either kingdom from being disclosed (which is improper). (1) Therefore no
subject of our empire shall hereafter presume to travel for the purpose of buying or
selling merchandise beyond Nisibis, Callinicum and Artaxata, nor think that he can
exchange merchandise with a Persian anywhere beyond the above-mentioned cities.
Because each party in the contract is aware of this, if one makes a contract under such
circumstances, any merchandise which has been either sold or purchased beyond the
said cities shall be confiscated by our sacred treasury, and, in addition to this
merchandise, the price which was paid in cash or in kind shall be surrendered, and the
offender sentenced to the penalty of perpetual exile. (2) Nor shall a fine of thirty
pounds of gold be lacking for judges and their subordinates for every contract entered
into beyond the above-mentioned limits; for through their borders travelled the
Roman or Persian to forbidden places for the purpose of trade; (3) with the exception
of those envoys of the Persians who have brought merchandise to be exchanged, to
whom, for the sake of humanity and on account of their character as ambassadors, we
do not refuse the privilege of trading beyond the prescribed limits; unless, under the
pretext of belonging to an embassy, and having remained for a long time in some
province, they do not return to their own country; for, as they engage in trade, the
penalty of this law will not unreasonably be imposed upon them, as well as upon those
with whom they have contracted while they reside (in Roman territory). (tr. Scott,
In this context we may note the recent discovery of a hoard of jewellery, counterfeit Roman
solidi (of Arcadius) and silver drachms of Yazdgerd I, at Humayma in Jordan (south of Petra).
The excavators have tentatively dated the hoard to the first decade of the fifth century, and have
suggested that the Roman solidiwere forged in Persia; according to them, the hoard ‘has pro-
vided archaeological confirmation for the literary evidence of political, commercial and ecclesi-
astical contact between the Byzantine and Sasanian regimes in the early fifth century.’ See de
Bruijn and Dudley 1995: 683–97 (quotation from 697), Maeir 2000: 17
2, 180–2.
The organisation of the Persian Church (410)
In February 410 the Council of Seleucia–Ctesiphon took place, the first such assembly of the
Persian Church. Instrumental in the organisation of the council was Marutha, as is acknowl -
edged in the proceedings. Yazdgerd was kept informed of the discussions which took place, and
agreed to give his support to the head of the Persian Church, the catholicosIsaac; he also received
considerable powers of intervention in Church affairs. According to the records of the council,
Yazdgerd, upon being praised for his favourable attitude to Christianity, declared, ‘East and
West form a single power, under the dominion of my majesty’ – a statement capable of being
interpreted in several ways. See Labourt 1904: 92–9, Brock 1985: 126, Sako 1986: 67–70,
Synelli 1986: 50–1, Blockley 1992: 54.

A raid on the eastern provinces (411)
Jerome,ep.126 describes ‘a sudden inroad of the barbarians’ which overran Egypt, Palestine,
Phoenicia and Syria. Its origin is unclear, and it is possible that the raiders came from Africa rather
than the desert to the East. See Graf 1989: 349–50, Shahîd 1989: 22–5, Isaac 1998a: 450.
The role of Marutha in fostering good relations (c.412)
Nonetheless relations between the two powers remained friendly for most of the second decade
of the fifth century. See Blockley 1992: 54–5, Schrier 1992: 77, van
Rompay 1995: 363–4.
Socr. HEVII.8.1–3, 18–20 (353.9–15, 354.20–7): About this time, 16it happened that
Christianity was disseminated in Persia for the following reason. (2) Frequent embassies
were constantly taking place between the Romans and the Persians: there were various
reasons why they were continuously sending embassies to one another. (3) Necessity led
to Marutha, the bishop of Mesopotamia, of whom we have made mention a little earlier
(VI.15), being sent at this moment by the Roman emperor to the king of the Persians.
( Various miracles are performed by Marutha, which impress the Persian king Yazdgerd I .)
(18) He (Yazdgerd) loved the Romans and welcomed their friendship towards them
(the Persians). He nearly became a Christian himself after Marutha, in conjunction with
Yabalaha, the bishop of Persia, had shown him another demonstration ;
(19) for both men, by taking to fasting and prayers, drove out a demon which was trou-
bling the king’s son. (20) Yazdgerd died before he completely embraced Christianity
and the kingdom fell to his son Bahram. Under him the treaty between the Romans and
Persians was broken, as we shall tell a little later (VII.18, translated below).
The situation on the frontier in the 410s
Even the frontier region in Syria enjoyed peace at this time. The ruler of the Persian-allied
Lakhmid tribe, Nu‘man I, was on friendly terms with the duxof Phoenice Libanensis,
Antiochus; according to the Lifeof Symeon the Stylite (the Elder), Nu‘man later claimed that
he would have become a Christian if he had not been under the control of the Persian king. See
Shahîd 1989: 161–4, dating the passage below to between 410 and 42
V. Sym. Styl. ch.101, 596–7: For Antiochus, the son of Sabinus, came to him, when
he was made duxat Damascus, and addressed the holy man in front of all (p.597),
(declaring) ‘Na‘man (Nu‘man) came up into the desert near Damascus and held a
banquet and invited me’. For at this time there was still no enmity between him and
the Romans.
19(tr. M. Greatrex)
Further evidence on life in this same frontier region comes from the Greek Life of Alexander
Akoimetos. Alexander received his education in Constantinople, but upon deciding to dedicate his
life to God moved to the East. After spending some time in Osrhoene and Syria, he assembled
some disciples and headed south from Osrhoene into what the Lifecalls ‘the Persian desert’

(Euphratesia) (V. Alex. Akoim.32 [682], cf. Gatier 1995: 451). 20Because he and his followers had
brought no supplies with them, they soon began to suffer; many began to complain, but Alex -
ander insisted that God would help them.
V. Alex. Akoim. 33 (683.9–13):And when they had gone a short distance, God, in
accordance with the word of the holy man, sent Roman tribunes and soldiers, carrying
good things to them from God. And they asked them to go to their castellato bless
them; for there are castellabetween the Romans and Persians to ward off the barbari -
ans, (situated) ten and twenty miles from one another.
Alexander then proceeded to journey along the limes, blessing the fortifications; mention is also
made of a raid by barbarians (no doubt Arabs) when a three-year drought ended and the land
was recovering. Clearly this region, although later dismissed by an envoy of Justinian as ‘alto -
gether unproductive and barren’ (Proc. WarsII.1.11), supported farmers, soldiers and nomads
in this period (V. Alex. Akoim. 33–4 [684–5] with Gatier 1995: 452–4).
Towards the end of Yazdgerd’s reign, relations were still sufficiently good for a Roman envoy
apparently to play a part in the internal politics of the Sasanian kingdom, if Tabari’s account can
be trusted. See Bardill and Greatrex 1996: 177 and n.28.
Tabari, I, 857–8/86 (Nöldeke 90–1): Then Bahram informed Mundhir that he was
going to return to to his father, (and) he set out to see the latter. But his father
Yazdgerd, because of his evil character, paid no attention to any of his children, and
merely took Bahram as one of his servants, so that Bahram suffered great hardship in
this. At that point an embassy came to Yazdgerd under a brother of the Roman
emperor (p.858), called Theodosius, seeking a peace agreement and a truce in fighting
for the emperor and the Romans. Hence Bahram asked Theodosius to speak with
Yazdgerd and to secure for Bahram (permission) to return to Mundhir. (tr. Bosworth)
The war of 421–2: background
In 417/18 the Persian catholicosYabalaha was sent to Constantinople ‘for the peace and reconcili -
ation of the two empires’. Evidently there had been some fraying in the relations between the two
powers, perhaps as a result of the success of the Christians in attracting converts among the
Persians. Nonetheless, in 419/20, the bishop of Amida, Acacius, was present at the second council
of the Persian Church, held at Veh-Ardashir. Even at this late stage there was no sign of imminent
confrontation between Rome and Persia, nor of the persecution of Christians in Persia. See
Labourt 1904: 100–2, Sako 1986: 71–7 (rightly arguing that Acacius did visit Persia in 420),
Schrier 1992: 77–8, van Rompay 1995: 363–4.
24But over the course of the year 420 the situation
deteriorated markedly. High-ranking Persians who had converted to Christianity refused to
apostasise and were martyred; a general persecution ensued.
25Yazdgerd died in autumn 420, and
was succeeded by Bahram V once his brothers, and rivals for the throne, had been eliminated. 26
C.J. VIII.10.10 (5 May 420): All persons who desire to do so shall be permitted to
surround their own lands, or premises established as belonging to them, with a wall, in

the provinces of Mesopotamia, Osrhoene, Euphratensis, 27Syria Secunda, Phoenice
Libanensis, Cilicia Secunda, both the provinces of Armenia,28both the provinces of
Cappadocia, Pontus Polemoniacus and Helenopontus, where this is most required,
and in other provinces.
29(tr. Scott, revised)
Marc. com. a.420.2–3: In the East the soldiers raised a tumult and killed their
general, named Maximinus. In Persia persecution raged violently against the Chris -
30(tr. Croke)
Augustine, De civ. deiXVIII.52 (652.61–4): What (of recent events) in Persia? Did
not persecution rage against the Christians to such an extent – if, however, it has even
now stopped – that some refugees fled from there even to Roman towns?
Ps.-Dion. I, 193.10–12: In the year 732 (420/1) there arose a severe, intense and piti -
less persecution against the Christians in the country of the Persians.
31(tr. M.
Cyr. Scyth. Vit. Euthym. 10 (18.15–19.9): Therefore this Terebon the Elder, the
grandfather of the Younger, while still very young and beardless, was afflicted with a
demon and the whole right side of him from head to foot was paralysed. His father,
called Aspebetus, spent much money, but it did not help. This Aspebetus, while a
pagan and a subject of the Persians, became an ally of the Romans in the following
manner. At the start of the persecution in Persia which then took place (around the
end of the reign of the Persian king Yazdgerd), the Magi wished to hunt down the
Christians; and they stationed the phylarchs of the Saracens under them at all points
on the roads, so that none of the Christians in Persia might escape to the Romans.
Aspebetus, then a phylarch, observing the harshness (p.19) and inhumanity of the
Magi in the city towards the Christians, felt compassion. He did not prevent some of
the Christians from fleeing, but rather the opposite; he cooperated (with them),
moved by sympathy, although he was an adherent of paganism, (having inherited
this) from his ancestors. When he was therefore accused before King Yazdgerd, he
took his half-paralysed son, I mean Terebon, and his whole family and wealth, and
sought refuge with the Romans. Anatolius, then magister militum per Orientem,
received them, made them allies of the Romans, and bestowed on Aspebetus the
phylarchate of the Saracens in Arabia allied to the Romans.
Quodvultdeus, Liber promissorum III.34.36 (179.29–37): Indeed in our time we
learnt of a persecution undertaken among the Persians while the devout and Christian
Emperor Arcadius was reigning.
33In order not to hand back the Armenians who were
fleeing to him, he undertook a war with the Persians. He gained a victory by means of
a sign before (the engagement) by which bronze crosses appeared on the clothes of the
soldiers as they went into battle.
34Hence the victor also ordered that a golden coin be
produced with the same sign of the cross, which continues in use throughout the
whole world, and in Asia especially.
Numismatic evidence concerning the war may also be found in Kent 1994: 256, nos.218–21,
solidi issued in Constantinople, 420–422, depicting Victory supporting a long jewelled cross, a
motif which remained popular for the rest of the century. See Holum 1977: 155, 163–7,
connecting the coins with the Persian war and the erection of a large jewelled cross at Golgotha

in 420/1; also Holum 1982: 108–9, Grierson and Mays 1992: 142, Kent 1994: 75, Key
Fowden 1999: 47.
An interesting insight into Persian fears of potential collaboration between Persian Christians and
the Romans comes from one of several martyr acts from this period. See Christensen 1944: 280–1
on Mihrshapur’s zeal in persecuting Christians. Peroz was martyred in September,
AG 733 (= 422).
Conf. Peroz(AMSIV.258–9): (Mihrshapur, the chief priest of the Zoroastrians of
Persia) got up in anger and went in to the king, and began his accusation (against
Peroz) by flattering the king. ‘From this moment on, my Lord,’ (he said), ‘all the
Christians have rebelled against you: they no longer do your will; they despise your
orders, they refuse to worship your gods. If the king would hear me, let him give
orders (p.259) that the Christians convert from their religion, for they hold the same
faith as the Romans, and they are in entire agreement together. Should a war interpose
between the two empires, these Christians will turn out as a thorn in your side in any
fighting, and through their playing false will bring down your power.’
(tr. S. Brock)
The course of the war of 421–2
The fullest account of the events of the war is provided by Socrates; scattered details are to be
found in other sources, but confusion soon arose in later historians between the conflict of
421–422 and that of 440. See Synelli 1986: 53–5, Croke 1984: 68–72, Blockley 1992: 56–7
and Greatrex 1993: 2.
Socr. HEVII.18 (363.2–365.24): When Yazdgerd, the king of the Persians, who in
no way persecuted the Christians there, 36died, his son, Bahram by name, received the
kingdom in turn. Persuaded by the Magi, he persecuted the Christians harshly,
bringing to bear on them various [Persian] punishments and tortures. (2) The Chris-
tians in Persia, oppressed by this coercion, therefore fled to the Romans, entreating
them not to allow them to be destroyed. (3) Atticus the bishop (of Constantinople)
gladly received the suppliants, and did his utmost to help them in whatever way was
possible; and he made the events known to the Emperor Theodosius. (4) It happened
that at this time the Romans had another cause of grievance towards the Persians,
since the Persians were unwilling to hand back the gold-diggers in their possession,
whom they had hired from among the Romans,
37and they were also seizing (the)
wares of (the) Roman merchants. (5) To this grievance therefore was added the flight
of the Christians there to the Romans. (6) For immediately the Persian (king) sent
ambassadors and demanded (back) the fugitives; but the Romans by no means
returned the refugees to them, not only because they wished to save them, but also on
account of their eagerness to do anything on behalf of Christianity. (7) For this reason
they chose to go to war with the Persians rather than allow Christians to perish. (8)
The treaty was therefore broken and a terrible war broke out, which I consider it not
inopportune to pass over briefly. (9) The Roman emperor acted first, despatching a
special army under the command of the general Ardaburius. He invaded Persia

through Armenia and laid waste one of the Persian districts called Azazene. 38(10)
Narses, a general of the Persian king, encountered (p.364) him with a Persian force; he
engaged (Ardaburius) in battle and, when beaten, retreated in flight. He realised that
it would be advantageous unexpectedly to invade Roman territory, which was
unguarded, through Mesopotamia, and thus to repel the Romans. (11) But Narses’
plan did not escape the Roman general. And so, having ravaged Azazene at great
speed, he himself also marched to Mesopotamia. (12) Consequently Narses, although
fielding a large force, was nonetheless not strong enough to invade Roman tory>. (13) When he reached Nisibis (which is a border city belonging to the Persians)
he suggested to Ardaburius that they should make war according to a treaty and fix a
place and day for an engagement. (14) To those who came he replied, ‘Report (this) to
Narses: the Romans shall make war not when you want, in their interests>’. (15) and forthwith, just as he was, the king of Persia made ready to go forth and engage
Ardaburius with a large force. But the emperor of the Romans>,
39learning that the
Persian (king) was making ready all his forces, simultaneously despatched reinforce -
ments, placing all his high hopes for the war in God. (16) And because the emperor
had faith , he found him to be beneficent, as became clear henceforth. (17)
While the people of Constantinople suspense and uncertainty as to the
fortunes of the war, angels of God appeared to some persons in Bithynia who were
travelling to Constantinople on their own business, and bade them announce (to the
people) to be of good cheer, to pray, and to have faith in God, since the Romans
would prove victorious; for they declared that they had been sent by God as arbitrators
of the war. (18) When this was heard, it not only encouraged the city, but also made
the soldiers more courageous. (19) When, as I have said, the war was transferred from
Armenia to Mesopotamia, the Romans laid siege to the Persians who were shut up the
city of Nisibis. (20) They constructed wooden towers and led them slowly up to the
walls by some device; and they slew many of those hastening to defend the walls. (21)
(p.365) Bahram, the Persian king, learning that his province of Azazene had been laid
waste and that those shut up in the city of Nisibis were under siege, made preparations
to come to an engagement in person with all his forces; (22) but, struck by the
strength of the Romans, he summoned the Saracens to his aid, whose leader
Alamundarus (Mundhir) was a noble and warlike man. He led an army of many thou -
40of Saracens, and bade the king of the Persians take heart. He announced to him
that the Romans would not hold out for long against him, and that they would hand
over Antioch in Syria (to him). (23) But the result did not fulfil his pronouncement,
for God inspired the Saracens with an irrational fear: thinking that the army of the
Romans was coming against them, they were thrown into confusion, and having
nowhere to flee, they threw themselves, armed, into the Euphrates. Around 100,000
men perished by drowning in the river. (24) This matter (turned out) thus. When the
Romans besieging Nisibis heard that the king of the Persians was leading a multitude
of elephants against them, they grew very afraid; and having burnt all their siege
engines, they withdrew to their own territories. (25) In case, events in turn>, I should seem to digress from the subject, I shall leave to one side the

battles which took place after this, and how another Roman general, Areobindus, slew
the noblest of the Persians in single combat; (nor shall I mention) how Ardaburius
ambushed and killed seven noble Persian generals or by what means another Roman
general, Vitianus, vanquished the remainder of the Saracens.
Mal. 14.23 (285.65–286.81/364.3–21):In that year Blasses, 42king of the Persians,
came, making war on the Romans. When the emperor of the Romans learnt of this, he
made the patrician Procopius magister militum per Orientem,
43and sent him with an
army to do battle. When he was about to engage in battle, the Persian king sent him a
message, ‘If your whole army has a man able to fight in single combat and to defeat a
Persian put forward by me, I shall immediately make a peace-treaty for fifty years and
provide the customary gifts’. When these terms had been agreed, the king of the Per -
sians chose a Persian named Ardazanes from the division known as the Immortals,
while the Romans selected a certain Goth, Areobindus, (who was) comes foederatorum.
The two came out on horseback fully armed. Areobindus also carried a lasso according
to Gothic custom. The Persian charged at him first with his lance, but Areobindus,
bending down to his right, lassoed him, brought him down off his horse and slew him.
(p.286) Thereupon the Persian king made a peace treaty. Areobindus returned to
Constantinople after his victory with the general Procopius and as a mark of gratitude
was appointed consul by the emperor.
44(tr. Jeffreys and Scott, revised)
Oracle of Baalbek, 112–14: And the Persians will rise up for a mighty war, and they
will be tripped up by the Romans and will propose peace for forty years.

Tabari, I, 868/103 (Nöldeke 108) provides a brief account of a war in which Mihr-Narse
(Narses), with 40,000 men, entered Constantinople and obliged the Romans to resume the
payment of tribute. He then returned home, ‘having accomplished everything that Bahram
wanted’. See Nöldeke 1879: 108 n.2, Synelli 1986: 55, Bosworth 199
9: 103 n.261.
Socr. HEVIII.20 (366.3–367.9): Let that much suffice on Palladius. 47The Roman
emperor in Constantinople, knowing clearly that the victory had been provided by
God, was so happy that, although his men were having such good fortune, he none -
theless longed for peace. (2) Accordingly he sent Helio,
48a man whom he treated with
great honour, bidding him to come to peace terms with the Persians. (3) When Helio
reached Mesopotamia, (he found that) there the Romans had made a trench for their
own protection; he sent Maximus, an eloquent man, who was the assessorof the gen -
eral Ardaburius, as an ambassador concerning peace. (4) This man, when he was in the
presence of the Persian king, declared that he had been sent concerning peace not by
the Roman emperor, but by his generals; for, he said, this war was not known to the
emperor, and were it known to him, would be considered of little consequence. (5)
When the Persian (king) readily chose to receive the embassy (for his army was
oppressed by hunger), there came to him those called by the Persians the Immortals
(this is a unit of ten thousand noblemen). They said that he should not accept peace
before they had made an attack on the Romans, who were [now] off-guard. (6) The
king was persuaded and kept the ambassador shut up under guard; and he sent off the
Immortals to lie in ambush for the Romans. When they had drawn near (to the

Romans), they divided themselves into two divisions, intending to surround some
section of the Roman army. (7) The Romans, catching sight of one of the detach-
ments of the Persians, prepared themselves for its onslaught; the other division was
not observed by them, however, for they attacked (so) suddenly. (8) Just as battle was
about to be joined, by the foresight of God a Roman army under the command of the
general Procopius emerged from behind a small hill. (9) Noticing that his fellow
countrymen were about to be in danger, he attacked the Persians in the rear and they
who had just previously surrounded the Romans were themselves surrounded. (10)
Once they had destroyed all these men in a short time, the Romans turned against
those attacking from their place of ambush; and in like fashion they slew every one of
them with their missiles. (11) Thus those known to the Persians as Immortals were all
shown to be mortals; , for Christ exacted justice (p.367) on
the Persians because they had killed many pious men, who were his servants. (12) The
king, apprised of the misfortune, feigned ignorance of what had occurred, and
received the embassy, saying to the ambassador, ‘I welcome peace, but am not yielding
to the Romans; rather, I am doing you a favour, because I have come to the conclusion
that you are the most intelligent of all the Romans.’ (13) Thus was settled the war
which had broken out on account of the Christians in Persia; in
the consulship of the two Augusti, the thirteenth of Honorius and the tenth of
Theodosius, in the fourth year of the 300th Olympiad; and the persecution of Chris-
tians in Persia also came to an end.
Thdrt. HEV.37.6–10 (340.22–341.21): ( … ) And in the previous war he (God)
made these same people (the Persians) look ridiculous when they were besieging the
city named after the emperor (Theodosiopolis).
50(7) (p.341) For when Gororanes
(Bahram) had been encircling the aforementioned city with all his forces for more
than thirty days, had brought many siege-engines
51to bear, had employed thousands
of devices, and had raised up lofty towers outside (the walls), the godly bishop alone
(whose name was Eunomius) stood firm and dissolved the strength of the devices
brought to bear.
52Our generals refused battle with the enemy and did not dare to bring
aid to the besieged, (but) this man opposed (the enemy) and preserved the city from
being sacked. (8) When one of the kings in the service of the barbarians ventured upon
his usual blasphemy, uttering (words like those) of Rabshakeh and Sennacherib,
madly threatening that he would set fire to the temple of God, that holy man did not
tolerate his raving; he gave orders that the stonethrower, which was named after the
apostle Thomas, be placed on the battlements and bade a huge stone be placed on it.
He (then) commanded (the men) to discharge the stone in the name of Him who had
been blasphemed. (9) It fell directly upon that impious king, landing right on his foul
mouth, destroying his face, smashing his entire head, and scattering his brains on the
ground. (10) Having witnessed this, (the king) gathered his army together and left the
city he had hoped to capture; he acknowledged his defeat through these events and, in
fear, made peace. Thus the great king of all looks after the most faithful emperor.(…)
Evagr. HEI.19 (28.8–16) = Eustathius frg.1 (FHG IV.138):He (Theodosius II) so
defeated the Persians, who had violated (the treaty) at the time when their king was
Yazdgerd the father
54of Bahram or, as Socrates thinks, when Bahram himself was

king, that he granted them peace after they had sent an embassy; (the peace) lasted
until the twelfth year of the reign of Anastasius. These things have been recounted by
others, but are most elegantly abridged by Eustathius of Epiphania the Syrian, who
also related the capture of Amida.
Chr. Pasch.579.19–20 (a.421): In the same year a victory over the Persians was
reported, in the month Gorpiaeus, on day 8 before the Ides of September, a Tuesday.
Theophanes, A.M. 5943 (104.1–4): In the bygone times, in which the Persian War
was undertaken, Marcian, being a common soldier, set out from Greece with his
detachment against the Persians; and when he arrived in Lycia he was struck with an
56(tr. Mango and Scott, revised)
The conclusion of the war of 421–2
Despite the triumphalist tone of the Roman sources, it is clear that the result of the war was a
stalemate. Not much is known of the peace terms agreed in 422. Procopius (below) notes one
clause in particular, while Malchus (also below) reports that each side undertook not to accept
defecting Arab allies from the other. Holum 1977: 170–1 reconstructs the conditions, although
it is unlikely that the Romans agreed to make any regular payments. See also Synelli 1986:
62–4, Letsios 1989: 527, Blockley 1992: 57–8, Rist 1996: 33–4, Isaac 1998a: 443, 451, Winter
and Dignas 2001: M19.
Marc. com. a.422.4: The Persians made peace with the Romans. (tr. Croke)
Malchus frg.1.4–7: The Persians and Romans (had) made a treaty after the greatest
war had broken out against them (the Persians) during the reign of Theodosius (II),
(according to which) neither side would accept the Saracen allies (of the other), if any
of them attempted to revolt. (tr. Blockley, revised)
Chr. Arb. 16 (67/91): In consequence of that (war) both parties agreed to give their
territories complete liberty in the matter of religion. (tr. Schrier 19
92: 83)
Proc. WarsI.2.11–15 (9.8–10.8): When Theodosius had grown to manhood and
was advanced in years, and Yazdgerd had been taken from the world by disease,
Vararanes (Bahram), the Persian king, invaded Roman territory with a mighty army;
however he did no damage, but returned to his home without accomplishing any -
thing. (And this came about) in this way. (12) The Emperor Theodosius happened to
have sent Anatolius, the magister militum per Orientem, as an ambassador to the Per -
sians on his own; as he approached the Persian army, he leapt down from his horse
alone and advanced on foot towards Bahram.
59(13) And when Bahram saw him, he
enquired from those who were near who this man could be who was coming forward.
And they replied that he was the general of the Romans. (14) Thereupon the king was
so dumbfounded by this excessive degree of respect that he himself wheeled his horse
about and (p.10) rode away, and the whole Persian host followed him. (15) When he
had reached his own territory, he received the envoy with great cordiality, and granted
the treaty of peace on the terms which Anatolius desired of him; one condition, how -
ever, (he added), that neither party should construct any new fortification in his own
territory in the neighbourhood of the boundary between the two (countries). When

this (treaty) had been executed, both (sides) conducted their affairs as they saw fit. 60
(tr. Dewing, revised)
The aftermath of the war of 421–2
The good cross-border relations established at the opening of the fifth century were not entirely
destroyed by the war – which had, after all, lasted for only two years. The actions of Acacius
described by Socrates were in effect a reciprocal gesture for the return of Roman prisoners cap-
tured from the Huns by Yazdgerd I (noted at the start of this chapter). But only four years later,
at its third council, the Persian Church severed its links with the west: it was no longer possible,
in the light of the recent hostilities, to rely on the intervention of western bishops to help resolve
the conflicts within the Persian Church. See Labourt 1904: 121–4, Bro
ck 1994: 74–5.
Socr. HEVII.21.1–6 (367.10–368.2): At this time a good deed made Acacius, the
bishop of Amida, more famous among all men. (2) For the Roman soldiers were by no
means willing to return to the Persian king the prisoners they had taken in their ravag -
ing of Azazene (Arzanene); the prisoners were perishing of hunger, being about 7000
in number, and this was a source of great grief to the Persian king. At this point
Acacius refused to allow these things to take place; (3) summoning the clerics under
his authority, he said to them ‘Men, our God needs neither dishes nor cups; for he
does not eat nor drink, because he is not in need of anything. Since therefore the
church here has acquired many gold and silver treasures through the goodwill of its
devotees, it is fitting that we should rescue the prisoners of the soldiers and feed them
by means of these.’ (4) Once he had gone through all these items and still others like
them, he melted down the treasures and paid out the ransom to the soldiers on the
prisoners’ behalf; he fed them, then gave them provisions and sent them off to their
king. (5) This deed of the remarkable Acacius astonished the Persian king all the more
because the Romans (thus gave the impression that they) were in the habit of being
victorious in both warfare and good works. (6) They say that the Persian king wished
that Acacius should come into his presence (p.368), in order that he might enjoy the
sight of the man; and, following an order from Theodosius, this took pla
Concern for the limitaneiof the East (438)
Relations between soldiers and civilians in the East had never been easy, but the peace which
prevailed for most of the fifth century did not, apparently, ease tensions. 62Civilians had tended
to get the better of soldiers in civil courts, it appears, and so the magister militum per Orientem
Anatolius persuaded Theodosius to issue a law by which the soldiers could only be tried before a
military judge.
Nov. Theod. IV.1:(Emperors Theodosius and Valentinian Augusti to Florentius, praeto -
rian prefect.) The efficiency of the limitaneidemands the support and assistance of
Our Clemency, since they are said to be afflicted by the complaints of certain persons
and by being produced before various judges, so that as between private life and

military science they are born to neither. 63Add to this the fact that they are compelled
to unlearn the use of arms for the observance of the forum of a civilian office, 64and
they are strangers in an alien life, inexperienced in litigation, ignorant of the court
actions which are instituted by shameless eloquence and popular doctrines. These
facts were brought to Our attention by the report of the Sublime Anatolius, magister
utriusque militiae per Orientem. Wherefore We do not allow any further that the loss
of so great an advantage shall be disregarded, since it is evident by the arrangement of
Our ancestors that whatever territory is included within the power of the Roman
name is defended from the incursions of the barbarians by the rampart of the fron -
65(1) For this reason We sanction by a law destined to live in all ages that
throughout all the region of the East from the farthest solitudes, no soldier of a dux,no
soldier of the limitanei, who with difficulty and hardship wards off the misery of
hunger by means of his meagre emoluments, shall be produced at all in Our most
sacred imperial court, either by the sacred imperial order of Our Divinity or by Our
divine imperial response which has been elicited by the prayers of a person who has
approached Us. In short, such soldiers shall not be produced in court by the mandates
of Our Eternity nor by the decision of any judge whatsoever. Furthermore, the regula -
tions of the Magnificent man (Anatolius) shall be unimpaired and shall be observed in
the future, since he established them by a careful investigation for the welfare of the
limitanei. All such soldiers shall now be returned to their own service who appear to
have been produced in court. (tr. Pharr, revised)
The war of 440
The war of 440 scarcely amounted to a war: a brief incursion by Yazdgerd was quickly deflected
by the handover of a sum of money to the Persian king. See Blockley 1992: 61 and Greatrex
1993: 2.
Marc. com. a.441.1: The Persians, Saracens, Tzanni, Isaurians and Huns left their
own territories and plundered the lands of the Romans. Against them were sent
Anatolius and Aspar, magistri militum, and they made peace with them for one year.
(tr. Croke, revised)
Thdrt. HEV.37.5–6 (340.13–21): Just such a thing (as the repulse of Rua by the
forces of nature in 435/40) happened in the Persian war. When they (the Persians)
learnt that the Romans were heavily occupied, they marched against the neighbouring
towns, violating the peace treaty. No one came to the aid of those under attack, for the
emperor, confident in the peace, had sent his generals and soldiers to other wars. By
hurling down a very fierce thunderstorm and a great hailstorm (God) hindered the
(Persian) advance and impeded the movement of their horses. (6) And in the course of
twenty days they were not able to traverse as many stades, and by then the (Roman)
generals had arrived and mustered the soldiers.
Mos. Khor. III.67 (347): He (Yazdgerd II), having forgotten the treaty, as soon as he

was reigning attacked the forces of the Greeks at Nisibis, commanding the army of
Azerbaijan to go into our country (Armenia). (tr. Greenwood)
E³ish¢7/61–2: And in his exceeding madness, like a ferocious wild animal, he
(Yazdgerd II) attacked the country of the Greeks; he struck as far as the city of
Nisibis and he ravaged through assault many districts of the Romans and he set on
fire all the churches. He collected plunder and captives and terrified all the forces of
the country.
68Then the blessed emperor Theodosius, because he was peace-loving in
Christ, did not wish to go out against him in battle, but sent him a man whose name
was Anatolius, who was his commander of the East, with many treasures. And the
Persians, those who had fled because of Christianity and were in the city of the
emperor, he seized and handed over to him.
69And whatever he (Yazdgerd) said at the
time he (Anatolius) fulfilled in accordance with his wishes and he prevented him from
much anger; and he returned from there to his own city Ctesiphon. (tr. Greenwood)
Isaac of Antioch, Hom.11.374–80: When the Persians plundered our frontiers, many
people from within the city of Nisibis joined (them). But after a little time the army
which had come to our border was lost, (together) with those who had joined it, so
that their records were destroyed.
70(tr. M. Greatrex)
Concern for the defence of the frontier (443)
The difficulties experienced by the eastern empire in warding off aggressors on several fronts in
the early 440s resulted in a substantial reform of the way frontier defences were administered.
From 443 the magister officiorum had the responsibility of ensuring that the limitaneiwere able
to protect the frontiers, and an annual report was required.
71As is clear from the following
extract, particular attention was paid to distribution of the annona(rations) to the Arabs, from
which, it appears, unwarranted deductions had been made. Clearly their contribution to the
defence of the frontier was judged sufficiently important for them to be specifically mentioned.
See Shahîd 1989: 49–50, Isaac 1998a: 446.
Nov. Theod. XXIV.2 (September 443): To these (duces), together with the
commandant (princeps) and the provosts of the camps (praepositi castrorum), as a
recompense for their labours, we assign a twelfth part of the annonaof the
limitanei only, to be distributed among them, of course, by the decision of the
73But from the annonaof the Saracen foederatiand of other tribes, we grant
that such ducesshall have absolutely no right to abstract and appropriate anything.
( … ) (tr. Pharr, revised)
The threatening situation in the East (447/8)
Despite the Roman concerns here reported by Priscus, no Persian attack materialised. The
Roman government might also have taken heart from some of the acclamations uttered at
Edessa in this very period, expressing not only support for Roman rule but also hostility to the
Persians (and Nestorians). See Frend 1972: 67 (citing Flemming 1917:
21, 27).

Priscus frg.10.9–15 (cf. Jord.Rom.333): They (the Romans) heeded his (Attila’s)
every bidding and considered it the command of their master, whatever order he
might issue. They were not only wary of starting a war with Attila, but were afraid also
of the Parthians who were preparing (for hostilities), the Vandals who were creating
havoc on the sea, the Isaurians whose banditry was reviving, the Saracens who were
overrunning the eastern parts of their dominions, and the Ethiopian tribes who were
in the process of uniting.
74(tr. Blockley, revised)
Peaceful state of the eastern provinces (450s)
Priscus frg.[19] (Suda A 3803): Ardaburius, the son of Aspar, a man of noble spirit
who stoutly beat off the barbarians who frequently overran Thrace. As a reward for his
prowess, the Emperor Marcian made him magister militum per Orientem. Since he
received this office in time of peace, the general turned to self-indulgence and effemi -
nate leisure. He amused himself with mimes and conjurors and stage spectacles, and,
spending his days in such shameful pursuits, he took no thought at all for things that
would bring him glory. Marcian, having proved himself a good emperor, quickly
passed away, and Aspar on his own initiative made Leo his successor.
75(tr. Blockley,
Roman negotiations with the Arabs (c.452)
The raid of 447/8 was not an isolated incident. The Romans therefore sought to come to terms
with the restive Arab tribe or tribes attacking the provinces. See Shahîd 1989: 55–6, dating this
passage to 453 and suggesting that the Arabs in question were probably Ghassanids or Kindites
rather than Lakhmids.
Priscus frg.26: Ardaburius, the son of Aspar, was fighting the Saracens around
Damascus. When the general Maximinus and the historian Priscus arrived there, they
found him in peace negotiations with the envoys of the Saracens. (tr. Blockley,
Allegations of treachery on the frontier, c.466
Nothing more is known of the allegations reported here; whether or not they were substanti -
ated, they at any rate indicate Roman suspicions of Persian intentions. See PLREII.136 and
Lane Fox 1997: 190 on the date. The Nestorian leader Narsai at Edessa in the late 450s had also
excited suspicions, prompting him to move to Persia (Barhadbeshabba HE31,PO 9.600).
Vit. Dan. Styl. 55 (53.22–54.1, 54.15–19): At around that time (c.466) someone
(called) Zeno, an Isaurian by race, came to the emperor. He brought with him letters
written by the then magister militum per Orientem Ardaburius, who was encouraging
the Persians to (undertake) an attack the Roman state and agreeing to join in assisting
(p.54) them. (The matter is then discussed by Leo and his advisers.) Having heard these

things, 77the emperor replaced Ardaburius, stripping him of all military office, and
ordered him to make for Byzantium as quickly as possible. In his place he appointed
Jordanes and sent him off; he also appointed Zeno comes domesticorum.
The strengthening of a frontier city (465/6)
Chr. Ede.70:In the year 777 (465/6) Leo built Callinicum in Osrhoene, which he
called Leontopolis after his name; and he also appointed a bishop in it. 78(tr. M.
Developments in southern Arabia (473/4)
Despite the terms of the agreement which had concluded the war of 421–422, towards the end
of his reign the emperor Leo accepted an alliance with an Arab, Amorcesus (Imru’ al-Qays) who
had defected from the Persians and had installed himself by force in Palaestina Tertia (in the
area of the Gulf of Aqaba). To the disapproval of the historian Malchus (frg.1), who alone
reports this development, the emperor even invited the ruler to Constantinople, appointed him
phylarch of the region and patrician, and bestowed many gifts and further territories on him.
Although this region was important economically because of the trade routes leading to the Red
Sea, it appears that Leo, in the wake of his disastrous expedition against the Vandals, felt he had
little choice but to accept the fait accompli with which he was presented. See Letsios 1989:
525–35, Z. Rubin 1989: 388–9, Shahîd 1989: 59–91, Greatrex 1998a: 28, 227 on these devel-
opments. In the late 490s Anastasius restored Roman control to this regi
Arab raids, c.474
Evagrius’ rather general notice on Arab raids (referred to as Scenitae, i.e. tent-dwellers) can be
supplemented by a homily of Isaac of Antioch, describing an attack on Beth Hur, in the vicinity
of Nisibis. See Klugkist 1987: 238–43, Shahîd 1989: 114–15, Gre
atrex 1998b: 287–91.
Evagrius HEIII.2 (100.6–9): In these ways therefore did Zeno change his way of life at
the opening (of his reign). His subjects both in the East and the West suffered badly, on
the one side (on account) of the barbarian Scenitae laying waste everything …
Isaac of Antioch, Hom.11.32–47:Why is our foolish community disquieted, so that
it complains against its chastisements? For it pleads with the judge, seeking vengeance
from him. Behold captivity and exile, along with the plundering of all our possessions.
The Arabs (‘Arbaye) harry the land in the portion which they took from it. The whole
world is in turmoil because the sons of Hagar, those wild asses, transgressed the
boundaries of peace and are slaying the good alongside the wicked. Brothers, let us
therefore consider Beth Hur, which was suddenly laid waste by the justice of the ruler
of all things, who chastises us in order that he may help us. (tr. M. G

Roman financial assistance to Peroz (mid-470s)
The whole of the last third of the fifth century saw the Persians embroiled in a bitter struggle
with their new eastern neighbours, the Hephthalite Huns. Peroz conducted several campaigns
against them, with little success: in the first he was captured, in the second he was killed. See
Szaivert 1987, Luther 1997: 121–4, Greatrex 1998a: 46–8.
Josh. Styl. 10 (243.8–13):By the help of the gold which he received from the
Romans, Peroz subdued the Huns, and took many places from their land and added
them to his own kingdom; but at last he was taken prisoner by them. When Zeno, the
emperor of the Romans, heard (this), he sent gold of his own and freed him, and rec -
onciled him with them.
80(tr. Wright, rev. M. Greatrex)
Border tensions in Mesopotamia in the 480s
Despite Zeno’s assistance, tensions on the Mesopotamian border heightened in the early 480s.
The emperor had to contend with an uprising in the eastern provinces in 484 led by Illus and
Leontius. The rebels sought the support of the Persians, although whether they obtained it is
uncertain; they did, however, receive the backing of all but one of the Armenian satraps. Edessa
resisted an attack by the rebels, who were soon defeated by Zeno’s generals (Joh. Ant. frg.214.2,
Josh. Styl. 15–17, Proc. Aed.III.1.25–6). See Brooks 1893: 227, 231, Stein 1949: 28–31. A
renewal of the persecution of Christians by Peroz prompted the outbreak of a revolt in
Persarmenia (see Chapter 4), and when in 483 the Persian catholicosBabowai sought to bring
the plight of Christians in Persia to the attention of Zeno, his message was intercepted at Nisibis
and he himself executed. See Blockley 1992: 84–5, Gero 1981: 98–107 (doubting that the
bishop of Nisibis, Barsauma, was responsible for betraying the catholicos, as later sources allege).
On Peroz’s persecution, see Labourt 1904: 129–30, Fiey 1977: 92–4, Rist 1996: 37. At the
same time, Zeno started to demand the return of Nisibis to Roman control (Josh. Styl. 18, cf.7).
See Chapter 1 p.5 and n.22.
The letters of Barsauma throw further light on the situation in Nisibis in the 480s. Two years of
drought forced Arabs allied to the Persians to encroach on Roman territory, almost leading to
full-scale war between the two powers; only by delicate negotiations was the situation restored,
as the letters make clear. The situation was exacerbated, according to Barsauma, by the presence
of anti-Chalcedonians (Monophysites) in Nisibis, who appear to have formed a fifth column,
eager to surrender the city to the Romans. See Gero 1981: 34–7, Blockley 1992: 85, Isaac 1992:
Barsauma, ep.3(to Acacius), p.528–9: ( … ) I have learnt, in fact, from the very
things which have happened to me, that as long as Nisibis is not in obedience to and
under the direction of the one who occupies the throne of the Holy Church of
Seleucia and Ctesiphon, the eastern region will endure damage and grave disasters.
Now Nisibis is troubled, shaken and tossed like a stormy sea, and if a letter of excom -
munication is not promptly despatched by Your Fatherhood and does not pacify the
city, I shall not maintain my bishopric of Nisibis; Nisibis itself will not remain subject
to the great empire of the Persians. There are, in fact, rebels, and if they are left at

Nisibis, they will destroy themselves and the country they inhabit. They have even
suggested to me the idea of rebellion, at the time when we were in opposition to the
seat of Your Fatherhood and in opposition to the two (cities),
82as I indicated above.
The marzban here supports these (p.529) rebels because he is unaware of their evil
intent. I fear that when they apprise him of their plan and inform him of their secret,
he will write to the king and tell him of the rebellion of these perverse men and that
the king will issue an edict against all Christians on account of the revolt of these
83The matter must be covered up as much as possible and not reach (the ears of)
people outside. With your great and divine authority, write a letter of excommunica -
tion in an imperious style and smite them, just as the doctor who does an amputation
on an ill person to deliver him of his illness. Threaten to denounce them to the king
and his officers in your letters if they do not pull back. ( … )
Barsauma, ep.2, p.526–7: (To patriarch Mar Acacius, the venerable one and friend of
God; your affectionate Barsauma. Peace in (the name of) our Lord.) We live in a country
considered beautiful by those who have not experienced it, where there are many
opponents of its tranquillity and where the obstacles to its prosperity cannot be
counted, especially at the present time. For now for two successive years we have been
afflicted by an absence of rain and a shortage of the necessary commodities.
assemblies of the tribes of the south are present here; the multitude of these people and
their animals has destroyed and laid waste the villages of the plains and the mountains.
They have dared to plunder and capture people and animals, even in the territory of
the Romans. A large army of the Romans then gathered and came to the frontier, with
their subjects the Tayyaye; they asked for compensation for what the Tu‘aye,
subjects of the Persians, had done in their country. The glorious and illustrious marzbanQardag Nakoragan
86restrained them from
this (demand) by his tact and wisdom. He made a pact for them, (undertaking) that
he would assemble the chiefs of the Tu‘aye and take back from them the plunder and
the captives, as soon as the Tayyaye of the Romans had returned the livestock and
captives which they had taken on several occasions in (the country of) Beth Garmai,
Adiabene and Niniveh, and they would give back some to the Romans and some to
the Persians;
87and he laid down a law in the treaty that they should establish bound -
aries (p.527), so that there would be no more of these evils, or other
s like them. But God knows when these things which were just mentioned will be completed!
With reference to and on account of these things, the king of kings has ordered the
king of the Tayyaye and the marzbanof Beth Aramaye to come here; but the leader of
the Romans, with all their soldiers and Tayyaye, remains on the frontier. At the start
of August, for the sake of peace and as a sign of great friendship we allowed the duxto
come to Nisibis, to the marzban; and he was received by him with great honour.
While they were drinking, eating and enjoying themselves together, four hundred
cavalry of the Tu‘aye had the temerity to go and fall upon the lesser villages of the
Romans. When this was heard about, it caused great distress to both sides – both to
the Romans and the Persians. The general and the nobles who had accompanied him
grew angry with us, for they believed that this had happened in order to insult the
Romans; (he thought that) they had entered Nisibis through our deceitf

It is therefore impossible for us at this time to respond to what Your Excellence
wrote until all the present differences between Romans and Persians have been
resolved. The marzbanwas completely unable to tolerate the matter, when he heard of
the king’s edict and your letter. If your wisdom heeds us, you will not summon the
other bishops there at this moment when famine reigns everywhere and there is a
shortage in every country, lest you should bring rumours and blame on yourself. But
this assembly which you are preparing to hold will be kept for the time after you have
gone up with the envoys to the country of the Romans and you have returned from
there. For thus the solid doctrine established at the previous reunion will give pleasure
to your knowledge when you have seen and learnt of the confusion which Satan has
sown in the country of the Romans. Even if you are absolutely determined that this
(assembly) should take place in the near future, I (shall) uphold by word and my deeds
all that they do at the assembly according to the laws of Christ; but it is not possible for
me to go there now on account of the difficulties over the frontiers. We have annulled
and completely destroyed the Synod of Beth Laphat from our letters; now we are for
ever the disciples and subjects of the throne of Your Fatherhood. May you be well in
Our Lord and pray for us.
88(tr. M. Greatrex)
Barsauma, ep.4, p.529: ( … ) After I received the edict in our favour sent by the mer -
ciful king of kings, on account of the grave matters taking place here between the
Romans and the Persians with regard to the frontiers – matters which demanded our
presence and diligence above all – the glorious and illustrious marzbanQardag
Nakoragan passed on our reply to the Gate of the king of kings and made this declara-
tion: ‘The Romans are going to send some of their leaders to make a treaty to maintain
peace on the frontiers; the bishop of Nisibis knows about the business of frontiers. His
presence is very necessary and he cannot come to the Gate until the treaty is con-
cluded.’ ( … ) (tr. M. Greatrex)
Further confirmation of tensions on the border comes from the Ecclesiastical Historyattributed
to Barhadbeshabba ‘Arbaya. The two major Christian communities in Persian territory, the
Monophysites and Nestorians, vied with one another for royal favour, and did their utmost to
impugn the loyalty of the other. The advantage was mainly on the side of the Nestorians, whose
doctrines were approved by a council in 486. See Fiey 1977: 118–19, Gero 1981: 52–3, Brock
1985: 126. Three years later, in 489, the School of the Persians at Edessa, a bastion of Nestorian
teaching, was expelled from the Roman empire. For as long as the Roman emperors tended
towards Monophysitism, it was an uneven struggle, and the Nestorians tended to get the upper
hand. See Segal 1970: 94–5, Blockley 1992: 85, Brock 1994: 75–6.
Barhadbeshabba, HE31 (PO 9.614):Once, when a messenger (of the king) was
passing by, certain people of the city accused him (Narses) of being a hater of the
kingdom (of Persia) and of spying for the Romans. When the messenger heard these
things, he promised that, once he returned from the land of the Romans, he would
crucify him. When the athlete (Narses) learned that punishment had been decreed
against him without a trial, he said, ‘If you return in peace, it is because the Lord has not
spoken to me’. And when he (the messenger) had left (Constantinople), had finished his

business and had reached Antioch, he died there in accordance with the saint’s word;
and he ceased his threat(s) and the (saint’s) accusers blushed. 90(tr. M. Greatrex)
Disruption in the Roman eastern provinces (490–500)
The first years of Anastasius’ reign (491–518) were troubled by events in the East. First, certain
powerful Isaurians who had been sidelined by the death of Zeno undertook a rebellion which
took the emperor’s generals until 498 to crush. See Brooks 1893: 231–7, Stein 1949: 82–4,
Elton 2000: 300. It was also around this time that the Ghassanids under Jafnid leadership
displaced the Salîhids as the Romans’ chief partner in the defence of the provinces against Arab
raids, although not without inflicting some damage on the provinces first. See Shahîd 1958:
147, Sartre 1982a: 157, Shahîd 1989: 282–5. One such raid penetrated as far as Emesa in
c.491. (Cyr. Scyth.Vit. Abram.1 [244]). See Sartre 1982a: 152, Shahîd 1989: 120. In the
meantime, successive Persian kings, most notably Kavadh, constantly requested funds from the
emperor. See Chapter 4 below.
Arab attacks resumed in earnest around the turn of the century. Our chief source for these
razzias is Theophanes, who evidently had access to good contemporary sources, for he provides
considerable details on who undertook the raids and what provinces were affected. From his
account it appears that the Lakhmids, Jafnids and the banu Tha‘laba were all involved in the
raids: it soon became clear that the Salihids were no longer capable of countering enemy tribes.
But around 502 a new configuration of alliances emerged, in which the Kindites, the Jafnids
and the tribe of Mudar were brought into the Roman orbit.
Evagrius, HEIII.36 (135.20–6): The Scenite barbarians launched assaults on the
Roman empire, but not to their own advantage; and they plundered the lands of Mes-
opotamia, of each Phoenice and of the Palestines. They suffered badly, however, at the
hands of those in command throughout (the region) and later kept peace with the
Romans, having agreed a truce en bloc.
Theoph. A.M. 5990 (141.1–17): In this year
92there was an invasion of the so-called
Scenite Arabs into Euphratesia. Eugenius, an earnest man in both word and deed,
who commanded the army in those parts, drew up (his forces) against them at a place
called Bithrapsa
93in the first region of Syria and defeated them in battle. The
vanquished (Arabs) were tributaries of the Persians of the tribe of the phylarch
94At that time Romanus was commander of the army in Palestine, an excel -
lent man. By good planning and generalship, he captured in battle Ogarus (Hujr), the
son of Harith (the latter being known as the son of Thalabane), together with a great
mass of prisoners.
95Before the battle Romanus had worsted and put to flight another
Scenite, Jabala by name, who had overrun Palestine before his arrival. 96At that time
also Romanus set free through fierce battles the island of Iotabe, which lies in the gulf
of the Red Sea and was subject to the Roman emperor, paying considerable tribute,
but which in the meantime had been seized by the Scenite Arabs, and gave it back to
the Roman traders to inhabit under its own laws, to import goods from the Indies and
to bring the assessed tax to the emperor.
97(tr. Mango and Scott, revised)
Theophanes, A.M. 5994 (143.21–5): In this year (501/2) there was again an incursion

incursion of the Saracens in Phoenice, Syria and Palestine. After the death of Hujr, his
brother Badicharimus (Ma‘dikarib) overran these regions like a hurricane and
retreated with the booty even more swiftly than he had invaded, so that Romanus,
who pursued him, could not catch up with the enemy.
98(tr. Mango and Scott, revised)
Theophanes, A.M. 5995 (144.3–6): In this year (502/3) Anastasius made a treaty
with Harith (known as the son of Thalabane),
99the father of Ma‘dikarib and Hujr,
and thenceforth all Palestine, Arabia and Phoenice enjoyed much peace and calm. (tr.
Mango and Scott, revised)
Rivalry between Nestorians and Monophysites in the Persian kingdom
It was around this same time that the rivalry between the Nestorians and Monophysites in
Persia came to a head. Deteriorating relations with Rome could only harm the Monophysite
community, since the Roman emperor himself was a Monophysite. Unsurprisingly therefore
the Nestorians exploited the charge that their opponents were disloyal subjects of the king.
See Joh. Eph.,Lives,PO17.146–8, on a debate (amid accusations of disloyalty) between
Simeon (a Monophysite) and Nestorians at the court of a marzbanbetween 499 and 504. On
the Roman side of the border, comparable accusations against prominent churchmen were also
made (Philoxenus, Letter to the monks of Senun, 94–5/79 [dating from 502/5]) and
Bardhadbeshabba, HE31 (PO 9.614), above. See Labourt 1904: 157–8, Guillaumont
1969–70: 42–3, Frend 1971: 201.
Joh. Eph. Lives,PO17.142: (The Nestorian bishops said to the king): ‘These men are
traitors to your majesty, as you can learn, since their faith and their rites (agree) with
those of the Romans’. It happened that the Magian (Kavadh) believed them, and he
ordered a persecution against the orthodox (Monophysites) in the whole of his
101 (tr. M. Greatrex)

The tightening Persian grip on Persarmenia (415)
The division of Armenia in 387 greatly reduced tensions between Rome and Persia in the
region, but Sasanian concerns over possible collaboration between Armenians and Romans
never receded entirely. When Vramshapur, the king of Persian Armenia, died in 414, he was
succeeded briefly by his brother Khosrov III, who had previously been removed from power for
his alleged Roman sympathies; and when he in turn died in the following year, the Persian king
Yazdgerd I appointed one of his sons, Shapur, to succeed him. See Grousset 1947: 178–9,
Chaumont 1987: 428–9 and Garsoïan 1989: 430. Cf. Mos. Khor. III.55.
´azar 18–19/52–3: And he (Shapur) plotted evil designs in his mind: first, this – that
the country of Armenia is large and productive and borders the neighbouring Greek
empire, under whose control many of the Arsacid nation are subject. Perhaps (he
thought) the people would become sympathetic to each other like brothers – that is,
those who were under our rule and those under Greek rule – and then trusting in each
other and forming an alliance, they might treat for peace with the Greek king, and
willingly becoming subject to him, they might rebel against us. Just as they have often
caused us trouble, by the increase of their numbers they would cause us even more
worry in war. And secondly, (he thought thus) because they (the Armenians) are
strangers to our religion and hate it, whereas they have the same cult and religion as
them (the Romans). (tr. Greenwood)
The foundation of Theodosiopolis (420–1)
But upon the death of Yazdgerd in late 420, Shapur did not succeed him. The Armenians took
the opportunity afforded them by the brief internal struggles in Persia to eject the Persian garri -
sons, and for three years Persian Armenia was in a state of anarchy (Mos. Khor. III.56, ´azar 19/
53). See Grousset 1947: 180 and Winkler 1994: 328. The Armenians remained in contact with
the Romans, in particular with Anatolius, the magister militum per Orientemat the time, who
was involved in the fortification of the border city of Theodosiopolis c.420. Roman interest in
the region clearly remained strong (Mos. Khor. III.57–9, cf. Koriun, V. Mesrop. 97).
1On the
war of 421–422 see Chapter 3.

Narratio de rebus Armeniae, 4–9:(4) In his (Arsaces’) days Armenia was divided,
when Theodosiopolis too was built (5) which was once in ancient times a village called
Kal¢ Arch¢. (6) For the great apostle Bartholomew, while going off to Parthia,
baptised the nephew of the king of the Persians and with him three thousand (people)
in the river Euphrates, (7) and built on the spot a church in the name of the most holy;
Mother of God (8) and when a town grew up in this place, he called it Kal¢ Arch¢. (9)
Theodosius the Great, when he beheld it and the water in it, was delighted and built a
very famous city, renaming it Theodosiopolis.
Despite the suspicions of his father, Bahram was content to accept an Arsacid king for Armenia,
Ardashir, son of Vramshapur. But he proved unpopular with some of the nobility, who accused
him of attempting to side with the Romans. He was deposed in 428 and replaced by a Persian
marzban; the Arsacids were never to return to the throne of Armenia (Mos. Khor. III.64, ´azar
24–5/59–60, Narratio, 16 [p.28]). No less importantly, the Armenian catholicosSahak was
ousted at the same time, as the Persian king (and the Persian Church) sought to win control of
Armenia (Mos. Khor. III.64–7, ´azar 24–5/59–60). See Blockley 1992: 60–1, Garsoïan and
Mahé 1997: 40–1, Garsoïan 1999: 59–65.
The defence of Roman Armenia (441)
As is evident from the following piece of legislation, the war of 440 (on which see above, Chap-
ter 3) had some impact on Roman Armenia. Although the dux Armeniaehad some troops at his
disposal, the sale of properties which had belonged to the Armenian crown had evidently led to
a reduction in the number of forces available for the defence of the region. Hence measures were
required to address the problems which had just been highlighted by the
Persian invasion.
Nov. Theod. V.3.1 (June 441): Therefore, since we learn through the well considered
report of Your Sublimity that the district of Armenia, which is situated almost upon
the very threshold of the border of the Persians, and which was formerly protected by
the troops and garrisons of the royal estates,
4has been exposed at the present time to
the invasions of the Persians, because the aforesaid estates, and especially those that are
adjacent or neighbouring to the municipalities of Theodosiopolis and of Satala, have
been transferred by the petitions of certain persons to their ownership at various times
in the past, and the original regular tax has been changed, We decree by this law which
shall live for ever that it must be established that all persons who obtained Our munifi -
cence in the matter (of these estates) shall be allowed to hold these landholdings as
granted, under the condition that they shall duly furnish to the account of the fisc the
5the baggage wagons, the supplementary post horses, the supplies, and also
the purchases enjoined upon them according to the former custom by the regulation
of your magnificent office, and all the other things, and they shall assume such pay -
ment as their responsibility. ( … ) (tr. Pharr, revised)

Persian repression in Persarmenia (449–51)
The king who had undertaken the war of 440, Yazdgerd II, showed himself to be increasingly
hostile to Christians in his kingdom as his reign progressed. In 449, urged on by his minister
Mihr-Narses, and encouraged by the Suanian prince Varazva³an, he required that the Armenians
embrace Zoroastrianism. See Chaumont 1987: 430, Rist 1996: 34–6. Cf. E³ ishe 15–27/69–80.
´azar, 42–3/78–9:(From a speech of Varazva³ an, a pro-Persian Armenian, to Yazdgerd
II.) And first and foremost, how suitable and productive is the great country of
Armenia, and likewise Georgia (Iberia) and Albania. Merely look at the advantages
which you receive from those countries. But (to) what is significant and urgent,
namely the salvation of so many lost souls, you pay no attention nor do you concern
yourself with it. You do not realise this – that you will have to give a reckoning to the
gods for such a great number of persons. For if you care about the salvation of so many
souls, know that their welfare and prosperity, which would come about, would bring
more profit and advantage to you than all the present prosperity of the kingdom
which you possess. And I see further profit and very great and significant advantages in
this matter for this country of the Aryans (Persia). Indeed you yourself, and all the
Aryans, know (p.43) how great and profitable is Armenia; it is a close neighbour of the
emperor’s realm and has the same religion and cult, for the emperor possesses
authority over them. (tr. Greenwood)
The programme of conversion which Yazdgerd introduced for Armenia proved intolerable to
the Armenian people. A rebellion broke out late in 449, and the following year was spent in
expelling the Persian garrisons. The leader of the uprising, Vardan Mamikonean, sought to
involve the Romans in the conflict, but without success; in 450 the Huns still posed a signifi-
cant threat to the eastern Roman empire, and matters were further complicated by the death of
Theodosius II and the accession of Marcian in the same year. See Grousset 1947: 194–9, Rist
1996: 34–6, Luther 1997: 141–2, Garsoïan 1998c: 1138–9.
´azar 63/105: Immediately (upon hearing of an imminent Persian attack) they
(Vasak, prince of Siwnik‘, and his associates) wrote letters to the emperor and to all the
nobles of the Greek court, and also to other princes and governors: to the bdeashkhof
A³ jnik (Arzanene), to the prince of Ange³ -tun (Ingilene), to Tsopk (Sophene) and
Hashteank‘ (Asthianene) and Eke³ eats‘ (Acilisene), and to other princes of their
respective regions, and to the great sparapetof Antioch.
7(tr. Greenwood)
Armenian forces were involved not only in Armenia itself, but also in Albania, in the eastern
Transcaucasus. Vardan Mamikonean seized control of the Derbend pass, and gave control of it
to the Albanians; in addition, he established friendly relations with the Huns north of the
Caucasus (´azar 66/108, E³ ishe 78/129–30). See Grousset 1947: 199–201.
8But in June 451
the Persians, assisted by Vasak and other Armenians loyal to them, were able to crush the rebel -
lion at the decisive battle of Avarayr (E³ ishe 116–21/169–73,´azar 69–73/112–17). See
Grousset 1947: 202–6.
9A few rebels continued to resist the Persians even after the defeat, oper -
ating from western and northern Armenia; some even managed to induce the Huns to destroy
the Persian force garrisoning the Derbend pass and to cause damage to Persian territories

beyond it (E³ishe 127–9/180–1). See Grousset 1947: 206–7. The situation in Persarmenia then
gradually stabilised under the marzbanAdhur-Hormizd, who proved more tolerant towards
Christians, while Vasak of Siwnik‘ was executed by Yazdgerd for alleged disloyalty (E³ ishe 134/
185, ´azar 83/129). See Grousset 1947: 209–11, Chaumont 1987: 430.
The defence of the passes through the Caucasus (450s–60s)
Although Yazdgerd II, like his successor Peroz, spent most of his reign campaigning in the
north-east of his kingdom, he was not unaware of the growing importance of the main Cauca -
sian passes. It had become evident during the Armenian rebellion of 449–451 that whoever
controlled them could bar the Huns from entering the Transcaucasus or could unleash them
against an enemy. At the same time, there was always the danger that the Huns, whether in col -
laboration with the Armenians or Romans or independently, might seek to force the passes and
annihilate the Persian garrisons stationed there.
11It is not surprising therefore to find the Per -
sians, who now held control of most of the Transcaucasus, making repeated attempts to
extract men or money from the Romans in order to bolster their defences of the passes. As the
Roman grip weakened, tensions arose in the 450s and 460s between the Lazi (in ancient
Colchis), who had traditionally come within the Roman orbit, and their allies, as the follow -
ing fragments from Priscus make clear.
Priscus, frg.33.1: The Romans went to Colchis, made war on the Lazi, and then the
Roman army returned home. The Emperor’s advisers prepared for a second campaign
and deliberated whether in pursuing the war they should travel by the same route or
through the part of Armenia bordering on Persian territory, having earlier persuaded
the monarch of the Parthians by an embassy (to allow this).
13For it was considered
wholly impracticable to take the sea route along the rugged coast, since Colchis had no
harbour. Gobazes himself sent envoys to the Parthians and also to the Roman
emperor. Since the monarch of the Parthians was involved in a war with the so-called
Kidarite Huns, he dismissed the Lazi who were fleeing to him.
14(tr. Blockley, revised)
Priscus, frg.33.2: Gobazes sent an embassy to the Romans. The Romans replied to
the envoys sent by Gobazes that they would abstain from war if either Gobazes himself
resigned his sovereignty or he deprived his son of his royalty, since it was not right that
both rule the land in defiance of ancient custom. That one or the other, Gobazes or his
son, should rule over Colchis and that war should cease there was the proposal of
Euphemius, the magister officiorum. Because of his reputation for sagacity and elo -
quence he was given oversight of the affairs of the Emperor Marcian and was his guide
in many good counsels. He took the author Priscus to share in the cares of his office. When the choice was put to Gobazes, he chose to hand over sovereignty to his son and
himself laid down his symbols of office. He sent envoys to the ruler of the Romans to ask
that, since Colchis now had one ruler, he should not take up arms in anger on his account.
The emperor ordered him to cross to the land of the Romans and give an explanation of
what he had decided, and he did not decline to set out. But he asked that the emperor
should hand over Dionysius, who had earlier been sent to Colchis over the disagreement
with this same Gobazes, as a pledge that no harm should befall him. Therefore, Dionysius
was sent to Colchis, and they composed their differences. (tr. Blockley, revised)

Peroz, occupied with constant warfare in the north-east, kept up diplomatic pressure on the
Romans to assist in the defence of the Caucasian passes. His repeated demands for money or
troops may have elicited occasional contributions, although these are generally reported only
in late sources.
Priscus, frg.41.1.3–27:An embassy also arrived from the Persian monarch which
complained to the Romans both about those of their people who were fleeing to them
(the Romans) and about the Magi who had lived from old in Roman territory.
embassy alleged that the Romans, wishing to turn the Magi from their ancestral cus -
toms, laws and rites of worship, harassed them and did not allow the fire, which they
called unquenchable, to be kept burning continually according to their law. It also
said that the Romans, through a contribution of money, should show interest in the
fortress of Ioureiopaach,
17situated at the Caspian Gates, or they should at least send
soldiers to guard it. It was not right that the Persians alone should be burdened by the
expense and the garrisoning of the place, since if they gave up (these expenditures),
evil would befall not only the Persians but also the Romans at the hands of the neigh -
bouring peoples. They further said that they (the Romans) should help with money in
the war against the so-called Kidarite Huns, since they (the Romans) would derive
benefit from their victory insofar as (that) people would be prevented from penetrat-
ing to the Roman Empire also. The Romans replied that they would send someone to
discuss all these issues with the Parthian monarch. They claimed that there were no
fugitives amongst them and that the Magi were not harassed on account of their reli-
gion, and said that since the Persians had undertaken the guarding of the fortress of
Iouroeipaach and the war against the Huns on their own behalf, it was not right that
they demand money from them (the Romans). Tatian, who held the rank of patrician,
was sent as ambassador to the Vandals on behalf of the Italians, while Constantius,
who was prefect for the third time and a patrician as well as a consular, was sent to the
18(tr. Blockley, revised)
Roman involvement in Lazica continued in the 460s, at which time the small (but strategically
important) kingdom of Suania succeeded in breaking away from Lazic control, an event which
was to have repercussions in negotiations between Rome and Persia a century later. There were
stirrings too in Persarmenia, where Vahan Mamikonean, nephew of the rebel leader of
450–451, sought help from the Romans. Although Leo agreed to lend his support, none was
Priscus, frg.44: After the fire in the city during the reign of Leo, Gobazes came with
Dionysius to Constantinople, with Persian clothing and a bodyguard in the Median
manner. The officials at court received him and at first blamed him for his rebellion,
but then treated him in a kindly manner and dismissed him. For he won them over
both by the flattery of his words and by bearing with him the symbols of the Chris -
20(tr. Blockley, revised)
Priscus, frg.51: A very serious dispute existed between the Romans and Lazi and the
nation of the Suani. The Suani were making war against … ,
21and the Persians wished
to go to war with him because of the forts that had been captured from the Suani. 22He

(Gobazes) therefore sent an embassy (to the Romans), asking that reinforcements be
sent to him by the emperor from amongst the troops who were guarding the borders
of that part of Armenia which was tributary to the Romans. Thus, since these were
close at hand, he would have ready assistance and would not be endangered while
waiting for troops to come from a distance; nor would he be burdened with the
expense of supporting them if they came and the war was postponed, should it turn
out in this way, as had happened earlier. For when Heraclius was sent with help, and
the Persians and Iberians, who were at war with him, were diverted to fighting other
active peoples,
23he dismissed the reinforcements since he was worried about the
provision of food. As a result, when the Parthians returned against him, he again
called upon the Romans. When the Romans had replied that they would send help and a man to command
it, an embassy arrived from the Persians which announced that the Kidarite Huns had
been vanquished by them and that their city of Balaam had been captured. They
reported their victory and in barbaric fashion boasted about it, since they wished to
advertise the very large force which they had at present. But when they had made this
announcement, the emperor straightaway dismissed them, since he was more con -
cerned about the events in Sicily.
24(tr. Blockley, revised)
The end result of these disputes between the Suani and their traditional overlords, the Lazi, was
that the latter were soon afterwards brought into alignment with Persia, while the former
retained their independence and their loyalty to the Romans. See Zuckerman 1991: 543,
Greatrex 1998a: 126.
25While Peroz may have been gratified by his success in gaining control of
almost the entire Transcaucasus, his joy will have been blunted by the burden this imposed on
his armed forces: for since he now ruled the region south of the Derbend and Dariel passes, it
fell to him to defend it. And, as is clear from Priscus and Caucasian sources, numerous tribes
were active in trying to penetrate the passes in this period.
Priscus, frg.47: The Saraguri, having attacked the Akatiri and other peoples, marched
against the Persians. First they came to the Caspian Gates, but when they found that the
Persians had established a fort there, they turned to another route, by which they came
to Iberia.
27They laid waste their (the Iberians’) country and overran Armenia. As a result
the Persians, apprehensive of this inroad on top of their old war with the Kidarites, sent
an embassy to the Romans and asked that they give them either money or men for the
defence of the fortress at Iouroeipaach. They repeated what had often been said by their
embassies, that since they were facing the fighting and refusing to allow access to the
attacking barbarian peoples, the Romans’ territory remained unravaged. When the
Romans replied that each had to fight for his own land and take care of his own defence,
they again returned having achieved nothing.
28(tr. Blockley, revised)
The escalation of Persian demands for subsidies (470s–80s)
Leo’s successor Zeno proved more sympathetic, at least initially, to Peroz’s requests. Joshua is
the first source to refer to a mutual assistance agreement between the two powers; a similar

engagement is reported by Malalas, who gives the text of a letter of the Persian king Kavadh
referring to the agreement. See Blockley 1992: 75 and n.32, Greatrex 199
8a: 15–16. 29
Josh. Styl. 8 (242.10–23):Further, there was a treaty between the Romans and the
Persians, that, if they had need of one another when they had a war with any of the
nations, they should help one another, by giving 300 strong men, with their arms and
horses, or 300 staters in lieu of each man. This was (to be) according to what the side
that had the need wished. Now the Romans, by the help of God, the Lord of all, did
not need assistance from the Persians; for there have been believing emperors from
that time until the present day, and by the help of heaven their power has been
strengthened. But the kings of the Persians have been sending ambassadors and receiv -
ing gold for their need; but it was not in the way of tribute that they received it, as
many thought. (tr. Wright, rev. M. Greatrex)
Mal. 18.44 (378.32–9/449.19–450.6): see Chapter 6, p.88, below.
Josh. Styl. 9–10 (242.23–243.2): For in our days the Persian king Peroz, because of
the wars that he had with the Chionites who are the Huns,
30very often received gold
from the Romans, not however demanding (it) as tribute, (but) by inflaming them
with zeal,
31as if (p.243) he was carrying on his (the emperor’s) fights on their behalf,
‘that’, (he said), ‘they may not pass over into your territory.’ (tr. Wright, rev. M.
Revolt in Persarmenia (482)
Roman subsidies availed Peroz little. As early as 471 he deposed the Armenian patriarch Giwt,
suspecting him on account of the links he had with the Romans (´azar 113–16/165–9). See
Garsoïan 1999: 129. Towards the end of his reign, he not only persecuted the Christians, but
also undertook further campaigns against the Hephthalite Huns. While he was away in the
north-east, a rebellion broke out in Iberia in 482. The Armenians soon followed suit, and
gained the advantage initially. Persian armies were deployed in 483, and succeeded in restoring
the situation. The response of the rebel leader, Vahan Mamikonean, was to take refuge in the
western regions of Armenia. Here, just as the Romans had operated with circumspection in
attacking Lazica in the 450s, the Persians had to proceed with care (´azar 118–54/172–213).
See Sanspeur 1975–6, Luther 1997: 142–3, Greatrex 1998a: 127–8,
Garsoïan 1999: 130.
´azar 136/193: In pursuit of them (Vahan and his forces, then in Tayk c) came
Mihran with his army. He was attempting either to slay them in battle or to seize them
by some means, or to persuade them to submit. The camp of the mass of the Aryan
(Persian) forces was in the village called Du, on the border of Persian and Roman terri -
tory. The Armenian commander Vahan Mamikonean was about two leagues distant
in a village called Mknarinch with around one hundred men, or perhaps a few more or
less. (tr. Greenwood)
Mihran, the Persian general, therefore sought to negotiate with Vahan, rather than attacking
him. 32Other Persian generals, however, actually entered Roman territory in their pursuit of
the rebels. Zarmihr crossed the border to the north of Theodosiopolis in search of Vahan

before proceeding against the Iberian king Vakhtang; but the king eluded the Persians by
also moving west, into Lazica. Shapur of Ray, another Persian general, actually attacked
some labourers in Roman territory south of Theodosiopolis when he was foiled in his
efforts to seize Vahan.
33In a battle fought in this same campaign,´azar reports that a certain
´ erpagos died, ‘who was Greek by origin’: some assistance from Roman Armenia and the
satrapies, perhaps unofficial, may then have been forthcoming. It should be stressed,
however, that even at this point, one hundred years after the partition of Armenia, the borders
between Roman and Persian territory will hardly have been obvious or precise. And while
Shapur was worried in case Vahan should obtain help from the Armenians of the satrapies, it
appears that these same regions were becoming involved in the insurrection of Illus and
Leontius against Zeno.
The waning of Persian power (484–98)
In late summer 484 the situation changed abruptly. Peroz was decisively defeated once again by
the Hephthalites, and on this occasion perished in battle (Josh. Styl. 11, Proc. I.4, ´azar 154–6/
35His successor, Balash, proved more conciliatory. Vahan was appointed marzbanof
Armenia, and Christianity tolerated; for their part, the Armenians helped Balash to defeat a
challenger to the Persian throne, Zarer (´azar 170–5/232–7). See Grousset 1947: 226–9,
Luther 1997: 129–30, 143, Garsoïan 1998c: 1139–40. Notwithstanding this aid, Balash was
faced with an intractable problem – the emptiness of the Persian trea
Josh. Styl. 18 (247.25–248.12): After he (Peroz) had been sought out and not found,
which I have mentioned above, 36his brother Balash reigned over the Persians in his
stead. He found nothing in the Persian treasury, (p.248) and his land was laid waste
by the depopulation (carried out) by the Huns. It is not forgotten by Your wisdom
what expenses and outlays kings incur in wars, even when they are victorious, to say
nothing about when they are defeated, and from the Romans he had no help of any
kind such as his brother had. For he sent ambassadors to Zeno, (asking him) to send
gold; but because he was occupied with the war against Illus and Leontius, and
because he also remembered the gold that had been sent by them at the start of their
rebellion, which still remained there in Persia,
37he did not wish to send him anything,
but he sent word to him, ‘The taxes of Nisibis which you receive are enough for you,
which for many years past have been due to the Romans’.
38(tr. Wright, rev. M.
Balash’s lack of funds and unpopularity hastened his downfall. In 488 he was deposed and
replaced by his nephew, Kavadh. But since he too had no source of funds, he was obliged to
make repeated demands of the Romans; unlike Balash, he believed it best to adopt a more
aggressive tone in his requests.
Josh. Styl. 19 (248.21–8): ( … ) For he (Kavadh) sent an ambassador, and a large ele -
phant as his present to the emperor, that he might send him gold. But before the
ambassador reached Antioch in Syria, Zeno died, and Anastasius became emperor
after him.
39When the Persian ambassador informed his master Kavadh of (this)

change in the Roman kingdom, he sent him (word) to go up with diligence and to
demand the customary gold, or to say to the emperor, ‘Accept war.’ (tr. Wright, rev.
M. Greatrex)
Anastasius refused, demanding that Nisibis be returned before he handed over any gold (Josh.
Styl. 20). As a result of his failure, Kavadh, like Balash before, found his position weakening.
Despite the concessions made to the Armenians by Balash, a rebellion broke out in 491. Joshua
ascribes the uprising to discontent with Persian attempts to force Zoroastrianism upon the
Armenians once again, although this would have been in contravention of the recent agree-
ment. The Armenians may therefore have been persuaded to rise up instead in the wake of a
Church council at Va³ arshapat at which they expressed their doctrinal unity with the Church of
Constantinople. See Luther 1997: 143–4, Garsoïan 1999: 133, 164–
Josh. Styl. 21 (249.15–23): When the Armenians who were under the rule of Kavadh
heard that he had not received a peaceful answer from the Romans, they took courage
and strengthened themselves, and destroyed the fire-temples that had been built by
the Persians in their land, and killed the Magi who were among them. Kavadh sent
against them a marzbanwith an army to exact capital punishment from them and
make them return to the worship of fire; but they fought with him, and destroyed
both him and his army, and sent ambassadors to the emperor, in order to become his
subjects. He however was unwilling to receive them, that he might not be assumed to
be provoking war with the Persians. ( … ) (tr. Wright, rev. M. Gre
Like all his predecessors before him in the fifth century, Anastasius refused to undertake any-
thing which might be seized on by the Persians as an excuse to break the peace. In the short
term, his refusal to bow to Kavadh’s demands paid off: not only the Armenians, but also the
Qadishaye (tribesmen living in the vicinity of Nisibis) and Tamuraye (in Persia) revolted
against the king (Josh. Styl. 22). See Luther 1997: 145–6, Greatrex 1998a: 76–7 and n.18.
The Persian kingdom was also disturbed by the Mazakites, members of a revolutionary sect,
to which the king gave some support (Josh. Styl. 20, Proc. I.5.1, Tabari, I, 885–7/132–6).
See Winter and Dignas 2001: M11, Crone 1991, Luther 1997: 137–41, Wiesehöfer 1996:
173, 208–10.
40Kavadh, however, having failed in another attempt to extract money from
Anastasius, was ousted from power in favour of his brother Zamasp, and only returned to the
throne in 498/9 with the assistance of the Hephthalite Huns (Josh. Styl. 23–4, Proc.
I.6.10–11). The stage was then set for the oubreak of hostilities in 502.

The causes of the war
Several factors underlay the termination of the longest period of peace the two powers ever
enjoyed. The Persian king Kavadh, having only just regained his throne in 498/9 with the assis -
tance of the Hephthalite Huns, had more need now of funds than ever: his allies required pay -
ment. The situation was exacerbated by recent changes in the flow of the Tigris in Lower
Mesopotamia, sparking famine and floods. When Anastasius again refused to provide any help
(in 501 or 502), Kavadh had little choice but to try to gain the money by force. See Morony
1995: 76, Luther 1997: 177–8, Greatrex 1998a: 51–2, 76.
Theoph. A.M. 5996 (144.21–3): In this year Kavadh, the Persian king, demanded
money from Anastasius. Anastasius said that if Kavadh wanted a loan, he would make
a written agreement; but if it were in any other form, he would not pay.
1(tr. Mango
and Scott, revised)
Proc. WarsI.7.1–2 (30.7–16): A little later Kavadh was owing the king of the
Hephthalites a sum of money; since he was not able to pay it back to him, he asked the
Roman Emperor Anastasius to lend him this money.
2He took counsel with some of
his friends and asked whether this should be done. (2) They did not allow him to
make the agreement, for they declared that it was disadvantageous to strengthen the
friendship of their enemies with the Hephthalites with their own money; rather it was
better (for the Romans) to stir them up against each other as much as possible. (tr.
Dewing, revised)
The outbreak of war (502)
Once restored to the throne, Kavadh enlisted the peoples who had revolted against him – the
Tamuraye, the Qadishaye, the Arabs and the Armenians; some joined his invasion force will -
ingly, others less so (Josh. Styl. 24). The king invaded Roman territory in August 502, quickly
capturing the capital of Armenia Interior, Theodosiopolis, perhaps with local support; the city
was in any case undefended by troops and weakly fortified (Josh. Styl. 48, Mal. 16.9 [398], cf.
Proc. I.7.3). See Luther 1997: 178–9, Greatrex 1998a: 79–80.

Zach.HEVII.3 (22.15–22): But Kavadh, who ruled after him (Peroz), and his
nobles, had a grudge against the Romans, as they said that they had caused the coming
of the Huns, and the pillage and devastation of their country. And Kavadh mustered
an army, and went out against Theodosiopolis in Armenia of the Romans, and
subdued the city; and he dealt with its inhabitants mercifully, because he had not been
insulted by them; but he took (prisoner) Constantine, the ruler of their city. (tr.
Hamilton and Brooks, rev. M. Greatrex)
The siege of Amida (502–3)
Anastasius only learnt of Kavadh’s preparations after his invasion. He immediately despatched
Rufinus to try to dissuade the king from his attack. But Kavadh, having seized Theodosiopolis,
quickly moved southwards, overrunning the vulnerable Armenian satrapies. He paused at
Martyropolis, where the satrap, Theodore, handed over two years’ worth of taxes from the
satrapy of Sophanene (Proc. Aed.III.2.4–8).
3Kavadh therefore left Theodore in charge and
continued to Amida. Here Rufinus met him, but was unable to deflect him from his course.
The siege of Amida, testament to Kavadh’s continuing appetite for plunder, proved to be a far
more difficult enterprise than he expected; the defenders, although unsupported by troops,
repelled his assaults for three months before they were finally beaten (Proc. I.7.5–35, Josh. Styl.
50, 53, Chr. 1234, 51 [147–50]). Zachariah of Mytilene provides one of the most detailed
accounts of a siege on the eastern frontier.
4On the siege itself see Luther 1997: 180, 182–6,
Greatrex 1998a: 83–94 and the notes in Trombley and Watt 2000: 53–63, 120 with map 5.
Zach. HEVII.3–4 (22.22–31.14): In the month of October he (Kavadh) came to
Amida of Mesopotamia. (But though he assailed it) with fierce assaults of sharp arrows
of bows, and with battering-rams, which pushed the wall to overthrow (it), and roofs
of hide, which protected the men who had piled up the mound and raised (it) up and
made (it) equal in height with the wall, for three months, day after day (p.23) yet he
could not take the city by assault. Meanwhile the people with him had great hardship
through work and fighting, and he heard in his ears the abuse of disorderly men on the
wall, and ridicule and mockery, and he was reduced to great straits. (He felt) indigna -
tion and regret, because the winter came upon him in its severity, and because the Per -
sians, being clad in their loose garments, appeared inefficient; and their bows were
greatly relaxed by the moisture of the air;
5and their battering-rams did not hurt the
wall or make any breaches in it, for (the defenders) were tying bundles of rushes from
the beds with chains, and receiving upon them the force of the battering-rams, and
preventing them from breaching the wall. But they (the defenders) breached the wall
from inside, and carried the material of the mound inside from the outside, and they
gradually propped up the cavity with beams from underneath. And when chosen Per -
sian soldiers ascended the mound and placed beams upon the wall to effect an
entrance (now they had armour, and the king was near with his army outside, and was
supporting them with arms
6and shooting of arrows, and giving encouragement with
shouting, and stimulating them and exhorting them by his presence and appearance;
they were about five hundred men [in number]), (the defenders) threw strings of skin

flayed from an ox, and soaked vetch mixed with myrrh-oil, from the wall upon the
beams; and they poured the liquid from the vetch upon the skins to make (them) slip-
pery, and placed fire among the props which were placed beneath (p.24) the mound.
When they had fought with each other for about six hours, and (the besiegers) had
been prevented from effecting an entrance, the fire blazed up and consumed the wood
of the props, and immediately the rest of the material was reduced to ashes by the vio -
lence of the fire, and the mound was destroyed and fell. The Persians who were on top
of it were burned, and they were also bruised, since they were struck (by stones) from
the wall. And (so) the king withdrew with shame and grief, since he was more and more
mocked and insulted by those daring, proud and boastful men. For there was no
bishop in that city to be their teacher and to keep them in order, (since) John the
bishop, a chaste and noble man of great character, had died a few days before. This
man had been summoned from the monastery of Qartmin, and he came (to Amida)
when he was elected and became bishop. However, he did not change his asceticism
and self-mortification and way of life, but was faithful in service by day and by night.
He warned and rebuked the rich men of the city in the days of the famine and the
incursion of the Arabs and the pestilence, when he said that they should not keep back
the corn in the time of distress, but should sell (it) and give (it) to the poor, in case
when they kept (it) back, they might only be hoarding it for the enemy, according to
the word of Scripture. And so, in fact, it happened. To him an angel appeared openly,
who was standing on the side of the altar-table, and he foretold to him concerning the
incursion of the enemy, and that he should be taken away as a righteous man from the
presence of the enemy; and he (John) revealed the saying when he published it in the
presence of the people of the city, that they might turn and be saved from the wrath. (p.25) The fourth chapter of this seventh book explains how the city of Amida was sub-
dued, and what happened to its inhabitants. When Kavadh and his army had been con-
quered in the various assaults which they had made upon the city, and many of his
soldiers had died, his power gave way and (so) he asked that a small gift of silver should
be given to him and he would withdraw from the city. But Leontius, the chief council -
lor, son of Pappus, and Cyrus the governor, and Paul Bar Zainab the steward,
through the men whom they sent to Kavadh, requested from him the value of the
garden vegetables which his army had eaten, and for the rest of the corn and wine
which they had gathered and brought from the villages. When he was greatly grieved
at these things, and was preparing to withdraw since he had been disgraced, the Mes -
siah appeared to him in a vision of the night, as he (himself) afterwards related, and
said to him that within three days He would hand over to him the inhabitants of the
city, because they had sinned against Him; and this happened in this way
. On the western side of the city by the Tripyrgion was a guard of monks who were
appointed from the monastery of John Urtaye, and their archimandrite was a Persian.
On the outside, opposite their watch-tower, one marzban, Kanarak the Lame, (was)
encamped. Day after day, as he vigilantly watched by night and by day, he was diligent
and clever in devising craftily how he could subdue the city. For there was a man
whom they called in the city Kutrigo, a seditious and thievish man; this (man) was

very daring in all kinds of attacks upon the Persians, and he made raids and snatched
away from them cattle and possessions, (p.26) so that they too, being accustomed to
hear those on the wall crying out, called him Kutrigo. Kanarak watched this man, and
perceived that he went out by the streams next to the Tripyrgion and plundered, and
went in again. For a time the Persians let him accomplish his will, as they noted and
examined his actions, and they ran after him and saw from where he came out and
went in.But it happened that on that night when the city was subdued, there was darkness,
and a black cloud sending down soft rain;
9and a man charitably made entertainment
for the monks who guarded the Tripyrgion, and he gave them wine to drink in the late
evening. Because of this, sleep overtook them, and they did not watch diligently upon
their guard, according to their usual custom. When Kanarak and a few of his army
came up as he pursued Kutrigo, and drew near to the wall, the monks did not cry out
or cast stones. The man perceived that they were asleep, and he sent for ladders and for
his army; and those with him went in by the streams, and ascended the tower of the
monks, and slew them. They took the tower and the rampart; and they set up ladders
against the wall, and sent (a message) to the king. But when their neighbours, who were guarding another tower, heard it, they cried
out; they sought to come to the monks who were being killed, but could not; but some
of them were wounded by arrows from the Persians, and died. When the report was
heard by Cyrus the governor, he came up; and torches were close to him and he was
easily (struck with arrows (p.27) which (themselves) carried torches).
10He was struck
and wounded by an arrow from the Persians, who stood in the darkness but were
unhurt; and he withdrew when he was wounded. But when it was morning, and the
king and his army reached the place, they set ladders against the wall, and he ordered
the army to go up. Many of those who went up died, since they were wounded by
arrows and by stones and driven back by spears. Those who through fear turned from
the ladders and went down were killed at the king’s behest as cowards and fugitives
from the battle. And on account of this the Persians were encouraged and determined
either to be victorious by conquering and subduing the city, or, after being wounded
in the actual conflict, not to be blamed and killed by their king, who was near and was
a spectator of their struggle. But the people of the city tried to loose from beneath the
keystone of the arch of the tower in which the Persians were, and they were in the act
of loosening the supports. Meanwhile another tower was subdued, and another and
another in succession, and the guards of the wall were killed. But Peter, a man of huge stature, from ‘Amkhoro, held the battlement of one side,
since he was clad in an iron coat of mail; and he did not allow the Persians, and those
who assailed him from outside and inside, to cross over, and repelled and hurled
(them) back with a spear, since he was firmly fixed (in place) and stood like a hero
until at last, when five or six towers on another side were subdued, he too fled and was
not killed. The Persians (thus) first got possession of the whole wall and held it; and
for a night (p.28) and a day and the following night (they) killed and drove back the
guards. At last they went down and opened the gates, and the army entered after
receiving the king’s command to destroy men and women, and all classes and ages, for

three days and three nights. But a certain Christian prince of the country of Arran 11
pleaded with the king (on behalf of) a church called the Great (Church) of the Forty
Martyrs; 12and he spared it, since it was full of people. After three nights and days the
slaughter ceased by the king’s command. Men went in to guard the treasure of the
church and of the nobles of the city, that the king might have whatever was found in
them. He ordered also that the corpses of those who were slain that were in the street,
and of those whom they had crucified, should be collected and brought round to the
northern (side) of the city, so that the king, who was outside on the south side, might
enter in. (So) they were collected, and they were numbered as they went out, eighty
thousand, besides those that were heaped up in the taverns and thrown into the
streams, and those left in the courtyards. Then the king entered the treasury of the
church, and when he saw there an image of the Lord Jesus, who was represented in the
likeness of a Galilean, he asked who it was. They said to him, ‘It is God’; and he bowed
his head before it, and said, ‘He it was Who said to me,
13“Stay, and receive from Me
the city and its inhabitants who have sinned against Me.”’ But he took away many
holy vessels of gold and precious items of clothing of Isaac Bar Bar‘ai, a consul and a
rich man of the city, which had come to the church by inheritance a few years before.
But he found there also good wine dried into its dregs, which was brought up (p.29)
and placed in the sun for seven years, and at last it became dry; from this the stewards,
when on journeys, were accustomed to take some, once it had been pounded, in clean
linen pouches. They would pour out a little of it (to make) mingled wine and water,
and drank (it) with the sweetness and taste of wine. They told the ignorant it was holy
water. The king admired it greatly, and took it. And (so) the technique of (practising)
this gluttony was lost to the sons of the church from this time. The gold and silver of the great men’s houses, and the beautiful garments, were
gathered together and given to the king’s treasurers. They also brought down all the
statues of the city, and the clock-towers and the marble; and they collected the bronze
and everything that pleased (them), and they placed (them) upon wooden rafts that
they made, and sent (them) by the river Tigris, which passes by the east of the city and
penetrates into their country. The king sought for the chiefs and nobles of the city;
and Leontius, and Cyrus the governor, who had been wounded by an arrow, and the
rest of the nobles, were brought to him; but the Persians had killed Paul Bar Zainab
the steward so that he did not reveal to the king that they had found much gold in his
possession. They clothed Leontius and Cyrus in filthy garments, and put swine-ropes
on their necks, and made them carry pigs, and led them along, while proclaiming and
exhibiting them, saying, ‘Rulers who do not rule their city well nor restrain its people,
so that they do not insult the king, deserve such insult as this.’ But at last the nobles and
all the chief craftsmen were bound (p.30) and gathered together, and set apart as the
king’s captives; and they were sent to his country with an army which brought them
14But knowledgeable men of the king’s army drew near and said to him, ‘Our
kinsmen and brothers were killed in battle by the inhabitants of the city,’ and they asked
him that one-tenth of the men should be given to them for the exaction of vengeance.
They gathered and counted some of the mass of men and gave to them (this group
selected) from the men; and they put them to death, killing them in every way.

The king bathed himself in the bath of Paul Bar Zainab, and after winter he departed
from the city. He left in it Glon the general as ruler, and two marzbans, about 3000
soldiers to guard the city, and John Bar Habloho, one of the rich men, and Sergius Bar
Zabduni, to rule (the people). (tr. Hamilton and Brooks, rev. M. Greatrex)
During the siege Kavadh ordered his Lakhmid allies, under their king Na‘man, to ravage the
area around Carrhae, while another detachment struck the region around Constantia; these
forces encountered Roman resistance, under the duxOlympius, but emerged victorious from a
battle at Tell Beshme. Na‘man’s forces, however, penetrated as far as Edessa. The duxof
Armenia, Eugenius, meanwhile succeeded in recapturing Theodosiopolis (Josh. Styl. 51–2).
See Luther 1997: 181–2, Greatrex 1998a: 87–9.
Rufinus, who had been detained by Kavadh during the siege, was then allowed to depart west -
wards. While the news he brought with him caused alarm in the cities of the East, Anastasius
had already reacted to news of the siege (Proc. I.8.1–5, Josh. Styl. 54, Theoph. A.M. 5997,
146–7 [below]).
A composite account of the events of 502
Theoph. A.M. 5996 (145.24–146.15): As a result 17Kavadh violated the peace treaty
that had previously been made with Theodosius the younger, and with a large army of
Persians and foreigners invaded first of all Armenia, and captured Theodosiopolis,
which was betrayed by Constantine, a senator who had been commander of the Illyr-
ian detachments. Next he went to Mesopotamia and besieged Amida, since no
Roman army worthy of note was yet stationed in that region: only Alypius was there
with a small force. He was praised by everyone and was a lover of philosophy and took
such care as he could both for the defence of the cities and the stores of food. He him-
self lived at Constantina (Constantia), 507 stades to the west of Nisibis and an equal
distance from Amida to the north. But some time having meanwhile elapsed, and sev-
eral engagements having taken place between the Romans and the Persians, in which
the Romans sometimes got the worse and sometimes the better of the foreigners in
different places, Amida was finally betrayed to the Persians, after being besieged for
over three months by the barbarians. It was betrayed during the night by monks who
were guarding one of the towers. The enemy used ladders and thus gained entrance to
the city, plundered and destroyed it all and captured considerable riches. King Kavadh
came to the city on an elephant on the third day after its betrayal and removed great
riches. He left Glon to guard the city while he himself returned to the city of Nisibis
and the Persian forces remained between Amida and Constantia.
18(tr. Mango and
Scott, revised)
The Roman reaction
Josh. Styl. 54 (281.6–27): ( … ) The emperor Anastasius too, when he heard (of the
fall of Amida), sent a large army of Roman soldiers to winter in the cities and garrison
them. All the booty that he had taken, and the captives that he had carried off, were

not, however, enough for Kavadh, nor was he sated with the great quantity of blood
that he had shed; but he (again) sent ambassadors to the emperor, saying, ‘Send me
the gold or accept war.’ This was in the month of April. The emperor, however, did
not send the gold, but made preparations to avenge himself and to exact satisfaction
for those who had perished. In the month of May he sent against him three generals,
Areobindus, Patricius and Hypatius, and many officers with them. Areobindus went
down and encamped on the border by Dara and Ammodius, facing the city of Nisibis;
he had with him 12,000 men. Patricius and Hypatius encamped against Amida, to
drive out from there the Persian garrison; they had with them 40,000 (men). There
came down too at this time thehyparchosApion, who dwelt at Edessa, to look after the
provisioning of the Roman troops that were with them. As the bakers were not able to
make (enough) bread, he ordered that wheat should be supplied to all the houses in
Edessa and that they should make the soldiers’ bread (bucellatum) at their own cost.
The Edessenes turned out at the first baking 630,000 modii.
19(tr. Wright, rev. M.
The Roman force assembled was the largest the eastern front had seen since Julian’s invasion of
Persia. While Areobindus initially gained the upper hand in his advance upon Nisibis, his army
was eventually forced back from there in July by superior Persian numbers; an appeal to
Patricius and Hypatius for support went unheeded, and so he quickly withdrew westwards
(Josh. Styl. 55). His two colleagues eventually came to his aid, but arrived too late: Areobindus
had already retreated further west, leaving his camp at Apadna to be plundered by the Persians.
The two generals therefore returned to Amida (Josh. Styl. 56). See Luther 1997: 187–9,
Greatrex 1998a: 96–8.
Zach. HEVII.4–5 (30.14–31.1): And then in the summer the Romans came, and
their chiefs of the army were Patricius the great strat¢gos, an old man, upright and
trustworthy, but with slight intelligence, and Hypatius and Celer the magister, (and)
at last also Areobindus; there was with them also the comesJustin, who ruled after
Anastasius. They gathered and made war on the city (Amida) with wooden towers and
trenches and all kinds of engines. They set fire also to the gate of the city, which was
called that of the house of Mar Za‘uro, (in order) to come upon the Persians, but they
were hindered because they (the Romans) were resting, and they did not go in for the
Persians shut (the gate).
20And the Romans did not subdue it nor take it from them by
assault, although the men who were found in it were afflicted by famine, day after
(p.31) day, until at last the people in it were eating one another.
Theoph. A.M. 5997 (147.6–24): ( … ) The army gathered in Edessa, a city of
Osrhoene, and at Samosata, a city of Euphratesia.
21The forces under Hypatius and
Patricius were engaged in freeing Amida from the Persian garrison. Areobindus, cam -
paigning with Romanus, the phylarch Asouades and several others against Kavadh
himself, who was then staying at Nisibis, prevailed against the Persians in various bat -
tles and drove Kavadh out of Nisibis, making him retreat many days’ journey inside its
22In one engagement there fell the greatest of the Persian generals, 23whose
sword and bracelet were brought to Areobindus by the Scythian who had killed him,

and then sent to the emperor, a noteworthy and particularly clear token of victory.
And so with the Persian army defeated to this extent by the Roman generals, Kavadh
got ready and sent against the Romans a very large army, with the result that
Areobindus urged the forces under Hypatius and Patricius to hasten to his help from
the area round Amida. When they declined out of envy, Areobindus wanted to with-
draw and to return to Byzantium; (and he would have done so) had not Apion, the
Egyptian, who was second in command of the army and in charge of supplies and gen -
eral supervision, with difficulty kept him in those parts. (tr. Mango and Scott, revised)
Meanwhile Timostratus, the duxof Osrhoene, based in Callinicum, inflicted a defeat on the
Lakhmid forces, and Arabs allied to Rome – the Tha‘labites – even undertook an attack on the
Lakhmid capital at Hira (Josh. Styl. 57). Cf. Robin 1996: 689–90, 697–9, Luther 1997:
189–90, Greatrex 1998a: 99 and n.77, Whittow 1999: 212–13.
Kavadh’s counter-offensive (summer 503)
In August, as Kavadh led a renewed invasion from Nisibis, Patricius and Hypatius moved
southwards to join forces with Areobindus in the Tur Abdin. They failed to meet up, and while
Areobindus decided to withdraw westwards again, the forces of the other two generals were
decisively defeated in the area between Apadna and Tell Beshme by Kavadh and retreated
towards Samosata (Josh. Styl. 57, Marc. com.a.503, Proc. I.8.11–19, Zach. VII.5 [below],
Theoph. A.M. 5997 [below]). Kavadh then proceeded due west, coming first to Constantia.
There, according to Josh. Styl. 58, the Jewish population sought to betray the city to the
invaders, but were thwarted by a message from a Roman officer who was a prisoner of the
Persians. Prominent in boosting the morale of the city’s defenders was the bishop Bar-Hadad.
Kavadh, frustrated in his wish to capture the city, but having received some supplies from its
inhabitants (Proc. II.13.8–15), continued west to Edessa; he arrived in its vicinity in early
September. The king attempted to extort 10,000 lbs of gold from the city, but Areobindus
refused his demands. Although elements of the Persian army, especially the Arabs, overran
much of Osrhoene, attempts to storm the city failed (Josh. Styl. 59–60). Following further
unsuccessful negotiations, Kavadh was forced to withdraw because of a lack of provisions and
the approach of further Roman reinforcements (Josh. Styl. 61–3); passing by Callinicum, he
followed the Euphrates back to Persian territory (Josh. Styl. 64). The legendary impregnability
of the city had been vindicated again. See Luther 1997: 191–202, Greatrex 1998a: 99–106,
Trombley and Watt 2000: 77–82.
Theoph. A.M. 5997 (146.24–147.16): 25While the generals were at odds with one
another, Kavadh came to Nisibis (p.147) and, having learned of the discord among
the generals, he himself, in a powerful position with his large force, divided up his own
army in many places and overran almost the entire Roman territory, so to speak, push -
ing as far as the Syrias themselves. Meanwhile he sent many envoys to Areobindus
about peace, saying that he would end the war on payment of money. He therefore
overran in particular the territory round Edessa, where Areobindus was. However, he
did not accomplish anything successfully there but, contrary to expectations, he came
off worst in a battle with Areobindus. Knowing, too, that his general Glon, with the

garrison at Amida, had been destroyed following a plot against him, 26he marched
back in distress along another route, neglecting the hostages whom he had given to
Areobindus during the peace-talks. They also retained, contrary to the agreements,
Alypius, that excellent man, and Basil of Edessa;
27so that after his retreat to his own
territory (since winter had already arrived), the Roman generals divided themselves
among the various cities of Euphratesia, Osrhoene, Mesopotamia, Syria and Armenia,
and encamped for the winter season. (tr. Mango and Scott, revised)
While Kavadh invaded Osrhoene, another Roman commander, Pharesmanes, continued to
harry the Persian garrison in Amida. Zachariah’s account concentrates on this, but describes
some of the other episodes of the campaign as well.
Zach. HEVII.5 (31.7–33.21): King Kavadh, as stated above, when he departed
with his army from Amida to his own country, left in it Glon, a general, and two
marzbans, and about 3000 soldiers to guard the city; and also two or three rich
men and some (other) independent people. These the Roman generals did not
overcome, nor did they subdue the city and take it. But at last Patricius went down
to Arzanene of the Persians, and carried off captives, and subdued fortresses in it.
And Areobindus and Hypatius (went down) to Nisibis and did not subdue it,
although the citizens were pleased with the Romans, and appeared lazy in the
29However, when the king heard, Kavadh came with an army against the
Romans; and they fled from him, and they left their tents and the heavy equip-
ment which was with them. Areobindus (fled) from Arzamon by Apadna, and
Hypatius and Patricius and others from Thelkatsro. And they lost many horses
and the riders who were on them, who fell from the cliffs of the mountains, and
were bruised, and perished, and were injured.
Pharesmanes alone, a warlike man, was successful in battle several times; he was
renowned and formidable against the Persians, and his name terrified them. His deeds
destroyed and weakened them, and they were found to be cowards and fell in his pres-
ence (p.32). This (man) at last came to Amida with 500 horsemen, and he observed
the Persians who went out to the villages, and he killed some of them, and he took the
animals which were with them, and also their horses. Now a certain cunning man, whose name was Gadono, of the town of Akhore,
whom I know, introduced himself to him, and made a compact with him, that when
he acted treacherously, he would bring out to him (Pharesmanes), on some pretext,
Glon, the Persian general, with three or four hundred horsemen. Because this man
Gadono, who was mentioned (above), was a hunter of wild animals and partridge and
fish, he was accustomed to go freely to Glon, carrying in his hands a present of game
for him; and he ate bread in his presence, and received from him goods of the city
equivalent to the game. At last he told him that there were about 100 Romans and 500
horses nearly seven miles from the city, at a place called Fold of the Shepherds; and as a
friend he advised him to go out and take possession of the beasts, having killed the
Romans, and to make a name (for himself). He (Glon) sent scouts, and they saw a few
Romans and the horses, and returned and informed him. Then he made preparation

and took with him 400 horsemen, and this Gadono upon a mule; and he (Gadono)
led him (Glon) and set him in the midst of the ambush of the Romans, who were lying
in wait. So the Romans massacred the Persians, and they brought away the head of
Glon to Constantia.
So then distress and rage (seized) the son of Glon, and themarzbanswho were there
became cautious; they did not allow the inhabitants who happened to be shut up in
the city to go out to the market, which was held beside the wall by peasants from the
villages. (These peasants) brought wine and wheat (p.33) and other produce, and sold
(them) both to the Persians and to the citizens, while horsemen were stationed by
them, and escorted them one at a time and conducted them in. By an excellent law of
the Persians, no one ventured to take anything from the villagers: they sold what they
liked and received payments and goods from the city. They assembled in the market
enthusiastically. However, on account of the slaughter of Glon and the horsemen, the
market ceased. And the nobles who were left in the city, and about 10,000 people,
were seized and shut up in the Stadium, and they were guarded there, and there was
no food; some of them died. They ate their shoes and their excrement, and they drank
their urine. At last they assailed one another; and when they had been on the point of
death for some time, those who were left in the Stadium were let loose like the dead
from graves in the midst of the city. Starving women, who were found in groups, laid
hold of some of the men by means of solicitation, deceit and trickery, and overcame,
killed and ate them; and more than 500 men were eaten by women. The famine which
was in this city increased, (so that) the suffering surpassed the blockade of Samaria and
the destruction of Jerusalem, which is written in Scripture and Josephus relates. (tr.
Hamilton and Brooks, rev. M. Greatrex)
Lakhmid raid on Palestine (late 503)
While Roman forces were heavily engaged in Mesopotamia, the new Lakhmid ruler Mundhir
undertook a daring raid further south. Mundhir’s father Nu‘man died in August or September
503, and this raid probably took place shortly afterwards. See Shahîd 1995: 17–19, Greatrex
1998a: 107.
Cyr. Scyth., Vit. Ioh.13, 211.15–19: Around this time Mundhir the son of Sikika
(Shaqiqa), 32who had acquired the dignity of king over the Saracens subject to the
Persians, invaded Arabia and Palestine in great anger against the Romans, laying
everything waste, enslaving countless thousands of Romans, and committing many
lawless acts, following the capture of Amida.
The Roman response (503–4)
In summer 503 Anastasius despatched further reinforcements to the East, under the command
of the magister officiorum Celer; he also cancelled the taxes due from Mesopotamia and
Osrhoene. When he arrived in Hierapolis in late September, Hypatius and Apion were recalled

and the other commanders assigned to winter quarters. During the winter Patricius moved
from Melitene to Amida in order to prevent the Persians there from replenishing their supplies;
although a Persian force was sent to drive him away, he defeated it and
invested the city (Josh
Styl. 65–6). At this point Kavadh was obliged to leave the front to deal with a Hunnic incur-
sion, while Celer brought increasing pressure to bear on the remaining Persian forces. From his
assembly point at Resaina, Celer moved north in spring 504 to join Patricius in besieging
Amida, having meanwhile ordered a successful raid on Persian livestock in the Jebel Sinjar
(Josh. Styl. 69). But the defences of Amida, despite the sufferings of the inhabitants, proved too
strong for the besiegers, who were unable to sustain the siege in winter (Josh. Styl. 70–3). See
Luther 1997: 202–4, Greatrex 1998a: 111–12.
While the siege of Amida continued, Celer harried Beth Arabaye, the Persian territory east of
Nisibis (Josh. Styl. 79, Marc. com.a.504). Areobindus conducted a simultaneous punitive raid
in Arzanene (Josh. Styl. 75). The weakness of the Persian position was highlighted by the defec -
tions of the renegade Constantine (Josh. Styl. 74), an Arab chief Adid, and the Armenian
Mushlek and his army (Josh. Styl. 75). See Trombley and Watt 2000: 93 n.443. This year, not
only was Amida’s tribute remitted, but the taxes of both Mesopotamia and Osrhoene were
cancelled again; a reduction was also granted to Hierapolis, where Roman troops had mustered
(Josh. Styl. 78). See Greatrex 1998a: 112–14. Now that the war was spilling over into Persian territory, Kavadh had no interest in
prolonging it. He sent a spahbadhto negotiate the surrender of Amida and returned the Roman
hostages taken earlier. Although initially Celer preferred to continue the siege of Amida,
believing its capture to be imminent, he decided to agree to a truce when the Roman army
proved incapable of sustaining the siege during the winter. A truce was therefore agreed, to be
ratified by both rulers (Josh. Styl. 71–3, 75–7, 80–2). See Luther 1997: 206–7, Greatrex
1998a: 114–15.
Theoph. A.M. 5998 (147.31–149.7): In this year themagisterCeler was sent out by
the emperor with a very large force under him and took over almost all authority
(p.148) together with the general Areobindus. To them the emperor entrusted the
management of the entire war. He recalled Apion and Hypatius with all speed to
Byzantium, thinking them to be unnecessary for the army because of their hostility
towards the general Areobindus, and he appointed the general Calliopius to be in
charge of supplies. Accordingly Celer managed the whole war extremely well together
with Areobindus, Patricius, Bonosus, Timostratus, Romanus, and the others in their
various regions; for he was a man of good sense and learning of every kind, filled with
God’s grace, and brave, an Illyrian by race, from where Anastasius also came. Many
forts in Persian territory were overrun in the incursions and destroyed by fire or by
other means, so that even Nisibis itself nearly fell to the Romans. For hunger was by
then seizing the Persians, and besides a tribal uprising of the so-called Kadousioi and
other races had occurred. Thus, in short, the Romans prevailed over the Persians, so
that Kavadh sent the general Aspetius
33to discuss peace urgently with the Romans,
(instructing him,) even if he gained little or nothing in return, to hand back Amida to
the Romans. (For) despite their enormous effort, they had not yet been able to capture
it from the Persians, even though hunger was oppressing the garrison, because of the
nature of the site and the unbreachable (nature) of the walls. But the generals, seeing

that winter was approaching again, and judging that it was preferable to redeem the
Roman army for a few talents from the harsh wintry conditions of those places where
the discussions with Aspetius took place, handed over three talents
34and got back
Basil of Edessa, who was still being held a hostage by the Persians (the excellent
Alypius had died after suffering an illness among them), and returned (p.149) the hos -
tages whom they held. They won back Amida and made the covenants for the peace
on the border between the forts of Amudis (Ammodius) and Mardin, and confirmed
them in writing. Such was the end of Anastasius’ Persian war in the 15th year of his
reign. It had lasted for three years and had harmed the territory of the Persians more
than previous wars, reaching this conclusion in the 15th year of Anastasius’ reign. (tr.
Mango and Scott, revised)
Zach. HEVII.5 (33.21–34.20): But at last Pharesmanes came to the city, and he
made a treaty with the Persians who were in it, for they too were weak. The chiefs of
the Romans and the Persians sat by the gate of the city, while the Persians went out
carrying as much as they could, and they were not searched. If any of the citizens
accompanied them, they were asked whether they wished to remain or would like
(p.34) to go with the Persians. So the evacuation of the city (took place). But 1100 lbs
of gold was given to Kavadh by Celer the magisterfor the ransom of the city and for
peace. And when the documents (were drawn up) they brought the drafts to the king
for his signature. The king fell asleep, and it was told him in a vision that he should not
make peace; (therefore) when he awoke, he tore up the paper, and departed to his
country, taking the gold.
But Pharesmanes remained in the city, since he governed its inhabitants and the
country. Now a remission of tribute was (granted) by the emperor for seven years. The
emperor dealt mildly with the inhabitants of the city: on those who returned from
captivity he bestowed gifts lavishly, and he received (them) peaceably, every man
according to his rank. (So) the city was at peace and was inhabited; and more building
was done to the wall. There was sent to the city a quiet and pleasant man, Thomas, a
monk, councillor (and) merciful bishop, by the advice of Dith. Again the providence
of God called and conveyed there Samuel the Just from the monastery of the
36, a miracle-worker and a ‘dissolver of doubts’; and he upheld the city by his
prayer and also aided its inhabitants. (tr. Hamilton and Brooks, rev. M
. Greatrex)
An account of the whole war
John the Lydian provides a very negative assessment of the Roman performance in the war.
Since he was commissioned to write a history of Justinian’s war of 527–532, he no doubt aimed
to diminish the achievements of earlier emperors, just as Procopius and Marcellinus comesdid.
See Greatrex 1998a: 73–6.
Joh. Lyd. De Mag.III.53 (214.7–20): 37In his (Anastasius’) time, when the elderly
Kavadh led out the whole of Persia against the Romans, war broke out. And though
the Romans were capable of victory by force, through the profligacy and weakness of
the last Areobindus (for he was a lover of song, of the flute and of the dance) and the

inexperience and cowardice of the generals Patricius and Hypatius, they were at first
beaten when the Persians suddenly overwhelmed them. Subsequently, when they (the
Romans) had pursued the Persians and liberated Amida again, which had been cap-
tured, the Persians began talks with Celer, who was the magister officiorumfor
Anastasius, concerning Biraparakh, as we mentioned before, and the expense incurred
by the Persians alone on its account. The strife came to an end when certain modest
favours were granted to Kavadh by Anastasius; for the magnanimity and organisation
of Anastasius piously sustained the loss for the sake of peace.
38(tr. Carney, revised)
Roman refortification work and the construction of Dara (505)
While Kavadh and Anastasius were consulted about the truce, the Romans took the opportu -
nity of upgrading their dilapidated fortifications. Improvements were made at Edessa, Batnae
and Amida, and the governor of Osrhoene, Eulogius, erected a wall to defend the fortress at
Batnae (Josh. Styl. 81, 89, 91). The Arab allies of either side attempted to continue hostilities,
carrying out raids in (hitherto) enemy territory; but both the Romans and the Persians pun -
ished their allies severely for these breaches of the truce (Josh. Styl. 88). Undoubtedly the most
important construction project undertaken by the Romans in this period was that at Dara, to
the north-west of Nisibis, just inside the Roman frontier; work was begun there in late 505. We
are fortunate to possess a wealth of sources concerning why and how it was founded. See Croke
and Crow 1983: 148–59, Whitby 1986b: esp. 751–2, Isaac 1992: 254–5, Gregory 1997: C6,
Luther 1997: 210–12, Greatrex 1998a: 115–16.
Josh. Styl. 90 (309.12–310.3): The year 817 (505/6). The generals of the Roman
army informed the emperor that the troops suffered great harm from their not
having any (fortified) city situated on the border. For whenever the Romans went
forth from Constantia or Amida to go about on forays in the ‘Araba, they were in
constant fear, whenever they halted, of the treachery of enemies; and if it happened
that they encountered a larger force than their own, and thought of turning back,
they had to endure great weariness, in that there was no city near them in which they
could find shelter. For this reason the emperor gave orders that a wall should be built
for the village of Dara, which is situated on the frontier. They selected workmen
from all Syria (for this task), and they went down there and were building it; but the
Persians kept going out from Nisibis and stopping them. (p.310) On this account
Pharasmenes set out from Edessa, and went down and dwelt at Amida, and went
forth to those who were building and gave them aid .(…)(tr. Wright, rev. M.
Zach. HEVII.6 (34.24–38.15): The Emperor Anastasius brought heavy criticisms
against the Roman strat¢gosand commanders (p.35) who set out to the royal city after
the war with the Persians, because they did not, according to his will, under the Lord,
prosper and succeed in the war, and conquer the Persians or drive them out from
Amida, save by the gifts and the gold that were sent from him. And they made the
excuse that it was difficult for the strat¢lat¢sto fight with a king who, according to the
word of God, although he was an Assyrian and an enemy, was sent by the Lord to the

country of the Romans for the punishment of sins, and, moreover, on account of the
greatness of the army which was with him; and again, that it was not easy for them
even in his absence to subdue Nisibis, since they had no engine ready, nor any place of
refuge in which to rest. For the fortresses were far away and were too small to receive
the army, and neither the water in them nor the vegetables that were left (there) were
sufficient.They (therefore) begged of him that a city should be built at his command on the
side of the mountains, as a place of refuge for the army and for their rest, and for the
preparation of weapons, and to guard the country of the Arabs from the forays of the
Persians and Tayyaye. Some of them spoke on behalf of Dara, and some on behalf of
39Then he sent (a message) to Thomas the bishop of Amida, and he
despatched engineers and he drew up a plan; and this holy Thomas brought (it) up
with him to the emperor. It semed good to the emperor and the nobles that Dara
should be built as a city. At that time (p.36) Felicissimus was dux, a brave and wise
man; and he was not covetous (or) a lover of money, but was upright and a friend of
the peasants and the poor. Now King Kavadh was fighting with the Tamuraye and
other enemies of his country. The emperor gave gold to Thomas the bishop as the
price of the Church’s village; and he bought it for the treasury. He liberated the settlers
who were in it, and granted to (each of) them his land and his house. For the building
of the church of the city he gave several gold centenaria. He promised with an oath
that he would give with a liberal hand everything the bishop might spend, and that he
would not deal falsely. And at last he made a royal decree without (resort to) a
summary, providing that the city be built immediately according to the word of the
bishop; so the craftsmen and slaves and peasants who were required for the collection
of material there gained and were blessed. He (Anastasius) sent (to the site) many
stonecutters and masons, and he ordered that no man should be cheated of the wages
of his labour, because he rightly perceived and cleverly understood that by that agree-
ment a city would quickly be built upon the frontier. When they began by the help of
the Lord and made a start, there were there as superintendents and commissaries over
the works the presbyters Cyrus ‘Adon and Eutychian, and Paphnutius and Sergius
and John the deacons, and others from the clergy of Amida. The bishop also was
(often to be) found there on visits.
40Gold was given in abundance without any
restraint to the craftsmen and for work of every kind (p.37) in this manner: there was
(for each) labourer four keratiaa day, and if he had an ass with him, eight.
41And from
this many grew rich and wealthy. Since the report was published abroad that (the
system) was honest and that the wages were given, from the East to the West workmen
and craftsmen gathered together. In addition, the superintendents in charge of the
work were well provided for and their purses were filled, for they found the man
(Thomas) very generous, gentle and kind: he believed in the just emperor and in his
promises which he (had made) to him. In two or three years the city was built, and, as we may say, suddenly sprang up on
the frontier. When Kavadh heard (of this) and sought to bring (it) to a halt, he could
not, for the wall had been raised and constructed (high as) a protection for those who
took refuge (behind it). A large public bath and a spacious storehouse were built, and a

conduit which came along the lower part of the mountain, and wonderful cisterns
within the city to receive the water. 42Persons to hasten (the work) were (sent) from the
emperor to the bishop, and they all brought back excellent reports of his integrity and
justice to the emperor. (So) he was greatly pleased with the man, and sent gold when
he answered his request, fulfilling (it) without delay. At last the hundred pounds
which he sent was counted out, and the bishop sent (a report) in writing to the
emperor, declaring in the presence of God that (the money) had been spent upon the
work, and that no part of it remained in his hand or had reached his church. He
(Anastasius) readily sent him a royal decree, which the treasury approved, (acknowl -
edging that) all the gold which had been sent by him (had been spent) on the building
work in the city. And Dara was completed, and it was named Anastasiopolis, after the
name of the just emperor. He swore (p.38) by his crown that no statement of accounts
(should be) required of Thomas or of his church, either by himself or those who were
appointed emperors after him. And he (Thomas) appointed there and consecrated as
first bishop Eutychian the presbyter, a diligent man, and accustomed to business
affairs; and he gave the privilege of (certain) rights to his church (taken) from the juris -
diction of the Church of Amida.
43To him was attached John, a Roman soldier from
Amida. Eutychian tonsured him and made him a presbyter and master of the hostelry;
and, when he went up to the royal city, he (John) was with him. When he was pre-
sented to the emperor, he (Anastasius) gave (him) an endowment for his church.
Abraham bar Kaili of Tel Bande
44was notarius at this time, who was the son of
Ephraem of Constantia, and he too was attached to Eutychian the bishop, who made
him a presbyter. He was sent as superintendent of the work and the building of the
bath, and eventually he became steward of the church.
Marc. com.a.518: Dara, a city of this kind, was founded in Mesopotamia. Dara, a cer-
tain estate located 60 miles south of Amida and 15 miles west from the town of
Nisibis, paid its dues to the church of Amida. So the Emperor Anastasius bought the
houses of this modest village at a fixed price for the purpose of founding a city there,
and he immediately despatched outstanding craftsmen and ordered (the city) to be
built. He put Calliopius, who was later patrician of the city of Antioch, in charge of
this work.
45Indeed, with remarkable insight, he marked out with a hoe a furrow for
locating the foundations on a hill ending up on level ground; and he protected (it) on
all sides by the construction of very strong walls as far as (this) boundary. So too he
enclosed the stream called Cordissus, which takes its name from the estate near which
it originates, and which winds its way as it roars along, and at the fifth milestone it
divides the same hill and the new city, falling into a concealed entrance at each end.
Moreover, he allowed the city which had been endowed with public buildings to
retain the name of the village. 47The city’s huge look-out, called the Herculean tower,
built on higher ground and connected to the walls, looked up to Nisibis to the east
and looked back to Amida to its north.
48(tr. Croke, revised)
Joh. Diakrin. 558 (157.9–11): The Emperor Anastasius built Dara, and after he built
it in a vision he beheld the apostle Bartholomew, saying that he was entrusted with the
defence of the city. And for this reason he (Anastasius) sent for his relic and deposited
it there.

Joh. Lyd.De Mag. III.47 (206.22–8): While many were the things which Anastasius
did in the service of the common good, one alone will suffice as an illustration – the
city established by him beyond the Euphrates (the natives call it Dara, (but) among us
they call it the city of Anastasius after him). If God had not clamped down on the
throat of the Persians through that man (Anastasius), the Persians would have taken
over the lands of the Romans, since they border on them. (tr. Carney, r
Proc. I.10.13–18 also describes Anastasius’ construction of Dara. Cf. Proc. Aed.II.1.4–3.26
(mostly on subsequent improvements by Justinian, playing down Anastasius’ work). See
Capizzi 1969: 216–21, Mundell 1974: 219, Whitby 1986b, Greatrex 1998a
: 120–1.
Despite these measures and remissions of tax by the emperor (Josh. Styl. 92, cf. Proc. I.7.35 and
Anecd. 23.7), many inhabitants of the frontier region remained discontented, chiefly because of
the behaviour of Gothic soldiers in the Roman army. Billeted on the population of Edessa, and
no doubt elsewhere, they had oppressed the population and nearly killed the duxRomanus
when he tried to limit their exactions (Josh. Styl. 93–4, 96). See
Greatrex 1998a: 115–16.
The conclusion of hostilities (506)
Negotiations between the two powers continued, interrupted by the death of the Persian negoti-
ator early in 506. In September that year, the two sides met on the frontier by Dara, each accom-
panied by their troops. Such was the distrust that remained that the Romans, suspecting
treachery, seized the Persian officials; once released, the Persians preferred to stay in Nisibis (Josh.
Styl. 97). At last in November 506 terms were agreed. Little is known of what these terms were.
Procopius (I.9.24) states that peace was made for seven years; it is likely that some payment was
made to the Persians (cf. Josh. Styl. 81 on gifts from Anastasius to Kavadh), but there is no reason
to suppose that the annual payments demanded by the Persians were conceded by the emperor.
See Luther 1997: 213–16 (who believes in Roman payments) and Greatrex 1998a: 117–18.
The aftermath of the war (506–18)
Celer returned west following the conclusion of negotiations and entered Edessa in triumph
(Josh. Styl. 100). Although no further large-scale conflict took place during Anastasius’ reign,
tensions remained, especially while work continued at Dara. Once Kavadh had concluded his
war with the Huns, he accused the Romans of breaching the treaty agreed in 422 (see Chapter
3, p.42), by which no further fortifications were to be erected near the frontier. But Anastasius
pursued the project nonetheless, deflecting the king’s complaints with money; Kavadh was in
any case unable to stop the work. (Proc. I.10.17). The walls were completed by 507/8; later,
probably during the reign of Justin, the city became the seat of the duxof Mesopotamia and a
bishop. See Whitby 1986b: 751, 769, Greatrex 1998a: 121.
The development of Sergiopolis and other sites
Although Justinian is well known for his strengthening of the fortifications of the east,
Anastasius too undertook important work in the region. The monastery of Mar Gabriel at

Qartmin, on the southwest flank of the Tur Abdin, was expanded with the assistance of the
emperor; work was finished in 512. See Whitby 1986a: 726, Palmer 1990: 1
Between 514 and 518 the city of Resafa, already an important shrine to the growing cult of St
Sergius and a meeting point for Arab tribesmen, was granted the status of a metropolis and
renamed Sergiopolis. It is possible that the emperor was also responsible for the construction of
the great basilica there (Joh. Diakr. 554, p.156.17–19). See Capizzi 1969: 214–15, Whitby
1986a: 724, 726, Whitby 1987: 102–5, Key Fowden 1999: 64–5, 92 and Chapter 3 generally.For epigraphic evidence of work on frontier defences (notably at Emesa), see Chapter 16
below and Greatrex 1998a: 122.
The northern frontier (506–18)
Already during the war against Kavadh, the Pontic region had suffered at the hands of the
mountainous Tzani, perhaps taking advantage of Roman involvement elsewhere (Theod. Lect.
HE 466). After the war, Anastasius undertook extensive work at Theodosiopolis in Armenia,
although Procopius plays down its importance in his Aed.(III.5.4–9, cf. WarsI.10.19). See
Capizzi 1969: 206–7, Whitby 1987: 106–7, van Esbroeck 1997: 367–8. The city of Euchaita
in Helenopontus was also fortified by Anastasius, probably before 515. See Mango and
®ev©enko 1972: 380–1 and Trombley 1985: 83 n.28.
50In the wake of a raid by Sabir Huns
across the Caucasus mountains in 515, work was started at Melitene, but probably only
completed during the reign of Justin I; fortifications in Cappadocia, where the invaders did
much damage, were also strengthened (Proc. Aed.III.4.19–20, Mal. 16.17 [406]). See Whitby
1987: 101, Greatrex 1998a: 122.
Further north still, in the Transcaucasus, the Persian grip was loosening. Although Kavadh had
brought Iberia under control once more around 500, an invasion by the Sabirs distracted him
from his war against the Romans in 504. An indication of growing Roman influence in the
region is provided by the offer of the Hunnic king Ambazuces to hand over control of the
Caspian Gates (i.e. the Dariel pass) to Anastasius, probably soon after the raid of 504 (Proc.
I.10.9–12). The emperor refused, however, and Kavadh continued to rebuild his influence in
the Transcaucasus, suppressing a minor Armenian uprising in 513/14, erecting fortifications,
building a new capital for Albania at Partaw, and installing a marzbanat Mtskheta in Iberia
around 520. See Kramer 1935–7: 613, Greatrex 1998a: 128–30 with bi
Southern Arabia
The conflicts between Ethiopians and Himyarites which occurred in the early sixth century in
modern-day Yemen lie beyond the scope of this work. The involvement of the Romans and
Persians was extremely limited: the former were called upon by the Ethiopians to assist them
against a Jewish Himyarite ruler, Dhu Nuwas, who had persecuted the Christians of his king -
dom. Some Roman ships sailed to help in the invasion of Himyar in 525, which swiftly over -
threw Dhu Nuwas. But the Ethiopian grip on Himyar did not last long: the ruler appointed by
the Ethiopian king was ousted by a certain Abraha in 530 or 531, who went on to reign for
approximately forty years. See Vasiliev 1950: 283–302, Smith 1954, B. Rubin 1960: 297–319,
Z. Rubin 1989: 383–420, Brakmann 1992: 753–62, Greatrex 1998a: 226

The revival of tensions (518–25)
The accession of Justin I was greeted in the East by a raid of the Lakhmid Arabs on Osrhoene in
519 or 520, perhaps following renewed demands for funds from Kavadh (Chr. 724, 111/14,
830, Mich. Syr. IX.16 [270–1a/178]). Probably in the course of this raid the Roman
commanders Timostratus and John were taken prisoner by the Lakhmid chief Mundhir (Proc.
I.17.43–5). The capture of aduxand another high-ranking officer is an indication of the return
of the Roman frontier to a state of ill-preparedness.
51Further south, in Palestine and Arabia,
Arab raids also caused problems, such that the aged St Sabas in 530 appealed to the Emperor
Justinian to build a fortress to protect the monks here, a request Justinian granted (Cyr. Scyth.
V. Sab. 72–3 [175, 178]). See Magoulias 1990–3: 301–3 (noting some other incidents), Shahîd
1995: 92, Di Segni 1999: 151.
The conference at Ramla (524)
The Emperor Justin despatched the presbyter Abraham to ransom the captured Roman
commanders, and at a significant gathering of Arab leaders and Roman and Persian diplo -
mats at Ramla, south-east of Hira, in February 524, their release was negotiated. The ‘con -
ference at Ramla’ was probably preceded in late 523 by discussions at Hira. Despite Justin’s
persecution of Monophysites within the Roman empire, Abraham intervened to support
them in Persia; their representative at the conference was the energetic Simeon of Beth
Arsham. Both gatherings also attracted envoys from southern Arabia. At Ramla, an envoy
of the newly installed Jewish ruler of Himyar, Masruq (also known as Dhu Nuwas), who
had unleashed bloody persecutions of the Christians at Najran and elsewhere, called upon
both the Persian king and Mundhir to persecute the Christians under their control.
Although Mundhir was a pagan, he certainly had Christians among his followers, and Dhu
Nuwas’ request went unheeded. See Vasiliev 1950: 278–83, Shahîd 1964: 115–31, De
Blois 1990 (on the dating), Shahîd 1995: 40–2, Evans 1996: 112–14, Greatrex 1998a: 131,
The forces of nature and doctrinal disputes had more impact on the eastern provinces in the
520s than any Lakhmid raids. Edessa, for instance, was struck by a devastating flood in 525,
while an earthquake destroyed Antioch in May 526. The Monophysite communities mean -
while met with increasing harassment from bishops and imperial troops. See Stein 1949:
241–2, Vasiliev 1950: 221–41, Frend 1972: 247–52, 257–61, Gr
eatrex 1998a: 131–2.
Roman successes in the Transcaucasus
The waning of Persian influence in the Transcaucasus was spectacularly confirmed in 521 or
522. See Whitby and Whitby 1989: 106 n.330, Scott 1992: 162–3, Braund 1994: 276–7,
Greatrex 1998a: 132 n.35.
Mal. 17.9 (340.53–341.94/412.16–414.16) (cf.Chr. Pasch. 613.3–615.4 and other
chronicles noted by Jeffreys and Scott 1986: 233): During his reign Ztathius (Tzath),
the king of the Lazi, grew angry and departed from Persian territory. (This was) while
the Persians were ruled by Kavadh, a friend of Ztathius, king of the Lazi, who had

once been subject to the rule of Kavadh. Thus whenever a king of the Lazi happened
to die, his successor, though from the race of the Lazi, was appointed and crowned by
the king of the Persians. (p.413) The king of the Lazi had rejected the belief of the
Hellenes and so as not to be appointed by Kavadh, the Persian king, and not to
perform sacrifices and all the other Persian customs, as soon as his father Damnazes
died, he immediately went up to the Emperor Justin in Byzantium, put himself at his
disposal and asked to be proclaimed king of the Lazi and to become a Christian. He
was received by the emperor, baptized, and, having become a Christian, married a
Roman wife named Valeriana, the granddaughter of Oninus the patrician and former
curopalat¢s, and he took her back with him to his own country. He had been
appointed and crowned by Justin, the Roman emperor, and had worn a Roman impe -
rial crown and a white cloak of pure silk. Instead of the purple stripe it had the gold
imperial border; in its middle there was a true purple portrait bust with a like -
ness of the Emperor Justin. He also wore a white tunic with a purple border, with gold
imperial embroideries, equally including the likeness of the emperor. The red shoes
that he wore, which he brought with him from his own country, bore pearls in Persian
fashion. Likewise his belt (p.341) was decorated with pearls. He received many other
gifts from the Emperor Justin, as did his wife Valeriana, since she had once and for all
been forced, or rather urged, to be wedded to him (and to go) to other
kingdoms. When Kavadh, the king of the Persians, learned of this, he sent (p.414) the follow-
ing message to the Emperor Justin by an ambassador: ‘Though friendship and peace
between us are being discussed and are now established,
53your actions are hostile.
Witness the fact that you have yourself appointed as king of the Lazi a subordinate of
mine, who does not come under the Roman administration but has from time imme-
morial belonged to the Persian state.’
54In reply the Roman Emperor Justin made the
following statement through an ambassador: ‘We have neither annexed nor won over
any of those subordinate to your empire. In fact, a man named Ztathius came to us in
our empire and begged us as a suppliant to be rescued from some defilement and Hel-
lenic belief, from impious sacrifices and from the errors of wicked demons, and asked
to become a Christian, worthy of the power of the eternal God, the Creator of all
things. It was impossible to prevent someone who wished to enter a better way and to
know the true God, so that when he had become a Christian and worthy of the heav -
enly mysteries, we sent him back to his own country.’ As a result there was enmity
between Romans and Persians. (tr. Jeffreys and Scott, revised)
Justin’s reply, together with the attention lavished on Tzath, clearly shows the important role
played by Christianity in dealings with the Transcaucasus. The emperor was being disingenuous,
however, in his reply, since it appears from another entry of Malalas that he had also been in nego -
tiations with Huns (presumably north of the Caucasus) and had secured their support. They
subsequently shifted allegiance to Kavadh, but Justin was able to turn this to his advantage by
claiming that they were planning to betray the Persians. Kavadh therefore killed their king,
Zilgibis, and destroyed their army (Mal. 17.10 [414.17–415.19], cf. Chr. Pasch.615.13–616.6).
See Vasiliev 1950: 264–5, Fowden 1993: ch.5, Greatrex 1998a: 133–4.

The proposed adoption of Khusro I by Justin I (c.525)
According to Malalas (17.10 [415.19–21], cf.Chr. Pasch.616.6–8), Justin’s information con
cerning Zilgibis inspired Kavadh to enter negotiations with Justin; he therefore despatched an
envoy called either Broeus or Labroeus to Constantinople. The priority of the Persian king,
now over seventy, was to secure the succession of his son Khusro, whose position was threatened
by rival brothers and the still strong Mazdakite sect. His proposed solution was that Justin
adopt Khusro as his son, a proposal greeted with enthusiasm by the emperor and his nephew
Justinian. The emperor’s quaestor Proculus, however, opposed the move, arguing that it would
entitle Khusro to inherit the Roman empire (Proc. I.11.1–22, Theoph. A.M. 6013 [167–8]).
Whether or not this would formally be the case under Roman law, it is highly unlikely that this
was Kavadh’s intention; and de factoit is difficult to see how Khusro could press his claim in any
55Negotations proceeded nevertheless, although it was made clear to the Persians that
Khusro could only be adopted ‘by arms’, a procedure already used by the Emperor Zeno for the
Ostrogothic king Theoderic. The negotiations are reported in most detail by Procopius and
must have taken place in 524 or 525. See Greatrex 1998a: 134–8.
Proc. I.11.23–30 (52.19–54.8): Accordingly the Emperor Justin dismissed the
(Persian) envoys, promising that men who were the noblest of the Romans would
follow them not long afterwards, who would arrange a settlement regarding the peace
and regarding Khusro in the best possible way. (24) He also answered Kavadh by letter
to the same effect. Accordingly there were sent (p.53) from the Romans Hypatius, the
nephew of Anastasius – who had reigned (until) recently – (who) also held the office
of magister militum per Orientem, and Rufinus, son of Silvanus, a man of note among
the patricians and known to Kavadh through their fathers; (25) from the Persians
came a most powerful man, wielding great authority, Seoses by name, whose rank was
adrastadaran salan¢s, and Mebodes (Mahbodh), who held the office of magister.
These men came together at a certain spot which is on the borderline between the land
of the Romans and the Persians; (there) they met and negotiated as to how they
should resolve their differences and make a good settlement of the issues surrounding
the peace. (27) Khusro also came to the river Tigris, which is about two days’ journey
from the city of Nisibis, in order that, when the details of the peace should seem to
both sides to be as well arranged as possible, he might set off himself in person to
Byzantium. (28) Now many words were spoken on both sides regarding the differ -
ences between them, and in particular Seoses (referred to) the land of Colchis, which
is now called Lazica, saying that it had been subject to the Persians from of old and
that the Romans had taken possession of it by violence for no reason. (29) When the
Romans heard this, (p.54) they took it ill that even Lazica should be disputed by the
Persians. And when they in turn stated that the adoption of Khusro must take place in
the manner that befits a barbarian, this seemed intolerable to the Persians. (30) The
two (sides) therefore separated and returned home, and Khusro left and went off to his
father with nothing accomplished, deeply injured at what had taken place and praying
that he might exact vengeance on the Romans for their insult to him. (tr. Dewing,

Despite the breakdown of negotiations about the adoption of Khusro in 524/5, it was not until
530 that full-scale warfare on the main eastern frontier broke out. In the intervening years, the
two sides preferred to wage war by proxy, through Arab allies in the south and Huns in the
north. Although the Emperor Justin died in August 527, the transfer of power to his co-ruler
Justinian was peaceful and straightforward.
The defection of Iberia to the Romans (524/5)
Tensions between the two powers were further heightened by the defection of the Iberian king
Gourgenes to the Romans. The successful transfer of allegiance of Tzath in 521/2 had clearly
had an impact in the region: Iberia had been securely in Persian hands since the late fourth cen-
tury. According to Procopius, the only source for this development, the Iberians were discon-
tented with efforts being made by Kavadh to impose Zoroastrian practices on them. Gourgenes
received pledges from Justin that the Romans would defend Iberia, and the emperor despatched
Probus to recruit Huns from north of the Caucasus to assist the Iberians. Although he proved
popular in the region between the Crimea and the Caucasus, he failed to raise any troops
(I.12.1–9, Zach. HEXII.7 [216]). See Angeli Bertinelli 1989: 123–4.
In 526/7 Justin therefore sent Peter, his former secretary, to Lazica, with some Huns, while
Kavadh countered by entrusting a large army to a certain Boes. The Persians easily overran
Iberia, forcing Gourgenes and the nobility to flee into Lazica. The Iberians then moved to
Constantinople and Peter was soon summoned there himself. In 528 he was sent back to Lazica
with other commanders, charged this time with defending Rome’s other new allies in the
region, the Lazi; evidently it was regarded as impossible to recover Iberia (Proc. I.10.9–14).
There was, according to the chroniclers (Chr. Pasch. 618, Mal. 18.4 [427], Theoph. A.M. 6020
[174]), heavy fighting, and the Romans lost control of the two forts guarding the approaches to
Lazica, Sarapanis and Scanda (Proc. I.12.15–19, although he does not name the forts there).
They were later to become important bargaining chips towards the end of the war. See
Lordkipanidse and Brakmann 1994: 85, Greatrex 1998a: 142–7. There were no further developments in the region during the war: the Romans, with Lazic
support, retained control of the rest of Lazica, while the Persians had regained Iberia. Negotia -
tions with the Huns north of the Caucasus continued, and, according to the chroniclers, a
Roman-allied queen Boa triumphed over a Persian-allied opponent in 528 (Mal. 18.13
[430–1], Joh. Nik. 90.61–5, Theoph. A.M. 6020 [175]). See Greatrex 1998a: 143.

The strengthening of Roman Armenia (528)
Just to the north of Armenia, the Roman commander Sittas spent some years in the early 520s
pacifying the mountainous Tzani. Thus the Romans gained control of an important region
bordering on both Armenia and Lazica (Proc. I.15.24,Aed. III.6).
1A brief and inconclusive raid
on Persarmenia in 526 by the young commanders Belisarius and Sittas may be noted in passing;
it may have been intended to relieve pressure on Iberia (Proc. I.10.20–2). See Greatrex 1998a:
The new Emperor Justinian, perhaps advised by these two former bodyguards of his, decided
on a large-scale reorganisation of Roman Armenia in 528. Hitherto, as has been seen, both
Armenia Interior and the former satrapies had had to rely primarily on their own defences: only
gradually had the Romans begun to exert their authority in this region. At the highest level, he
created a new post – that of magister militum per Armeniam – the holder of which supplanted
the magister militum per Orientem in Armenia and the Transcaucasus. Beneath him were duces
and Roman forces. See Toumanoff 1963: 133–4, Adontz 1970: 106–11, Garsoïan 1998a: 249,
Proc. Aed.III.1.27–9 (86.8–16): Even so, Roman soldiers did not serve these offi -
cials, 2but rather certain Armenians that had been accustomed to (do so) previously,
and consequently they were unable to ward off invading enemies. (28) Having exam-
ined these matters, the Emperor Justinian immediately removed the name of ‘satraps’
from the region and stationed two duces, as they are called, in these provinces. (29) He
joined to them many numeriof Roman soldiers so that they might guard together
with them the Roman borders.
3( … ) (tr. Dewing, revised)
C.J. I.29.5: The Emperor Justinian to A. Zeta (Sittas), vir illustrisandmagister
militum for Armenia, Pontus Polemoniacus and the gentes. Having, through God’s
grace, received the Roman power, and having considered (this matter) with solicitous
care and vigilant concern, we have found it appropriate to create by this law a special
military commander for parts of Armenia, Pontus Polemoniacus and the gentes.We
chose your highness, which has been greatly commended to us by its former activity,
confident that you would be suitable for such an honour. We have entrusted to your
care certain provinces, namely Greater Armenia, which is called Interior, and the
gentes (namely Anzitene, Ingilene, Asthianene, Sophene, Sophanene, in which lies
Martyropolis, Belabitene),
4as well as First and Second Armenias and Pontus
Polemoniacus, together with their duces. Thecomes Armeniae is to be abolished alto -
gether. We have placed under your command certain numeri, not only those which
have now been formed, but also (some) detached from the praesental, eastern and
other armies. Furthermore, the number of soldiers in them shall not be diminished;
rather, because we have added many (numeri) to them without burdening the state or
raising taxes, (although) we have now withdrawn some, the result is nevertheless that
they remain more numerous than they had been before our blessed time.
Adontz–Garsoïan, revised)
Mal. 18.10 (358.7–359.19/429.16–430.8): In the year of Justinian’s reign men-
tioned above, a man named Sittas was dispatched as magister militum per Armeniam,

for in previous times Armenia did not have amagister militumbut satraps,duces, gov-
ernors and comites. The emperor gave (p.359) the magister militum numeriof soldiers
from the two praesental (armies) and (the army) of the East. Sittas enrolled indigenous
(p.430) scriniarii and made them his own military scriniariiin accord with an imperial
rescript, having requested the emperor to enroll natives since they knew the regions of
Armenia. The emperor granted him this, and the rights of the Armenian ducesand
comites, and also their consulars who were formerly milites castrensiani; for the former
offices had been abolished.
6He also took four numerifrom the magister militum per
Orientem and from that time (Armenia) became a great bulwark and help for the
7(tr. Jeffreys and Scott, revised)
There were now therefore ducesbased at Tzanzakon, Horonon, Artaleson, Citharizon, Martyropolis
and Melitene, while the headquarters for the magister militumlay at Theodosiopolis. Fortifications
were also strengthened: Mal. 18.4 (427) notes building work at Martyropolis in 528, and work was
probably also carried out at Citharizon, Theodosiopolis and elsewhere (Proc. Aed.III). See Adontz
1970: 112–25, Whitby 1984, Whitby 1986a: 727, Howard-Johnston 1989: 218–20, Greatrex
1998a: 154.
Skirmishes along the Mesopotamian border (527)
Here Kavadh, through his Lakhmid vassal Mundhir, was able to continue exerting pressure on
the Romans to obtain funds from them: clearly none had been forthcoming since the end of the
Anastasian war. But the Roman treasury was in any case being severely drained by the need to
fund repairs to several cities of the East (see Chapter 5 above), compounded by a further series
of earthquakes which hit the region in 528 and 529. See Stein 1949: 420, Moorhead 1994: 96.
Zach. HEVIII.5 (77.9–24): Kavadh, the Persian king, insistently made demands for
the payment of the tribute of five gold centenariawhich was given to him by the king
of the Romans on account of the cost of the Persian army guarding the gates facing the
Huns; and for this reason he used from time to send his Tayyaye into the territory of
the Romans, and they plundered and took off captives.
9The Romans also crossed into
Arzanene, his (Kavadh’s) country, and the district of Nisibis, and inflicted damage.
Because of this, negotiations (took place), and the two kings sent (envoys), Justin
sending Hypatius and the old man Pharesmanes, and Kavadh Asthebid;
10and much
discussion took place on the frontier, which was reported to the two kings by their
nobles through couriers; and no peaceful message was sent by them, but (rather) they
were enemies of one another.
11( … ) (tr. Hamilton and Brooks, rev. M. Greatrex)
The fact that negotiations took place even after the defection of Iberia is an indication that Kavadh
genuinely wished to reach a compromise with the Romans; but their continuing refusal to offer him
payments will have rendered such a compromise impossible. Even before the raids of Mundhir
described by Zachariah (below, p.86), theRomans launched an attack of their own in mid-summer
527: the duxof Mesopotamia, Libelarius, led a force into Beth Arabaye, but failed to take either
Nisibis or Thebetha. Roman losses were considerable, and Justinian, having now become sole
emperor, replaced Libelarius with Belisarius (Proc. I.12.24). See Greatrex 1998a: 148–9.

Zach.HEIX.1 (92.3–17): ( … ) But since the Persians and the Romans were at
enmity with one another in those times, while Timostratus, the magister militum, was
dux on the frontier,
13the army with its officers assembled around him to fight against
Nisibis; and they fought, but could not crush it, and retired from there to the fortress
of Thebetha; and the (Roman) army came close up to the wall and breached it; and it
was the hot time of the summer. And through some cause they were hindered and did
not take the fortress, which was about fifteen parasangs from Dara. And the army was
ordered to go to Dara; and, because they greedily coveted as food honey and the flesh
of many swine, some of the infantry died of thirst on the march and were lost to the
army; others threw themselves into the wells of the desert and were drowned, and the
rest were burnt up on the march, but the cavalry reached Dara; and so the army was
broken up. (tr. Hamilton and Brooks, rev. M. Greatrex)
This raid may have been intended as a diversionary measure, for the Romans were seeking to
strengthen their fortifications elsewhere along the frontier, as is clear from Zachariah’s accounts
(below). While these efforts were unsuccessful too, other work proceeded unhindered, as to the
north (already noted) and at Dara. See Whitby 1986b: 758, Greatrex 199
8a: 149–50.
Zach. HEIX.2 (92.19–25): During the lifetime of the Emperor Justin, 14who had
learnt that Thannuris was an advantageous place for a city to be built as a place of
refuge in the desert, and for an army to be stationed as a protection for the ‘Arab
against the forays of the Tayyaye, Thomas the silentiariusof Apadna was sent to build
it. And when he had got ready a certain amount of material, the works which had
started to be carried out were halted by the Tayyaye and Qadishaye from Singara and
Thebetha. ( … ) (tr. Hamilton and Brooks, rev. M. Greatrex)
Zach. HEIX.5 (96.2–7): In the days of Belisarius the dux, in the (year) five,
Romans, having been prevented from building Thannuris on the frontier, wished to
make a city at Melabasa.
16And therefore Gadar the Kadisene was sent with an army by
Kavadh; and he hampered the Romans and put them to flight in a battle which he
fought with them on the hill of Melabasa. ( … ) (tr. Hamilton and Brooks, rev. M.
The strengthening of Palmyra (late 527)
One of Justinian’s first acts as sole emperor was to take steps to counter Mundhir’s raids. See
Greatrex 1998a: 151, Gregory 1997: II, 194–5.
Mal. 18.2 (354.13–355.19/425.10–426.5): In the month of October of the sixth
indiction the emperor appointed an Armenian named Patricius as comes Orientisin
Antioch. To him he gave a large sum of money with instructions (p.426) to go and
reconstruct the city in Phoenice on the limes, known as Palmyra, and its churches and
public buildings. He also ordered a numerusof soldiers to be stationed there with the
limitanei, and also the duxof Emesa, to protect (p.355) the Roman territories and
Jerusalem. (tr. Jeffreys and Scott, revised)

Cross-border warfare in the south (528–529)
In late 527 or early 528 Harith, the phylarch of Palaestina I, quarrelled with the localduxand
fled eastwards into the desert. There he was defeated and killed by Mundhir. Justinian reacted
swiftly, writing personally to the ducesof Phoenice, Arabia and Mesopotamia and to various
phylarchs, instructing them to avenge Harith’s death. Their force failed to encounter Mundhir,
but sacked several Persian forts, and returned with much booty in April 528 (Mal. 18.16
[434–5]). See Shahîd 1995: 70–6, Greatrex 1998a: 152.
17Mundhir’s riposte followed in spring
529. Clearly the duxrecently moved to Palmyra was unequal to the task of deterring Mundhir.
See Shahîd 1995: 43–4, 79–80, Greatrex 1998a: 152–3.
Zach. HEVIII.5 (77.24–78.4): ( … ) And Mundhir, the Tayyaye king, went up into
the territory of Emesa and Apamea and the district of Antioch twice; 19and he carried
off many people and took them down (p.78)with him. And four hundred virgins,
who were suddenly captured from the congregation (of the church) of Thomas the
Apostle at Emesa (?), he sacrificed in one day in honour of ‘Uzzai.
20Dada the ascetic,
an old man, who was captured from the congregation, saw it with his eyes, and told
me. (tr. Hamilton and Brooks, rev. M. Greatrex)
Theoph. A.M. 6021 (178.7–15):
21On March 21st of the 7th indiction, Mundhir son
of Zekik¢, prince of the Saracens, invaded and looted Syria I as far as the boundaries of
Antioch, up to a place called Litargon, and the estates of Skaphathae. He killed many
people and burned the territory outside Chalcedon and the Sermian estate and the
Kynegian country.
22And, hearing (this), the Roman commanders went out against
him. When they realised this, the Saracens, with the Persians, took their booty and
prisoners and fled through the inner limes.
23(tr. Mango and Scott, revised)
In response, Justinian deployed contingents of Lycaonian infantry in the East (Mal. 18.34
[445], Theoph. A.M. 6021, 178.15–18). See Greatrex 1998a: 152–3.
The attempt to fortify Thannuris (528)
The most important Roman initiative on the southern front in 528 was Belisarius’ expedition
to Thannuris. Here he tried, albeit unsuccessfully, to protect Roman workers undertaking the
construction of a fort right on the frontier. (Zach. HEIX.2, p.85 above) The region to the south
of Dara was poorly defended, and had always been subject to Arab or Persian raids (cf.
Ammianus XIV.3). Now measures were being taken to plug this gap in defences. Although this
effort failed, it is clear from Proc. (Aed. II.6.14–16) that at a later stage Justinian did succeed in
establishing a fort there. See Kennedy and Riley 1990: 118–21 (on the site), Shahîd 1995: 76–9
and Greatrex 1998a: 150 n.33, 156–9.
Zach. HEIX.2 (92.25–93.27): Now because, as we have made known above, the
Romans had equipped themselves (p.93)and fought against Nisibis and Thebetha,
afterwards the Persians also similarly came and made an entrenchment
24in the desert
of Thannuris. And when the duxTimostratus, the magister militum25had died,
Belisarius had succeeded him; and he did not hanker after bribes, and was kind to the

peasants, and did not allow the army to injure them. For he was accompanied by Solo-
mon, a eunuch from the fortress of Edribath; 26he was a shrewd man, and accustomed
to the affairs of the world. He had been notary to the duxFelicissimus, and had been
attached to the other governors and had gained cunning through experience of
difficulties. Accordingly, a Roman army was gathered in order to enter the desert of Thannuris
against the Persians; with (it were) Belisarius, Coutzes, the brother of Bouzes, Basil,
Vincent and other commanders, and Atafar, the chief of the Tayyaye.
27When the Per -
sians heard, they craftily dug several ditches among their trenches, and concealed
28all round outside by triangular stakes of wood, and left several openings. When
the Roman army arrived, they did not foresee the Persians’ deceitful strategem, but
the generals entered the Persian entrenchment at full speed; and when they fell into
the pits, they were seized, and Coutzes was slain. Of the Roman army those who were
mounted turned back and, fleeing, returned to Dara with Belisarius; but the infantry,
who did not escape, were killed or taken captive. Atafar, the Tayyaye king, during his
flight was struck from nearby and died;
29he was a warlike and skilful man, and he was
very practised in Roman arms, and in various places had excelled and was celebrated.
(tr. Hamilton and Brooks, rev. M. Greatrex)
Mal. 18.26 (441–2) describes an invasion by 30,000 Persians, who were opposed by Coutzes,
Sebastian, Proclianus, duxof Phoenicia, the comesBasil, Belisarius and the phylarch Tapharas.
Proc. I.13.1–8 refers to the fort as Minduos, apparently situating it near Nisibis, and mentions,
apart from Belisarius, the two ducesof Phoenice, Coutzes and Bouzes. See Greatrex 1998a:
156–9 for an attempt to reconcile these three sources.
While continuing work at fortresses already constructed, such as Dara and Martyropolis
(above), Justinian shored up defences in the East still further by transferring soldiers from the
Balkans and entrusting the defence of Beroea, Sura, Edessa, Amida and Constantia to specific
nobles despatched from Constantinople for the purpose (in late 528) (Mal. 18.26 [442]). See
Greatrex 1998a: 159.
Attempts at a compromise (529)
So far no significant campaigning had taken place anywhere along the eastern frontier. Armies
had clashed close to the border, Roman fort-building had been thwarted on several occasions,
and Arab raids had been carried out, but nothing more. The considerable Roman building
programme in the late 520s should be seen as a defensive effort, designed to allow the Romans
to refuse Persian demands for money with impunity – not as a measure with which to threaten
Persian territories.
30After the hard winter of 528–529, during which Antioch and Laodicea
were struck by earthquakes, followed by Mundhir’s daring raid in March 529, Justinian was
eager to negotiate. Already by the time his envoy Hermogenes had reached the Persian court, a
large uprising of the Samaritans had broken out, which was suppressed only with difficulty
(Mal. 18.35 [445–7], Theoph. A.M. 6021, 178–9).
31The response of Kavadh to Hermogenes
is recorded by Malalas. See Scott 1992: 160, Greatrex 1998a: 160.

Mal. 18.44 (378.32–47/449.19–450.15):Koades (Kavadh), king of kings, of the
rising sun, to Flavius Justinian Caesar, of the setting moon. 32We have found it written
in our ancient records that we are brothers of one another and (p.450) that if one of us
should stand in need of men or money, the other should provide them. From that
time to the present we have remained constant in this. Whenever nations have risen
against us, against some we have been compelled to fight, while others we have per -
suaded by gifts of money to submit (to us), so it is clear that everything in our treasury
has been spent. We informed the Emperors Anastasius and Justin of this, but we
achieved nothing. Hence we have been compelled to mobilize for war, and having
become neighbours of Roman territory, (we have been compelled) to destroy the peo -
ples in between on the pretext of their disobedience, even though they had done noth -
ing wrong. But, as pious Christians, spare lives and bodies and give us part of your
gold. If you do not do this, prepare yourselves for war. For this you have a whole year’s
notice, so that we should not be thought to have stolen our victory or to have won the
war by trickery. (tr. Jeffreys and Scott, revised)
Justinian received this reply in late 529. In response he did not send money, but despatched
Hermogenes again, this time with Rufinus. In March 530 they were in Antioch, from where
they proceeded to Dara. They appear to have expected negotiations to continue, but Kavadh
had now mustered an army and was prepared to back up his demands with force. The Romans
too had an army ready at Dara, commanded by Belisarius, recently promoted to the post of
magister militum per Orientem (Mal. 18.50 [452], Theoph. A.M. 6022, 180, Proc. I.13.9–11).
See Scott 1992: 161, Greatrex 1998a: 162–3.
The promotion of Harith the Jafnid (529)
The opening of Justinian’s reign saw significant changes of command on the eastern frontier:
men of Justin’s generation, such as Pharesmanes, Hypatius and Timostratus disappear from the
scene. Sittas and Belisarius were entrusted with high commands, as was the Jafnid ruler Harith.
According to Procopius, who in general is critical of the chief, Harith received ‘the dignity of
king’ from the emperor. He thus became a supreme phylarch, wielding authority not just in
Arabia, his own phylarchate, but over the phylarchs of other provinces too. Justinian thereby
hoped to make him a match for the Lakhmid Mundhir, who exercised undisputed control over
the tribes allied to Persia, and whose power was extended by Khusro in 531 to cover the whole
Arabian coast of the Persian Gulf (Proc. I.17.46–8, Tabari 958/253). See Sartre 1982a: 170–2,
Potts 1990: II 249, 334–5, Shahîd 1995: 95–124, Greatrex 1998a:
The battle of Dara (June 530)
Kavadh’s decision to cross the border in force in June took the Romans by surprise:
Hermogenes, who was with Belisarius at Dara, had expected negotiations to continue, and
Rufinus was waiting at Hierapolis to take part in a diplomatic mission (Proc. I.13.10–11). But
when Kavadh’s army advanced as far as Ammodius, 7.7 km from Dara and in Roman territory,
Belisarius prepared his army for battle outside Dara; the fact that he chose to fight in the field,
rather than defending the city from within, probably indicates that work on the city’s defences

was continuing. To compensate for the uncertain quality of his forces, Belisarius dug protective
trenches in front of his infantry (Proc. I.13.12–18). See Whitby 1986b: 758–9, Greatrex
1998a: 168–72.
Belisarius positioned his forces with care, assigning a key role to the Hunnic forces. 600 Huns
were stationed on each side of the centre, just in front of the trenches and the infantry. From
there they could help either the infantry in the centre or the cavalry on the flanks. The Persians
divided their forces into three parts, the central section of which was under their commander-
in-chief Mihran. When both sides were ready for battle, some skirmishing and single combat
took place; in the latter, a Roman former bath attendant emerged victorious in two duels,
despite having been forbidden to fight again after his first victory (Proc. I.13.19–39). On the
next day, reinforcements arrived for the Persians from Nisibis, bringing their force up to
50,000; the Romans faced them with 25,000 men. Nevertheless, no fighting took place.
Belisarius and Hermogenes sought to reopen negotiations, but the Persian general Mihran
rebuffed the attempt (Proc. I.14.1–12). The two sides therefore prepared for battle on the
following day. Mihran kept some of his forces in reserve, so as to relieve those who were hard
pressed in the fighting, thus exploiting his superiority in numbers. On the Roman side,
Belisarius accepted a suggestion from Pharas, a Herul, that he should position his men behind a
hill on the edge of the battlefield, from where he might fall on the Persians unexpectedly (Proc.
I.14.13–33). See Greatrex 1998a: 173–85 (with diagrams). Procopius’ eyewitness account of
the battle itself provides an important description of the tactics and h
abits of the two sides.
Proc.WarsI.14.34–55 (70.23–74.4): But up to midday neither side began battle. As
soon, however, as the noon hour had passed, the barbarians began operations, having
postponed the engagement to this time of day for the reason that they (p.71) are
accustomed to partake of food only towards late afternoon, while the Romans have
their meal before noon; and for this reason they thought that the Romans would never
hold out so well, if they assailed them while hungry.
33(35) At first, then, both sides
employed arrows against each other, and the missiles by their great number made, as it
were, a vast cloud; and many men were falling on both sides, but the missiles of the
barbarians fell much more thickly. (36) For fresh men were always fighting in turn,
giving the enemy the least possible opportunity to observe what was being done; but
even so the Romans did not have the worst of it. For a steady wind blew from their
side against the barbarians, and did not allow their arrows to be very effective. (37)
(Then), when all the missiles of both sides had been exhausted, they began to use their
spears against each other, and the battle had come still more to close quarters. On the
Roman side, the left wing was suffering especially. (38) For the Kadiseni, who with
Pityaxes were fighting at this point, came to (his) assistance suddenly in great num -
bers, routed their enemy, and, pressing hard upon the fugitives, were killing many of
them. (39) Seeing this, the men under Sunicas and Aigan advanced against them at
great speed. But first the 300 Heruls under Pharas from the high ground got in the
rear of the enemy and made a wonderful display of valorous deeds against all of them,
and especially the Kadiseni. (40) And the Persians, seeing the forces of Sunicas too
already coming up against them from the flank, (p.72) turned to flight. (41) And the
rout became complete, for the Romans here joined forces with each other, and there

was a great slaughter of the barbarians. (42) On the Persian right wing not fewer than
3000 perished in this action, while the rest escaped with difficulty to the phalanx and
were saved. (43) And the Romans did not continue their pursuit, but both sides took
their stand facing each other in line. Such then was the course of these events. (44) But
Mihran secretly sent to the left wing many men and the so-called Immortals. And
when Belisarius and Hermogenes saw them, they ordered the 600 men under Sunicas
and Aigan to go to the angle on the right, where the troops of Simmas and Ascan were
stationed, and behind them they placed many of Belisarius’ men. (45) So the Persians
who held the left wing under the leadership of Baresmanas, together with the Immor-
tals, charged on the run against the Romans opposite them, who failed to withstand
the attack and turned to flight. (46) Thereupon the Romans in the angle, and as many
as were behind them, advanced with great ardour against the pursuers. (47) But inas -
much as they came upon the barbarians from the side, they cut their army into two
parts, and the greater portion of them they had on their right, while they kept some
also who were left behind on their left. Among these happened to be the standard-
bearer of Baresmanas, whom Sunicas charged and struck with a spear. (p.73) (48) And
already the Persians who were leading the pursuit perceived in what straits they were,
and, wheeling about, they stopped the pursuit and advanced against them (the
Romans), and thus became exposed to the enemy on both sides. (49) For those
(Romans) in flight before them understood what was happening and turned back
again. The Persians, on their part, with the detachment of the Immortals, seeing the
standard inclined and lowered to the earth, rushed all together against the Romans at
that point with Baresmanas. There the Romans held their ground. (50) And first
Sunicas killed Baresmanas and threw him from his horse to the ground. As a result of
this the barbarians were seized with great fear and thought no longer of bravery, but
fled in utter confusion. (51) And the Romans, having made a circle as it were around
them, killed about 5000. Thus both armies were all set in motion, the Persians in
retreat, and the Romans in pursuit. (52) In these straits all the infantry who were in
the Persian army threw down their shields and were caught in disorder and killed by
(their) enemy. However, the pursuit of the Romans took place only for a short dis -
tance. (53) For Belisarius and Hermogenes refused absolutely to let them go further,
fearing lest the Persians through some necessity should turn about and rout them
while pursuing recklessly, and it seemed to them sufficient to preserve the victory
unmarred. (54) For on that day the Persians had been defeated in battle by the
Romans after a long time (without any such success).
34Thus the two armies separated
from one another. (55) And the Persians (p.74) were no longer willing to fight a
pitched battle with the Romans straightaway. However, some sudden attacks were
made on both sides, in which the Romans were not at a disadvantage. Such, then, was
the fortune of the armies in Mesopotamia. (tr. Dewing, revised)
Zach. HEIX.3 (94.1–95.2): (The third chapter of the ninth (book), concerning the
battle by Dara.) The Persians were proud and puffed up and boastful indeed; the
mihran and the marzbans assembled an army and came against Dara and encamped at
Ammodius, being justifiably confident in thinking they could subdue the city because
the Roman army had been diminished by their sword. Their cavalry and infantry

approached and came up on the south side of the city, (in order) to encompass it all
around for the purpose of blockading it; but a Roman army encountered them by the
help of our Lord, who chastises but does not utterly deliver over to death. For a certain
Sunicas, a general who was a Hun (and) had been baptized when he took refuge with
the Romans, and Simmas, a Roman tribune, and their armour-bearers with twenty
men drove the whole Persian army away from the city several times, passing boldly
and vigorously from one side to the other, and slaying (men) right and left. They were
practised with the spear and the sword; their cry was vehement and powerful, and they
seemed terrible to the Persians, so that they fell before them; and two of their leaders
were killed, along with a not inconsiderable force of cavalry, while the Heruls who
were with Bouzes to the east of the city slew and hurled back thefaige, who are the Per-
sian infantry. The Persians, when they realised how many had been killed, craftily sent (a mes -
sage) to Nisibis, asking them to bring as many (pack animals) as they could, and to
come at once to Dara and take as much spoil as they were able. And when many came,
they loaded up the bodies of the slain, and they returned in shame. (p.95) However
the rest of the Persian force poured over the Roman ‘Arab and burned it with fire.
Hamilton and Brooks, rev. M. Greatrex)
The battle of Satala (summer 530)
Kavadh did not attack Roman territory only in Mesopotamia. To the north, the Persians had
gained ground recently, having retaken Iberia and advanced into Lazica. An advance into Roman
Armenia therefore seemed appropriate. Mihr-Mihroe (Mermeroes) was placed in command of a
mixed force of Persians and allies, including Persarmenians, Sabirs and Sunitae. Ranged against
them were the ducesand troops of the magister militum praesentalis Sittas and his successor as
magister militum per Armeniam Dorotheus. Even before the Persians invaded, they were able to
inflict some losses on the enemy, having heard of the preparations underway. Mihr-Mihroe, when
his 30,000 troops were assembled, made no attempt on Theodosiopolis, preferring to penetrate as
far as possible into Roman territory. The Romans took a stand at the important base of Satala,
where they succeeded in defeating their more numerous opponents by a ruse (Proc. I.15.1–17).
The Roman successes of 530 then began to have an impact on the region: first, two Persarmenian
chiefs, the brothers Narses and Aratius, defected to the Roman side. They received a generous
welcome from the Romans at the hands of the eunuch Narses, and this in turn persuaded their
youngest brother, Isaac, to follow suit, handing over control of the fortress of Bolum at the same
time. The Romans also gained the fort of Pharangium, which was a particular blow to the
Persians, since it lay in a gold-producing region (Proc. I.15.18–19, 26–33).
The resumption of negotiations (autumn 530)
In August 530, probably after the defeat at Satala, Rufinus and the comesHermogenes arrived at
Kavadh’s court. The king complained bitterly of the continuing Roman refusal to pay for the
defence of the Caspian Gates, as well as of the construction of Dara. He was now obliged, he
claimed, to maintain two armies – one to defend the Caspian Gates, and one to defend against

the Romans in Dara. He therefore demanded that the Romans either dismantle Dara or con-
tribute towards the costs of defending the Caspian Gates; he hinted to the envoys that the latter
was the option he preferred (Proc. I.16.1–10). When the ambassadors returned to Constanti -
nople in winter 530, Justinian believed the peace was within his grasp: he may have been
assured that a one-off payment to the Persians would satisfy Kavadh, without the obligation to
contribute money every year. He wrote a letter to the king, welcoming the development; but
when Rufinus reached the frontier in early 531, it soon became clear that Kavadh was no longer
interested in peace. The defections and losses in Armenia may have changed his mind, as may
exaggerated rumours of the Samaritan uprising: some Samaritans had contacted the king,
urging him to march on Jerusalem, but were intercepted on their way back to Roman territory
(Mal. 18.54 [455–6], Theoph. A.M. 6021 [179]). See Scott 1992: 161, Greatrex 1998a:
The campaign of Callinicum (spring 531)
According to Procopius, the inspiration for Kavadh’s next line of attack was the Lakhmid chief
Mundhir. A Persian army of 15,000, under Azarethes, consisting exclusively of cavalry,
together with 5000 men under Mundhir’s command, crossed into Roman territory by
Circesium. They advanced up the Euphrates, meeting no resistance. Only when they passed
Callinicum was Belisarius informed; he instantly moved west with 3000 men and 5000
Ghassanid allies under Harith. The remainder of his forces he left behind to defend Dara. The
Persians had by this time reached the region where the Euphrates runs north–south and
proceeded to ravage the provinces of Euphratesia and Syria I, but the arrival of Belisarius at
Chalcis barred any further advance. The duxSunicas harried the Persians near Gabbulon
successfully, although Belisarius had given orders not to engage the enemy. Meanwhile
Hermogenes arrived at Hierapolis with reinforcements, and from there proceeded south to
Barbalissus. Here the Roman army assembled, and Hermogenes tried to reconcile the magister
militum per Orientem with his subordinate commanders. This proved to be a difficult task, since
Belisarius was inclined to let the Persians now withdraw unmolested. Local commanders felt
differently, no doubt spurred on in part by the fact that prisoners seized by Mundhir in 529 had
only just been ransomed for a large price; others had been killed in captivity (Mal. 18.59
[460–1]). Furthermore, when the citizens of Antioch had heard of the recent Persian advance,
they had fled to the coast in panic. Hermogenes therefore led the army forward against the
invaders, who were now withdrawing eastwards. The Roman army shadowed the Persians as far
as a site opposite Callinicum, on the south side of the Euphrates. Here Belisarius was prevailed
upon to engage the Persians, just before they retreated into the uninhabited regions beyond
Callinicum (Mal. 18.60 [460–3]; Proc. I.17.1–3, 29–39, 18.1–13). See Greatrex 1998a:
The Roman army, having set off from Sura early on Saturday 19 April 531, came upon the
Persians as they were about to leave their camp opposite Callinicum. Although the Roman
troops had already marched some distance that day and were fasting during Holy Week, they
insisted on engaging the enemy immediately in spite of the objections of both Belisarius and
Hermogenes (Proc. I.17.14–29). The ensuing battle was a substantial victory for the Persians.
Both Procopius and Malalas offer a detailed account of events, but their emphasis is somewhat
different: the former exonerates Belisarius as far as possible, insisting that he prevented a rout by
drawing up his infantry in dense formation to ward off the Persian cavalry, and that the Persians

suffered considerable losses (I.17.30–56). 37Malalas, on the other hand, accuses Belisarius of
fleeing the battle early, leaving Sunicas and Simmas to wage a successful infantry struggle
against the Persians (Mal. 18.60 [463–5]). Both accounts agree on the poor performance of the
Lycaonian or Phrygian infantry and note suspicions of treachery on the part of the allied Arabs.
Procopius insists that although Roman firepower was less rapid than the Persians’, it was more
effective, but the evidence of Zachariah (below) implies otherwise, at least in this instance.
the battle itself see B. Rubin 1960: 287–9, Shahîd 1995: 134–42
, Greatrex 1998a: 200–7.
Zach. HEIX.4 (95.4–26): The Persians, since they had become wise from experience
on account of the great (damage) which they had suffered from the attacks of the
Romans whenever they approached the city and went out against them, went up into
the desert land of the Romans and encamped on the Euphrates; and according to their
usual practice they made a trench. Belisarius at the head of a Roman army and tri -
bunes came up against them to battle; and they (the Romans) arrived in the last week
of the fast. The Persians were found to be like a little flock, and so they appeared in
their (the Romans’) eyes; and the astebid, their commander, was afraid of them, and
those who were with him; and he sent (a message) to the Romans, (asking them) to
honour the fast, ‘for the sake of the Nazarenes and Jews who are in the army that is
with me, and for the sake of yourselves who are Christians.’
39And when Belisarius the
magister militum had considered this, he was willing (to agree); but the commanders
complained greatly, and would not consent to delay and honour the day. When they
prepared for battle on the eve of the first day of the week, (the day) of unleavened
bread, it was a cold day with the wind against the Romans, and they appeared feeble,
and turned and fled before the Persian attack. Many fell into the Euphrates and were
drowned, and others were killed; Belisarius escaped, while the nephew of Bouzes was
seized (for he himself was ill at Amida, and did not go to the battle, but sent his army
to Abgersaton with Domitziolus), and went down to Persia. But he eventually
returned; and how this happened I shall relate in the chapter below. (tr. Hamilton and
Brooks, rev. M. Greatrex)
The Roman failure was followed by a commission of enquiry, the result of which was the dis -
missal of Belisarius from his post in May or June. See Greatrex 1998a: 194–5, questioning the
impartiality of the commission.
The aftermath of Callinicum (April–June 531)
Hermogenes visited Kavadh immediately after the battle to reopen negotiations, but without
success (Proc. I.20.1). The Persians were meanwhile pressing home their advantage elsewhere,
seizing the fortress of Abgersaton in Osrhoene (Mal. 18.61 [465–6], Zach. HEIX.4, above).
Justinian therefore took steps to bolster the Roman position. Sittas was charged with the
defence of the entire East in the wake of Belisarius’ dismissal; and the former praetorian prefect
Demosthenes was sent to the frontier with funds to provide the cities with adequate granaries
(Proc. I.20.3, Mal. 18.63 [467]). At the same time the emperor continued to try to engage
Kavadh diplomatically. The role of Mundhir as intermediary is interesting, as is Justinian’s
attempt to enter into relations with the Lakhmid chief. See Scott 1992: 163–4, Shahîd 1995:

142–3 (suggesting that Sergius should be identified with a Sergius, bishop of Sergiopolis),
Greatrex 1998a: 208.
Mal. 18.61 (390.86–8/466.18–467.14):In the month of June, while the Roman
magistri militum were making preparations against the Persians, Alamundarus
(Mundhir), the prince of the Saracens, wrote to the Romans for a deacon called
Sergius to be sent to him so that he could convey peace terms through him to the
Roman emperor. Sergius was sent to the Roman emperor with the letter (p.467) sent
by Mundhir. The emperor, having read the letter, did not stop his campaign against
the Persians. He sent Rufinus as an ambassador to Persia with a letter for him (the
king, recommending) that he accept friendship: ‘for it is honourable and glorious to
make the two states to live in peace. If he does not do this, I shall seize the Persian land
for myself.’ At the same time Sergius the deacon was sent to king Mundhir with impe -
rial gifts. In that year gifts were sent from the emperor of the Romans to the Persian
king. Likewise the Augusta sent gifts to the Persian empress, who was his sister.
When Rufinus and Strategius reached the city of Edessa, they sent a message to
Kavadh, the king of the Persians. He put off receiving them, since he had secretly sent
a force against the Romans.
41(tr. Jeffreys and Scott, revised)
Campaigns in Armenia (summer 531)
The Persians were also active further north, as Zachariah relates.
Zach. HEIX.5 (96.7–27): ( … ) And he (Gadar) was in the confidence of Kavadh,
and had been stationed with an army to guard the frontier eastwards from Melabasa in
the country of Arzanene as far as Martyropolis. This man boasted and talked nonsense
against the Romans, and blasphemed like Rabshakeh, who was sent by Sennacherib.
He brought about 700 armed cavalry and infantry, who accompanied them to amass
plunder; and they crossed the Tigris into the district of Attachas (in the territory) of
Amida. Bessas was duxin Martyropolis, and it was summer in this year nine (531).
With Gadar was Yazdgerd, the nephew of the bdeashkh, who, as a neighbour, knew
the region of Attachas. When Bessas heard (of this), he went out against him with
about 500 horsemen from Martyropolis, which was about four stades distant.
encountered him at Beth Helte and massacred his army by the Tigris, killed Gadar,
took Yazdgerd prisoner, and brought him to Martyropolis. This man after the peace,
which was (made) in (the year) ten, was given in exchange for Domitziolus, who
returned from Persia. But Bessas the dux, after massacring Gadar and the Persian cav -
alry, who were guarding the frontier of Arzanene, entered the country and did much
damage there; and he carried off captives and brought (them) to Martyr
Cf. Mal. 18.65 (468–9) for an account of the same events, taken from a report compiled by
Hermogenes. He claims that 6000 Persians invaded, of whom 2000 were killed. Malalas also
(18.66 [469]) refers to successes of Dorotheus in Persarmenia, where he defeated Persians and
Persarmenians and captured fortresses. These events spurred Kavadh to make a further attempt

to redress the position: his victory at Callinicum had been narrow, little booty had been gained,
and no major fortresses captured. He therefore now despatched an army under Aspabedus,
Mihr-Mihroe and Khanaranges to take Martyropolis, a city vexingly close to the frontier and
recently strengthened by Justinian, and to avenge Bessas’ raid on Arzanene. See Greatrex 1998a:
Zach.HEIX.6 (97.2–98.18): The villages in the country of Arzanene belong to the
Persian kingdom, and no small sum is collected as poll-tax from their inhabitants for
the king’s treasury and for the office of the bdeashkhwho is stationed there (he is the
king’s prefect). To this country, as related above, Bessas the duxdid much injury; he
captured the nephew of the bdeashkh, and also kept him prisoner in Martyropolis.
King Kavadh was much distressed when he learned from the bdeashkhabout the dev -
astation of the country. This same Hormizd left no stone unturned in order to subdue
Martyropolis by force and cunning,
43for it acts as a (place of) ambush and refuge for a
Roman army (from which) to damage Arzanene. And an army was, so to speak,
equipped by the Persian army: Mihr Girowi was sent to hire a large number of Huns
and bring them to their assistance. They got ready and were gathered together by
Martyropolis at the beginning of the (year) ten;
44and they made a trench against it,
and a ‘mule’ and many mines; and they pressed hard upon it in an attack and
oppressed it. In it was a Roman force of no small size and Bouzes, and they drove large
numbers of Persians back in battle. But Nonnus, the bishop of the city,
had died. Now Belisarius, because he was blamed by the king on account of the massacre of
the Roman army by the Persians at Thannuris and by the Euphrates, had been
dismissed and had gone up to the king; and after him (the commander) at Dara was
45A large Roman army was assembled, and Sittas was strat¢gos; and Bar
Gabala, the Tayyaye king was with them. 46And they reached Amida in November 47of
the year ten; and John, the hermit of Anastasia, (p.98) a man of honourable character,
who had been called (to the bishopric) accompanied them. When they had gone to
Martyropolis and the winter came on – the country is northerly and cold – the
Persians were held back by rain and mud, and endured hardship, while they were also
afraid of the multitude of the Roman army; (in addition), Kavadh their king had died
while they were there, and (so) they made a truce with the Romans to withdraw from
the city. Soon after they had withdrawn and Martyropolis had been relieved, and the
Roman army had returned, the Huns, who had been hired by the Persians, arrived.
This great people suddenly entered the territory of the Romans, and massacred and
slew many of the tillers of the soil, and burned villages and their churches. They
crossed the Euphrates and advanced as far as Antioch, and no one stood before them
or harmed them, except only this Bessas, the duxof Martyropolis, who fell upon some
of them while they were leaving, and killed them, and captured about 500 horses and
much spoil; and the man became rich. At the fortress of Citharizon the duxthere
repulsed some of them – about 400 men – and captured their baggage animals. (tr.
Hamilton and Brooks, rev. M. Greatrex)

Proc. I.21.6–8 also mentions the siege; he plays up the weakness of the fortifications, probably
excessively. Cf.Aed.III.2 with Whitby 1984: 182. Mal. 18.66 [469–70] provides more details,
describing how the defenders foiled all the tactics pursued by the besiegers. As Zachariah
describes, the arrival of Sittas and another Roman army at Attachas, 20 km from Martyropolis,
forced the Persians to withdraw in November or December 531 (Proc. I.21.9–11). Procopius
(I.21.11–16) adds that the Romans encouraged the Persian decision to withdraw by suborning
a Persian spy, who reported that the approaching Sabir Huns were allied to the Romans rather
than them. A further factor prompting the Persian withdrawal was news of the death of Kavadh;
Malalas (18.68 [471.4–10]) actually attributes his death (in September) to news of Persian
losses. See Greatrex 1998a: 210–11.
In autumn 531 Rufinus and Strategius were still at Edessa, having been refused access to
Kavadh while the attack on Martyropolis proceeded. Justinian, learning of the attack on
Martyropolis, forbade them to proceed to the king with their gifts until he so ordered (Mal.
18.66 [470.16–18]). When Khusro, having succeeded to the throne, therefore invited the
envoys to proceed, they refused, citing the emperor’s command. Khusro was obliged to write to
Justinian through Hermogenes, asking to see the envoys and to arrange terms. The emperor
refused, answering Khusro: ‘We do not give permission for our ambassadors (p.472) to come to
you, nor do we recognize you as king of the Persians.’ (Mal. 18.68 [471.22–472.2], tr. Jeffreys
and Scott). Justinian thereby hoped to destabilise Khusro’s position, clearly aware of the insta -
bility within Persia. However, Khusro was able to purge his brother and his Mazdakite
supporters swiftly, and Justinian therefore agreed to a three-month truce and the exchange of
hostages, presumably at the very end of 531 or in early 532 (Mal. 18.69 [472.7–14]). See Crone
1991: 31–3, Scott 1992: 164, Greatrex 1998a: 211–12. Before peace was agreed in the following year, several eastern provinces had to endure one
further invasion – that of the Sabir Huns described by Zachariah (above). They had been
summoned by the Persians earlier in the year, but only arrived in December, after the Persians
had withdrawn from Martyropolis. According to Malalas (18.70 [472–3]) they penetrated as
far as Euphratesia, the region around Cyrrhus, and Cilicia. Little effort was made to repel them,
perhaps because their arrival was unexpected so late in the year. But as they withdrew counter-
operations got underway, and Dorotheus, the magister militum per Armeniam, retook much of
the booty they had plundered (Mal. 18.70 [472–3], Proc. I.21.28 [playing down the raid],
Zach. HEIX.6 [above], Chr. Ede.103 [
AG 843]).
The Eternal Peace (532)
In spring 532 negotiations began in earnest to end the war. The new Persian king, Khusro,
needed to devote his attention to securing his own position, while Justinian may already have
been looking westwards and have been intending to redeploy his forces now engaged in the
East. The Roman envoys Rufinus, Hermogenes, Alexander and Thomas met Khusro in Persian
territory and soon came to an agreement. The Persians would receive 110 centenaria(11,000 lbs
of gold), and the Romans would withdraw the base of the duxof Mesopotamia from Dara to
Constantia. Khusro insisted on retaining control of Sarapanis and Scanda, the forts in Lazica,
and would also receive back the Armenian forts of Pharangium and Bolum. Khusro underlined
that the payment was to be made to cover the expense of defending the Caucasian passes. Before
the agreement was finalised, Rufinus returned to Constantinople to acquire permission to con -
cede the forts to Khusro (Proc. I.22.1–8). Justinian at first agreed to the terms, but soon

afterwards ordered his ambassadors not to give up the Lazic forts to Khusro. The negotiations
were therefore broken off, and Rufinus recovered the money already paid only with difficulty
(Proc. I.22.9–14). In summer that year, however, Hermogenes and Rufinus succeeded in con-
cluding a treaty by which the Romans recovered the Lazic forts and the Iberians who had fled
their country were allowed to remain in Roman territory or to return to their native land (Proc.
I.22.15–19, Zach. IX.7 [99–100]).
Mal. 18.76 (401.19–33/477.13–478.7): In that year Hermogenes and Rufinus
returned from Persia, bringing with them a peace treaty between the two states of
Rome and Persia for the duration of the life of both.
49The region of Pharangium was
restored to the Persians with all their prisoners, while the forts that had been captured
by the Persians were restored to the Romans, with those captured in them. The two
rulers agreed and stated explicitly in the treaty that they were brothers according to the
ancient custom, and that if one of them required money or men in a military alliance,
they should provide it without dispute.
50After these proceedings (p.478) both armies,
the Roman and the Persian, withdrew. The war had lasted 31 years from the time that
Kavadh, the king of the Persians, had advanced in hostility into Roman territory, as
was mentioned above during the reign of Anastasius, and the capture of Amida, men -
tioned above, and the restoring of that city of Amida to the Romans, and the local
wars with raiding Saracens.
51(tr. Jeffreys and Scott, revised)
Chr. Ede. 104:In the year 843 (532) in the month of September the patrician Mar
Rufinus made peace between the Romans and Persians; and this peace lasted until the
year 851 (539/40). (tr. M. Greatrex)
A column from Hierapolis, with a fragmentary inscription on every side, commemorates the con-
clusion of the peace.
Roussel 1939: I (366) The lord Justinian alone, by divine counsel, made peace, and he
put a stop to the lamentations of war (which had been) upon (the) cities for three
52. With Rufinus the general … .
Roussel 1939: II (367) By the good will of the Lord Christ and for the preservation of
our city of the holy apostles Peter and Paul, peace was made. Being cons
ul … .
Roussel 1939: III (367) + The cross extinguished the terrible roarings
53of war and the
measureless hardships of life, as if (they were) a rough wave or a fir
Roussel 1939: IV (367) Peace was made between the Romans and Persians with (the
help of) holy God; the money was handed over and … .
The peaceful state of relations in the 530s
By the mid-530s, Justinian could take pride in his achievements on the eastern front alongside his
more recent successes in the West. He appears, however, not to have taken the current peaceful state
of relations with the Persians for granted.
Nov.J.1.proem: While we have been busied with the concerns of the whole state and

have preferred to concentrate not on trivial matters, but (on) how the Persians may
remain at peace, (how) the Vandals, along with the Moors, may obey us, (how) the
Carthaginians, having regained their former liberty, may enjoy (it), (and how) the
Tzani, for the first time (brought) under the Roman state (something which God had
not granted to the Romans until now, only during our reign) may be enrolled among
its subjects, individual concerns from our subjects have constantly been reported and
have flooded in; to each of these we give an appropriate response.
The following passage describes a meeting between the anti-Chalcedonian bishop of
Constantia (Tella), John, who had been captured in the Jebel Sinjar by a joint Roman and
Persian force, and themarzbanat Nisibis.
55John’s replies to the marzban’s questions clearly
show how good relations were between the two states, at least up to 537, and how permeable the
frontier was at this time. It is not surprising therefore that envoys from the Gothic king Vitiges
were able to make their way from Italy to Persia in order to establish r
elations with Khusro.
Elias, V. Joh. Tel. 71.21–73.11: When themarzbanheard this (that John was a
bishop) he at once gave an order (summoning him) and sat down on the ground in his
presence. And he spoke with him through an interpreter, who said to him in Greek:
‘How did you dare, a man such as you, to cross over into our territory without our
(permission)? Do you not know that this is another state?’ The blessed man replied,
speaking to him through (p.72) the interpreter in Greek, ‘It is not the first time that I
have crossed over into this land. This is the third time that I have crossed over, in order
that I might pray among these saints who have lived for many years on the mountain
(Jebel Sinjar) from which you took me away as an evil-doer. For who am I that your
greatness knows of me and (knows that) I had crossed over then? For I am a poor man,
just as you see me. Today, while there is complete peace between these two kingdoms,
I did not know one state from another. For the two kings are brothers in love; and, if I
am here, I think I am among Romans, and, if I am among Romans, I am here on
account of (that) peace.’ The marzbansaid, ‘How is that you say your are a poor man,
when much gold was retrieved from you by those who came to you?’ The confessor
replied, ‘Lo! Suddenly those whom you sent seized me; let them tell what they found
with me when they seized me and searched me. And again I ask that, if it is easy while I
am imprisoned here, your greatness send trustworthy men to go round our whole
land. Let them question both my enemies who have accused me, and my friends, and
let them show that anything has been taken by me from any man. And if any of those
(charges) with which I have been accused are found (to be true), let me be crucified in
the middle of this city. What I had from my parents I have given away, and have
received (items) from others.’ The marzbansaid, ‘Why have you rebelled against
Caesar and those who hold authority (p.73) in his land? Do you not know that the
man who rebels against his masters behaves wrongly? And, if you wish, I shall recon -
cile you with the Romans and you will behave according to their will, and live in peace
and quiet.’ The holy man said, ‘I have not rebelled against our victorious and peaceful
emperor, who is beloved by men. I, according to my duties, as we are bound, am his
servant, since I am of the world. And I pray on his behalf that he may judge his empire

in accordance with God’s will.’ And when (themarzban) had heard him praising the
emperor (so) fully, he shouted out and said, according to their custom of praise, ‘May
the fortune of the king of kings, who is the equal of God, be with you.’
57(tr. M.
The Martyrdom of Grigor confirms this picture. Cf. Joh. Eph. Lives,PO18.543–4 (Life of
Susan), also in Brock and Harvey 1987: 134–5 with Lee 1993a: 56–7 on the openness of the
frontier at this time.
Mart. Grig. 360.5–10:God made peace between these two kingdoms. In the year
forty (of Kavadh’s reign) 58King Khusro came to the throne, and in the third year of his
reign 59he ordered an ambassador to go to the Roman emperor, in order to confirm the
peace between the two kingdoms. (For this mission) Zabergan, one of the nobles of
the kingdom, was chosen.
60(tr. M. Greatrex)
Deterioration of Roman defences
Although Bury confidently ascribed Justinian’s refortification and upgrading of cities in Meso -
potamia, Osrhoene and Syria to the period of the Eternal Peace, there is no evidence to support
this. On the contrary, given that in the fifth century a ban on fortifications near the frontier had
been part of a peace agreement (see Chapter 3 above), and that Justinian had been obliged to
move the headquarters of the duxof Mesopotamia from Dara to Constantia, it is highly
unlikely that Khusro would have tolerated such a building programme.
61Furthermore, Justin-
ian seems to have wished to cut his expenditure on the defences of the East to the maximum
possible extent.
Proc. Anecd. 24.12–14 (148.11–149.14): And I shall add one further item to those I
have mentioned, since the subject of the soldiers leads on to it. Those who had ruled
the Romans in earlier times stationed a very great multitude of soldiers at all points of
the state’s frontier in order to guard the boundaries of the Roman domain, [and]
particularly in the eastern portion, thus checking the inroads of the Persians and Sara -
cens; these troops they used to call limitanei. (13) These the Emperor at
first treated so casually and meanly that their paymasters were four or five years behind
in their payments to them, and whenever peace was made between the Romans and
the Persians, (p.149) these wretches were compelled, on the supposition that they too
would profit by the blessings of peace, to make a present to the Treasury of the pay
which was owing to them for a specific period. And later on, for no good reason, he
took away from them the very name of an army. (14) Thereafter the boundaries of the
Roman empire remained without guardposts, and the soldiers suddenly found them -
selves obliged to look to the hands of those accustomed to works of piety.
Dewing, revised)

Reorganisation of Armenia (536)
Following his military reforms in Armenia, Justinian now undertook a systematic reorganisa-
tion of the Roman Armenian territories, seeking to integrate the former satrapies and Armenia
Interior into the empire. He established four provinces of Armenia. The frontier provinces were
Armenia I, which incorporated both Theodosiopolis and Satala, and Armenia IV, which took
over the old satrapies and had its chief city at Martyropolis (Nov.J.31, tr. in Adontz 1970:
133–6). See Toumanoff 1963: 174–5, Adontz 1970: ch.7, Hewsen 1992: 18 (with a map),
Garsoïan 1999: 195–6. Roughly simultaneously with these developments in Armenia, the
Roman grip on the Transcaucasus tightened. Roman troops were deployed in Lazica and a city
established at Petra. In addition, the new commander there, John Tzibus, restricted trade, to
the vexation of the Lazi (Proc. II.15.1–26). See Lordkipanidse and Brakmann 1994: 85–6,
Braund 1991: 222–3, Braund 1994: 290–5. ( … ) Pontus Polemoniacus comprises five other (cities) – Neocaesarea,
Comana, Trapezus, Cerasus and Polemonium, for Pityus and Sebastopolis are to be
counted among forts rather than cities. As far as these extend the two Ponti (i.e.
Pontus Polemoniacus and Helenopontus). After these, our Lazica is established, in
which is also the city of the Petraeans, which has taken its civic identity and title from
us, using the name of our piety and being called Justiniana. (There lie also)
Archaeopolis and Rhodopolis, very large and ancient forts. Amid these are also the
forts of Scandis and Sarapanis, which we took from the Persians, and Mourisius
Lysiris and whatever works we have undertaken among the Lazi. Thereafter the land
of the Tzani takes over, now in our day, for the first time, held by the Romans, which
is receiving recent civic foundations, and which will receive foundations yet to be
made. But after that land the peoples are the Suani and Scymni and Apsilae and
Abasgi and others now, by the grace of God, friendly and ours.
64(tr. Braund, revised)
Justinian clearly had his sights on the whole western Transcaucasus. From Procopius
(VIII.3.13–21) we know that it was at just this time that the Abasgi were converted to Chris -
tianity through the efforts of the Abasgian Euphratas. See Colvin 2003.
Euphratesia threatened by an Arab incursion (536)
The fragility of the eastern frontier is well illustrated by this rare glimpse of border negotiations. 65
Marc. Com. Addit.a.536.11: Indeed in that very year, on account of the excessive
drought, pasture land in Persia was refused and about 15,000 Saracens with the
phylarchs Chabus (Ka‘b) and Hezidus (Yezid) were driven across the borders of
Euphratesia by Mundhir. There the duxBatzas encouraged them partly by flattery
and partly by peaceful restraint and repressed their desire for war. (tr. Croke, revised)

Roman–Persian cross-border cooperation (late 536)
According to Elias, the biographer of John of Tella, the patriarch of Antioch, Ephraem, was so
eager to seize the anti-Chalcedonian bishop of Constantia that he enlisted the help of the Per-
sians (albeit by a deception).
Elias, V. Joh. Tel. 65.24–66.21: (p.66) And, in order that he (Ephraem) might alto -
gether block (that) fount of benefits, he schemed and found men to perform his will;
and he sent them to the marzbanof Nisibis of the Persians. He persuaded him
through them, since he had agreed (to pay) a not inconsiderable bribe to him, to send
urgently some of his soldiers, and the Romans with them, (in order that) this blessed
man, famous for (his) victories of the spirit, might be seized from the mountain called
Sinjar (Singara). For the saint had retreated here on account of the many things of use
(to him) in that hard time. Those who had been sent to that pagan marzbanby
Ephraem repeated to him many lies against that man, and supposed that he had
acquired a lot of gold. That pagan, whose name was Mihrdades, inflamed by desire,
sent off many cavalrymen against him with a certain commander from among his own
(men), who was also himself a pagan. And the Persians and Qadishaye and some of the
Romans reached the mountain, since they had also brought scouts with them,
comprised of Romans and Persians who knew the place. And, furthermore, there was
(a man) who had been sent by Ephraem to complete the task, who was from the for-
tress of Barbalissus (Beth Balash) and who had been made what it called a brigand-
hunter in the territory of Antioch, whose name was Cometas.
66(tr. M. Greatrex)
A coup at Dara (537)
The most detailed source for the uprising is Procopius (Wars I.26.5–12), according to whom
the initiator of the coup, John, was an infantry soldier, and enjoyed only limited support for his
venture. See Greatrex 1998a: 220.
Marc. Com. Addit.a.537.4: In the East too, John Cottistis was killed at Dara while
usurping power, before he could undertake any hostile action. (tr. Crok
e, revised)
Zach. HEX.1 (176.8–15): Now there went up … in teen, and also …
the king … much … he told him about
named But … ed a
rebellion in Dara in the summer of the year, who was … And he freed the king from
distress of mind, in what way I cannot truthfully state, and therefore keep
silence. (tr. Hamilton-Brooks, rev. M. Greatrex)

The southern front (540–5)
The background to the war
The second Persian war of Justinian broke out in early summer 540, when a large Persian army
commanded by Khusro invaded Roman Mesopotamia and Syria. According to Procopius,
Khusro was encouraged to undertake the campaign both by an emissary from the Gothic king
Vitiges and by Armenians discontented with Roman rule (Proc. II.2–3).
1Justinian’s continuing
campaigns in North Africa and Italy and the consequent reduction in manpower on the Roman
eastern front must have encouraged Khusro’s ambitions; he may too have been aware of the
poor morale among the limitanei, whose pay had been cut since the Eternal Peace.
likely contributory factor was Khusro’s continuing need for funds, since his kingdom was still
having to make tributary payments to the Hephthalite Huns.
3Khusro initially (in 539)
attempted to exploit a dispute between the Lakhmid and Ghassanid Arabs (over ownership of
grazing land known as the ‘Strata’, i.e. the region through which passed the Strata Diocletiana)
to justify an attack on Roman territory, but the Romans successfully defused the issue by pre-
varicating. Next the Persian king alleged that Justinian had sought to bribe the Lakhmid ruler
Mundhir through his envoy Summus, and that he had encouraged some Huns to invade Per-
sian territory (Proc. II.1). See Bauzou 1993: 36 (on the name of the Strata), Shahîd 1995:
210–16 (critical of Proc.’s portrayal of the affair), Key Fowden 1999: 66.
The Persian tradition is also aware of the quarrel between the two Arab leaders, but
(unsurprisingly) portrays the Jafnid ruler Harith as the aggressor. See Shahîd 1995: 216–18
(noting an acccount similar to Tabari’s in Firdausi, tr. Mohl 1868: 195–201).
Tabari, I, 958/252–3 (Nöldeke 238–9): Between Khusro Anushirvan and Justinian,
emperor of the Romans, there was, as has been mentioned, a total cessation of hostili -
ties. But enmity arose between a certain Arab whom Justinian had appointed king
over the Arabs of Syria and who was called Khalid ibn Jabala,
4a man of Lakhm whom
Khusro had appointed king over the territory between Oman, Bahrain and Yamama
as far as Ta’if and the rest of the Hijaz together with its Arab inhabitants and who was
called Mundhir ibn al-Nu‘man. Therefore Khalid ibn Jabala invaded Mundhir’s
domain, killed a great number of his men and took much of his wealth as booty.
Mundhir complained of that to Khusro and asked him to write to the king of the
Romans that he might obtain justice for him from Khalid. So Khusro did write to

Justinian, mentioning the treaty of peace between them and informing him what
Mundhir, his prefect over the Arabs, had met with at the hands of Khalid ibn Jabala,
whom he (Justinian) had appointed king over the Arabs in his lands, and asking him
to order Khalid to return to Mundhir the booty he had taken from his domain and to
pay him bloodmoney for the Arabs belonging to it, whom he (Harith) had killed. (In
short Khusro asked that) he (Justinian) obtain justice for Mundhir from Khalid and
that he not treat lightly what he had written to him about, in case the peace treaty
between them thereby be nullified. Repeatedly he wrote letters to Justinian to obtain
justice for Mundhir, but he (Justinian) did not heed them. (tr. Hoyland)
The Persian invasion, early summer 540
In late 539 Justinian, aware of Persian plans, despatched a message to Khusro through
Anastasius of Dara to dissuade him from his expedition (Proc.WarsII.4.14–26). 5But in May
540 6Khusro entered Roman territory nonetheless. He invaded Euphratesia at the confluence of
the Khabur and Euphrates, proceeding along the southern bank of the Euphrates, thus avoiding
the fortress at Circesium (II.5.1–6). He made a half-hearted attempt to induce the fortress of
Zenobia to surrender (II.5.7), but when he met with no success, he quickly moved on to Sura.
The garrison of Sura, under the commander Arsaces, resisted the first Persian attack with some
sucess, but Arsaces himself was slain in the fighting. Dispirited by this development, the inhab-
itants sent their bishop to negotiate with Khusro. He appeared to accept the entreaties of the
bishop, but instead seized the opportunity of taking the city by surprise and sacking it
(II.5.8–27). The Roman prisoners were then ransomed by the bishop of nearby Sergiopolis,
Candidus, who pledged to pay the Persian king two centenaria(II.5.28–33).
The general in charge of the Roman defences in Belisarius’ absence was Bouzes. He proposed to
defend his base at Hierapolis by positioning some of his forces in the neighbouring hills; but
having led off the best troops to do this, he lost contact entirely with those remaining in the city
(Proc. II.6.1–8). Justinian, apprised of the invasion, despatched his cousin Germanus to bolster
the defences of Antioch, but provided him with only 300 men. Discovering that the city was
vulnerable to assault, Germanus and the city’s inhabitants decided to negotiate with Khusro
through the agency of Megas, bishop of Beroea (II.6.9–16).
7Despite Megas’ intervention,
Khusro proceeded to threaten Hierapolis, whose defenders quickly agreed to buy off the king
for 2000 lbs of silver. Following further entreaties from Megas, Khusro agreed to call off his
invasion if the Romans paid him ten centenaria(II.6.17–25). While Megas returned to Antioch
to report back to Germanus, the Persians advanced against Beroea and sacked the city; most of
the population gained refuge in the acropolis, however. At Antioch, the bishop discovered that
his offer could not be followed up because John, the son of Rufinus, and Julian, Justinian’s a
secretis, had arrived in the meantime and forbidden (on the emperor’s orders) that any money be
handed over to the enemy. The patriarch Ephraem, having been accused of favouring the city’s
surrender, withdrew to Cilicia, where he was joined soon afterwards by Germanus (II.7.1–18).
Other sources offer scraps of additional information.
Mal. 18.87 (405.65–9/479.23–480.5): In the month of June of the third indiction
Antioch the Great (p.480) was captured by Khusro, the Persian king. Germanus was
sent with his son Justin to carry on the war, after being appointed magister militum.

Having achieved nothing, he stayed in Antioch buying silver for two or three
nomismataalitra from the Antiochenes. 9( … ) (tr. Jeffreys and Scott, revised)
Evagrius, HEIV.25 (172.11–14): He (Ephraem) is said to have preserved the church
and everything around it by adorning (it) with the holy offerings, so that they might
be a ransom for it.
Megas returned to Beroea and attempted to negotiate the release of those besieged in the city’s
acropolis. Khusro allowed them to depart (II.7.19–35). The defection of some soldiers to the
Persians, noted here by Procopius, is remarkable, and indicates the low level of morale in the
East at the time, perhaps resulting from the cut-backs in pay implemented in the wake of the
Eternal Peace. See p.99 above.
Proc. WarsII.7.36–7 (183.19–184.4): Then the Beroeans, after coming into such
great danger, left the acropolis free from harm, and departing went each in the direc -
tion he wanted. (37) Among the soldiers a few followed them, but most (p.184) came
as willing deserters to Khusro, putting forth as their grievance that the government
owed them their pay for a long time; and with him they later went into the land of
Persia. (tr. Dewing, revised)
The sack of Antioch (June 540) 11
As Khusro approached Antioch, many of the population fled. Some were encouraged to
remain, however, by the arrival of 6000 Roman troops under the ducesof Phoenice Libanensis,
Theoctistus and Molatzes. Once Khusro had arrived, he proposed to depart from the city for
the payment of ten centenaria. His offer was vigorously rebuffed by the Antiochenes, and
Khusro ordered that the city be attacked. Thanks to a fatal weakness in the city’s defences, the
Persians succeeded in gaining control of the walls; the newly arrived Roman forces therefore
immediately abandoned the city to the king. Despite the valour of the citizens of Antioch, and
in particular of the circus partisans, the city was taken by storm (Proc. II.8.1–35). Once enough
plunder had been removed from the city, it was razed to the ground with the exception of one
church (II.9.14–18). See Downey 1961: 542–4, Stein 1949: 489–
90, Evans 1996a: 156–7.
V. Sym. Styl. Iun. 57 (50–2):At that time what God intended to do in the city of the
Antiochenes was revealed to the holy man – that he intended it to be burnt from gate
to gate by the Assyrians, through whom the impending devastation (would occur).
And the servant of God prayed on its behalf, and God said to him, ‘Lo, the shouting of
the residents in it (Antioch) has gone up into my presence and the time for its requital
is at hand, on account of the lawlessness which they practise: they set forth a table,
libations, and a sacrifice to demons. Their excuse is the fortune of the city, (but) they
provoke me to jealousy against them. For that reason I shall hand them over to a sense -
12people.’ And lo, he beheld the Spirit, sent bearing a dagger into the city. This
vision the servant of God related to the brothers and certain Antiochenes devoted to a
pious life who had come to him. After a short interval of time, God aroused Khusro,
the Persian king, and the saying of the just man was close (to fulfilment), for a sub -
stantial multitude of Persians encamped outside the gates of the city of Antioch. And

the holy man cried out to (p.51) the Lord, (asking) whether he had changed his mind
about the things he had shown him earlier, and whether he would hand the city over
to the Assyrians (i.e. the Persians); and there was no explanation from the Lord,
because his heart was filled with anger. Again, therefore, for a second (time), he
(Symeon) earnestly prayed about this (event) and the Lord said to him, ‘I shall hand
over the city; nor shall I conceal from you what I intend to do. I shall fill it with the
enemy and hand over most of those who inhabit it to the slaughter; many of them will
also be led off as prisoners of war. Now therefore I dispense to you the prize of peace –
the symbol of my cross – as a protection for you, and in a short time you will see what
this vision (is).’ And again, in a trance, the Spirit, bearing the life-giving cross, came to
the holy man. And he saw two angels wearing linen robes and (bearing) drawn bows in
their hands and masses of arrows; and the servant of God asked what these things
were, and they said, ‘The cross is salvation and security against any enemy. The arrows
are for the heart of those wielding daggers in battle as they come against you, so that
they may be turned back in fright and so they may not prevail against you, through the
power of the Lord who sent us to guard you.’ After this divine vision had appeared to
him, he again beheld the city being besieged and captured by means of ladders
(placed) on the wall, and (he saw) the barbarians arriving inside. And (he saw) that
wailing and shouting rang out, and a massive flight from the town took place, while
many threw themselves from the wall; then (he saw) that the two gates of the city – the
one facing the sea and (the one facing) south – being opened; for on account of the
clamouring of the holy men, God forbore to destroy everyone down to the last man.
He also saw two of the monks with him deserting their posts and retreating; the head
of one was cut off, while the other was taken prisoner. (p.52) The divine vision, then,
took this form, and the event itself followed not long afterwards. For it happened that
the city was besieged and taken by the Assyrians and set alight from one gate to the
other, through not all of it was burnt. And some threw themselves from the wall, while
others, once the two gates were opened, fled to the south and south-west.
13And two
monks, growing afraid, retreated, and one of them was decapitated, while the other
was taken prisoner. Not one person of those remaining with him (Symeon) on the
mountain fell, because the prayer of the holy man was a strong tower for them in the
face of the enemy.
Joh. Lyd. De Mag.III.54 (216.14–22): But as the city (Antioch) was recovering, as if
from a nether gloom, through much effort, an abundance of resources and collabora -
tion between crafts, Justin (I) met his end, and the evil genius Khusro (came through)
Arabia with a countless host and fell upon Syria.
15The recently demolished city itself,
which appeared to him an easy prey since it was not fortified, he took in war and
burned, wreaking countless slaughter. The statues, however, with which the city was
adorned, he seized as booty, along with slabs of marble, precious stone and pictures; it
was the whole of Syria that he carried off to Persia. There was not a farmer or a tax -
payer left to the treasury.
16(tr. Carney, revised)

Negotiations for a Persian withdrawal
After the capture of the city, Khusro was approached by Roman ambassadors who urged him
not to continue his attacks on the empire; the king, in response, claimed that it was Justinian
who had violated the treaty (Proc. II.10.1–18).
Proc.WarsII.10.19–24 (196.16–198.4): Finally Khusro made the demand that the
Romans give him a large sum of money, but he warned them not to hope to secure
peace for all time by giving money at that moment only. (20) For friendship, (he said),
which is made by men on terms of money (p.197) is generally spent as the money (is
used up). (21) It was necessary, therefore, that the Romans should pay some fixed
annual sum to the Persians.
17‘For thus,’ he said, ‘the Persians will maintain a constant
peace with them, guarding the Caspian Gates themselves and no longer feeling resent -
ment at them on account of the city of Dara, in return for which the Persians them -
selves will be in their pay forever.’ (22) ‘So,’ the ambassadors said, ‘the Persians desire
to hold the Romans liable to the payment of tribute to them.’ (23) ‘No,’ Khusro said,
‘but the Romans will have the Persians as their own soldiers for the future, providing
them with a fixed payment for their service; for you provide an annual payment of
gold to some of the Huns and to the Saracens, not as tributary subjects to them, but in
order that they may guard your land unplundered for all time.’
18(24) After Khusro
and the ambassadors had held such discussions with one another, they later came to an
agreement by which Khusro should forthwith take from the Romans fifty centenaria
and receive a tribute of five more centenariaannually for all time. (In return) he should
do them no further harm, but taking with him hostages from the ambassadors (as
pledges) for the keeping of the agreement, he should make his departure (p.198) with
his whole army to his native land. There ambassadors sent from the Emperor Justinian
would put the peace terms on a firm (footing) for the future. (tr. De
wing, revised)
The progress of Khusro’s invasion
From Antioch Khusro continued westwards to Seleucia, the port of Antioch, where he bathed
in the Mediterranean. Having explored further the suburbs of Antioch, he then proceeded to
Apamea. Despite pledges given to the Roman ambassadors, he extracted great wealth from the
city, but forbore from taking the fragment of Christ’s cross (Proc. II.11.1–38, Evag. HEIV.26).
Khusro continued his predatory tour past Chalcis, which deflected him by a payment of two
centenaria. Crossing the Euphrates at Obbane, north of Barbalissus, and passing through
Batnae en route, he headed towards Edessa. Here he extracted a further two centenariaand
agreed not to damage the surrounding countryside (II.12.1–34). At this point Justinian
confirmed in writing the terms negotiated by his ambassadors earlier. Khusro therefore
prepared to ransom his captives, and the citizens of Edessa collected money for this purpose; but
Bouzes, who was in the city, prevented the exchange. The Persian army now moved south-east
to Carrhae, and from there to Constantia. Here too Khusro received a sum of money. Before
returning to Persian soil the king undertook an assault on Dara, despite the agreement made
with Justinian. A Persian mining operation was nearly successful, but was thwarted by a Roman
19Khusro therefore left the city, having received 1000 lbs of silver from the

defenders, and passed back into his own kingdom. His attempt on Dara had brought to
nothing the agreement reached earlier, and the two powers consequently remained at war
(II.13.1–29).The inhabitants of Antioch, whom the citizens of Edessa had been unable to ransom, were
given their own city, Veh-Antioch-Khusro (the better Antioch of Khusro) near Ctesiphon by
the king (II.14.1–4).
20And just as Justinian had adorned the Chalke in Constantinople with
mosaics depicting Belisarius’ triumphs, Khusro commissioned a mural of the fall of Antioch for the
throne room of his palace in Ctesiphon.
General notices on the campaign of 540
Chr. Ede. 105:In this year ( AG 851 = 540), in the month of May, Khusro, the Persian
king, broke the peace, invaded the lands of the Romans and took by force Sura,
Beroea and Antioch. He also gained control of Apamea; and, as he returned, he
reached as far as Edessa. But the grace of God protected it, and he inflicted no damage
on it; but once he had received two centenariaof gold,
22which the leading citizens held
out to him, he returned to his (own) land. (tr. M. Greatrex)
Jord. Rom. 376 (49.25–50.4): When the Parthian (i.e. Khusro) discovered this (Justin -
ian’s victories in the West), he blazed with the flames of envy and moved (his army),
mobilised for war, into Syria; laying waste Callinicum, Sura and Neocaesarea,
23he came
to Antioch. When Germanus, the patrician, together with his son Justin, the same man
who was consul, (reached Antioch) after returning from the province of Africa, he left
the city and withdrew to the region of Cilicia since he was unable to block the arrival of
the Parthians. The Persians, having gained control of Antioch, (which was) devoid of an
army, looked on as the populace, mingled with soldiers, fled along the course of the
Orontes to Seleucia Maritima. They did not pursue (them), but zealously seized
plunder throughout the city. They passed by the neighbouring cities and towns,
attacking some and exacting a sum of money from others; and (so) the Parthian gath-
ered for himself the wealth of the whole of Coele Syria within the space of one year.
Marc. Com. Addit. a.540.1–2: The Parthians invaded Syria and overthrew many
cities. Germanus took up arms against them, and brought with him his son Justin the
consul, while he was actually in office. (2) Antioch the Great was ravaged and demol -
ished by the Persians (tr. Croke, revised)
Ps. Dion. II, 69.7–15: The year 850 (538/9): there was a siege by the Persians of
Antioch and it was subdued, ravaged and burnt. Thus at that time, in the fourteenth
year of Ephraem’s (period as bishop), powerful Persian troops, coming with their king
Khusro, attacked and subdued the city of Antioch. They burned it with fire and
pulled it down. They stripped it and removed even the marble slabs which overlaid the
walls, and took them away to their country, since they also were building in their
country a city like this one and named it Antioch (as well).
24(tr. M. Greatrex)
Tabari, I, 898/157–8 (Nöldeke 165–6): When his (Khusro’s) rule was on a firm foot -
ing and all countries were subject to him, he moved, after a few years of his rule,
against Antioch, inside which were the commanders of Caesar’s army. He captured it,

then he ordered that a plan be drawn up for him of this city of Antioch, taking account
of its dimensions, of the number of its dwellings and roads, and of everything that was
in it, and (he ordered) that a city be constructed for him according to this plan, next to
Mada’in. Thus was the city know as Rumiya
25built, on the model of Antioch, and he
transported the people of Antioch to take up residence in it. When they entered the
gate of their city, each resident proceeded to a house which so resembled his own in
Antioch that it was as though he had never left it. Then he (Khusro) headed for the
city of Heraclea and captured it, next Alexandria and its environs.
26He left a portion
of his army behind in the land of the Romans after Caesar had submitted to him and
conveyed to him the ransom.
27(tr. Hoyland)
The Roman invasion of Assyria (541)
Two campaigns took place in the following year. While Belisarius undertook an invasion of
Persian territory in retaliation for Khusro’s attack, the Persian king opened up a new theatre of
war in the Transcaucasus (on which see Chapter 8). Belisarius arrived in the East to take charge
of the Roman campaign and quickly sought to bring his forces up to strength: according to
Procopius (II.16.3), many soldiers had no weapons or armour. Learning that Khusro was else -
where, he prepared to invade Persian territory at once. The two ducesof Phoenice Libanensis,
Theoctistus and Rhecithancus, were unwilling to join the expedition because of anticipated
raiding by Mundhir, but were obliged to do so by Belisarius, who was aware that the (pagan)
Lakhmid Arabs were in the habit of abstaining from raids in spring for religious reasons.
agreed, however, to release their forces when this period was ended (II.16.1–19). The Roman
army approached Nisibis from Dara and encamped a few miles away; two commanders, Peter
and John Troglita, the duxof Mesopotamia, insisted on taking their troops much closer to
Nisibis, despite Belisarius’ warnings. They were therefore easy victims to a sally by the Persian
garrison, led by the general Nabedes; but the attackers were quickly driven back by the forces
under Belisarius which came to the rescue (II.18.1–26). A tendentious account of the incident
is preserved in Corippus’ poem in praise of John, the Iohannis.
Corippus,Ioh.I.56–67: For the glory of the man (John Troglita), and the signs of dis -
tinguished service, and the wars won please (the emperor), as well as the more weighty
(conflicts) against a proud kingdom.
29(The emperor considers) how he expelled the
Persians, striking down and wounding the confident Parthians, and standing up to
the dense (volleys of) arrows on the occasion when the broad fields of Nisibis ran with
the blood of the Persians. Nabedes, second (in importance) only to the Parthian king,
relying on (his own) courage in the fierce battle, lost his allied units, even as he himself
was winning; and, in his flight, fearing for his post, he was barely able to shut the gates
(in time), so that the Roman cavalry was unable to burst into the midst of the citadel
of Nisibis.
30And the victorious John shook his spear at the lofty gates of the Persians.
Passing by the well defended Nisibis, Belisarius made instead for Sisauranon, a smaller Persian
fort to the east. Here he divided his forces, sending off Harith with 1200 Roman troops under
John the Glutton and Trajan to plunder Assyria (Beth Aramaye) and then report back to him.
With the remainder of his men, Belisarius soon brought the defenders to surrender; the Persian

garrison he sent to Justinian, while the Christian inhabitants were permitted to depart and the
city-walls razed (Proc.II.19.1–25). 31Harith’s forces meanwhile pillaged Assyria, but the Jafnid
leader sent off the Roman detachment which had accompanied them; John and Trajan there -
fore returned to Roman territory along the southern bank of the Euphrates, ending up, having
crossed the river, at Resaina. Belisarius, lacking information on Harith’s progress, and under
pressure from his commanders to withdraw, moved back westwards (II.19.26–46). The limited
success of the Roman invasion of Assyria in 541 was controversial. In the Wars, Procopius
accuses Harith of deliberately concealing his whereabouts from Belisarius in order to keep his
32he also stresses the poor condition of the Roman forces, as well as the fact that other
commanders insisted that the Romans withdraw. But in the Anecdota(2.18–22) he claims that
Belisarius’ decision was driven entirely by his desire to meet his wife Antonina, who, he had
heard, was on her way to the East. In this context he makes the followin
g assertion.
Proc. Anecd. 2.25 (17.4–10): And yet if he had been willing in the first place to cross
the Tigris river with his whole army, I believe that he would have plundered all the
lands of Assyria and have reached the city of Ctesiphon without encountering any
opposition whatever, and, after rescuing the prisoners from Antioch, and all the other
Romans who chanced to be there, he would finally have returned to his native land.
( … ) (tr. Dewing, revised)
General notices of the campaign of 541
Agap. PO8.431: In this year 34a certain Arab called al-Harith ibn Jabala attacked the
Persians. Since Khusro was afflicted with these ulcers, he (Harith) despatched troops
against him and the Persians were defeated. He destroyed many of their cities and
took captive a great number of them. Then one of Khusro’s marzbansmarched against
them and defeated them and recovered all the captives from them. (tr. H
Marc. Com. Addit. a.541.1: Since the Parthians continued to be hostile, Belisarius
undertook the eastern campaign after Germanus had returned to the royal city. (tr.
Croke, revised)
Jord. Rom. 377 (50.4–9): And he (Khusro) did not therefore desist (then), but con -
tinuously attacked the Roman state. Against him was appointed the consul, the victor
over Vandals and Goths, as usual.
35Although he did not subdue him (Khusro) as he
had other peoples, he nonetheless forced him to regroup his forces within his own ter -
ritory; and the fortunate commander would have obtained a victory from this people
too, had not the disaster in Italy, which had broken out after his departure, required
him to be swiftly replaced by Martin. ( … )
Jac. Ede. Chr.p.320: ( … ) And the Romans laid waste the lands of
C and Arzanene and ‘Arabaye.
36(tr. M. Greatrex)
The campaign of 542
The eastern provinces, already reeling from the rapacious invasion of Khusro of 540, received a
double blow in 542. For not only did a Persian army return to Euphratesia and Osrhoene, but

the whole eastern Roman empire was visited by plague. Although the extent of depopulation
brought about by the plague remains a matter of dispute, its arrival was sufficient to persuade
Khusro to withdraw from Roman territory not long after his invasion had
Khusro invaded Euphratesia in early summer. As the army advanced towards Roman territory,
a Persian convert to Christianity, Pirangushnasp (who had taken the name Grigor), was
martyred near Pirisabora.
38Candidus, the unfortunate bishop of Sergiopolis, unsuccessfully
pleaded for forgiveness from Khusro, having failed to hand over the money he had promised by
way of ransom for the prisoners taken at Sura. The inhabitants of Sergiopolis therefore handed
over many treasures,
39but no Persians were allowed in the city. A Persian attempt to capture the
city with 6000 men failed; although there were only 200 soldiers stationed there, all the citizens
took part in the defence of the city walls (Proc. II.20.1–16, Evagr. HEIV.28 [a more pious
version]). See Key Fowden 1999: 133–4.
Proc. WarsII.20.17–19 (240.13–241.3): But when Khusro arrived at the land of the
Commagenae, which they call Euphratesia, he had no desire to turn to plundering or
to the capture of any stronghold, since he had previously taken everything to hand as
far as Syria, some things by capture and others by exacting money, as has been set forth
in the preceding narrative. (18) And his purpose was to lead the army straight for Pal -
estine, in order that he might plunder the treasures (there) and especially all those in
Jerusalem. For he had it from hearsay that this was an especially rich land with wealthy
inhabitants. (19) And all the Romans, both leaders and soldiers, were in no way
inclined to confront the enemy or (p.241) to stand in the way of their passage, but
taking to their strongholds as each one could, they thought it sufficient that they
should guard these and be safe themselves.
40(tr. Dewing, revised)
Justinian again despatched Belisarius to retrieve the situation. He established Europus on the
Euphrates as his base, and there he summoned Bouzes and the other Roman commanders who
had sought refuge in Hierapolis (Proc. II.20.20–28). Belisarius’ presence just to the west of the
Euphrates effectively prevented Khusro from advancing further, and the king, over-estimating
the forces available to the Romans, decided to retreat. He crossed the Euphrates and returned to
Persian territory, receiving John of Edessa as a hostage.
41Although Khusro had agreed in return
not to damage Roman territory as he withdrew, he could not resist the opportunity afforded by
the weakness of Callinicum. Belisarius was nonetheless acclaimed throughout the East for his
success in repelling the Persians (II.21.1–29).
Proc. WarsII.21.30–32 (248.12–25): But in the meantime Khusro, disregarding
what had been agreed, took the city of Callinicum which was defended by absolutely
nobody. For the Romans, seeing that the wall of this city was altogether unsound
and easy of capture, were constantly demolishing some portion of it in turn and
restoring it with a new construction.
43(31) Now just at that time they had demol -
ished one section of it and had not yet built in the space left; when, therefore, they
learned that the enemy were close at hand, they carried out the most precious of
their treasures, and the wealthy inhabitants withdrew to other strongholds, while
the rest without soldiers remained there. (32) And it happened that a very great

number of farmers had gathered there. These Khusro enslaved and razed everything
to the ground. 44(tr. Dewing, revised)
Jac. Ede. Chr.p.320: Khusro went up and laid waste Callinic and all the south -
ern land(s) of Mesopotamia. (tr. M. Greatrex)
Roman efforts to refortify the East (540–65)
The defences of Callinicum were not alone in receiving attention at this time. Inscriptions from
above the gate of the citadel at Cyrrhus testify to building work being carried out there in the
540s, which has been connected with Belisarius’ presence at nearby Europus in 542.
45From 540
onwards serious efforts were made to upgrade Roman fortifications, and particularly those in
Syria which had been neglected during Justinian’s earlier phase of construction (527–532). See
Lauffray 1983: 35, Whitby 1986a: 729–30, Whitby 1988: 211–12 and C
hapter 16 below.
IGLS 145:Up with the victory of the general Belisarius. 46
IGLS 146:Many years to Justinian the emperor and to the Augusta Theodora many
IGLS 147:Many years to Eustathius the domesticus, by the grace of God.
The conclusion of Khusro’s invasion of 542
While Khusro himself presumably continued his retreat along the Euphrates to Persian terri-
tory, he may have detached some of his forces, under the command of Mihr-Mihroe
(Mermeroes), to strike at fortified cities further to the north-east.
Zach.HEX, index (173.23–174.2): The eighth (chapter), 49concerning Khusro, who
went up (p.174) and took Callinicum and the other camps on the frontier of the
Euphrates and the Khabur.
Agap. PO8.431: In year 17 of Justinian(’s reign), Khusro, son of Kavadh, attacked
Kafr-tut and Resaina. But Belisarius
50marched against them with the Byzantine army
and drove him back before he (Khusro) could take them (the two cities). 51(tr.
Corippus, Ioh.I.68–98: Before the eyes of the emperor all the brave deeds of the
faithful man pass by. He considers and ponders (his) labours. (He recalls) how a thick
(mass of the) enemy had blockaded Theodosiopolis in an oppressive siege, and how,
in the shadows of night, he (John) had swiftly brought aid to the ramparts of the
endangered city, entering the friendly gates (having passed) through the midst of the
enemy. (The emperor was) terrified as the huge Mihr-Mihroe moved off from those
walls, and as, with weapons bristling, he more fiercely headed for Dara, where the
leader (John) kept his standards, (a place whose) towering ramparts are surrounded by
a brilliant wall, and put the Roman formations to test in battle. But after the watchful
leader had snatched the first city from the enemy, he then followed those fleeing and
occupied the roads and took control of all the fields, so that the cruel enemy might not
lay (them) waste or harm anyone. He reached the ramparts of the lofty wall (of Dara)

first, and he did not allow (the issue) to be delayed for long. For at once he boldly ven-
tured to engage the enemy in the middle of the plains and, in a favourable battle, he
struck down innumerable formations, eminent leaders and allied nations. He put to
flight the vanquished and confounded leader of the Parthians, Mihr-Mihroe. In the
plains every Persian, in fear of the pursuing Romans, cast his sword from his hand
(and threw off) his gleaming decorations into the midst of the fields. A Median sword
and an empty scabbard gleam on every expanse of sand; and spears, shields, hairs
bodies, as well as the horse of the armour-bearer of the proud commander, (all) tan -
gled up with his weapons. He too (Mihr-Mihroe) would have lain there, laid low in
the plains, had not the magnanimous leader (John) wished to take him alive. Mihr-
Mihroe, however, saw the towering ramparts (of Dara), as he entered (the city) with a
few companions.
Harith’s involvement in ecclesiastical matters (542/3)
John of Ephesus, in his Life of James and Theodore (PO19.153–4), recounts how the Jafnid
phylarch Harith appealed to Justinian’s wife Theodora to appoint some bishops for the
Monophysites of Syria: few now remained in the wake of the persecutions in the 520s and 530s.
Theodora therefore arranged for the consecration of Jacob (known as Jacob Baradaeus) and
Theodore as bishops for the region, the former of whom proved to be an energetic convert of
pagans and organiser of the Monophysite community. Jafnid support of the Monophysite
Church was consistent throughout Harith’s reign and that of his successors. When, under
Maurice, an attempt was made to convert the Jafnids, it proved wholly unsuccessful. See Shahîd
1995: part 2 (on ecclesiastical history), esp. 755–75 on this episode, cf. Evans 1996a: 185–6
(dating Harith’s intervention to 541), Stein 1949: 625, Frend 1971: 285, Bundy 1978: 78–9.
The Roman counter-offensive of late summer 542
This took place in Armenia, but involved troops from Mesopotamia. See below, Chapter 8 with
n.8 (for the dating of this campaign, usually placed in 543).
The defection of a Persian prince (c.543)
Kavadh, the grandson of King Kavadh and son of Zames, who had rebelled unsuccessfully
against Khusro, sought refuge at the Roman court soon after the death of his protector, the
kanarang Adhurgundadh (Adergoudounbades). See Mosig-Walburg 2000: 69–70.
Proc.WarsI.23.23–4 (121.26–122.6): Not long after this (the death of Adhurgundadh)
(p.122) either Kavadh himself, the son of Zames, or someone else who was assuming
the name of Kavadh came to Byzantium; yet he resembled very closely king Kavadh in
appearance. (24) And the Emperor Justinian, though in doubt (concerning him),
received him with great friendliness and honoured him as the grandson of Kavadh.
( … ) (tr. Dewing, revised)

The campaign of 543
Khusro, presumably again in spring, directed his next invasion at Edessa. Despite initial losses,
he energetically laid siege to the city, erecting a huge mound to tower over the walls; elephants
were also employed in the operations. Attempts by the defenders, and in particular a doctor,
Stephanus, who had served Kavadh, to induce the king to raise the siege proved vain (Proc.
II.26.1–46, VIII.14.35–7). The defenders destroyed the Persian mound by undermining it and
then setting fire to it (II.27.1–17),
56and so another Persian attack was brought to nothing. Fur-
ther assaults had no more success, and negotiations were begun; a Roman envoy, Rhecinarius,
had arrived during the siege. The Edessenes paid five centenariato Khusro, and the Persians
departed after nearly two months (II.27.18–46, Chr.123456, 192–3 [below]). See Bury 1923:
II, 107–10.
Jac. Ede. Chr.p.321: Khusro continued to wage w against Edessa and laid waste
Batnae. 57(tr. M. Greatrex)
A truce is declared, 545
In the wake of the Persian retreat, two Roman envoys, the newly appointed magister militum
Constantianus, and Sergius, then proceeded to Ctesiphon to arrange a truce with Khusro (Proc.
Proc. WarsII.28.7–11 (283.7–23): But Khusro said that it was not easy for them to
come to terms with each other, unless they should first declare an armistice. Thus they
might go back and forth to each other more securely and resolve their differences and
henceforth arrange matters for a peace in safety. (8) And it was necessary, (he said),
that in return for this continuous armistice the Roman emperor should give him
money and should also send a certain physician, Tribunus by name, with the condi-
tion that he spend some specified time with him.
58(9) For it happened that this physi-
cian at a former time had rid him of a severe disease, and as a result of this he was
especially beloved and greatly missed by him. (10) When the Emperor Justinian heard
this, he immediately sent both Tribunus and the money, coming to twenty centenaria.
(11) In this way the truce was made between the Romans and the Persians for five
years, in the nineteenth year of the reign of the Emperor Justinian (545–546). (tr.
Dewing, revised)
Marc. Com. Addit.a.546.4: In the East a treaty was entered into with the Parthians
through Constantianus, the magister militum, and the army returned to Constantino -
59(tr. Croke, revised)
Jord. Rom. 377 (50.9–11): (…)He(Martin), though unequal in strength (to
Belisarius), was not inferior in planning (even if he was collaborating with
Constantianus), and although he was not strong enough to resist the Parthians, he
brought about peace, so that he would not have to maintain hostilities for a long time.
In fact, hostilities resumed in Lazica in 548, and it seems most likely that, as at other times, the
northern sector of the frontier (or in fact just Lazica in this case) was excluded from the truce. 60

Composite entry concerning the war in Mesopotamia, 540–545
Chr. 123456, 192.29–193.10: At this time Khusro, the Persian king, went up
against (p.193) Edessa. And he besieged it for 55 days; 61and when he proved unable to
subdue it, he laid waste the whole region, and Sarug (Batnae), as far as the Euphrates.
From there he set out and laid siege to Antioch, and subdued it, dismantled it and laid
it open; and he took all the treasure of the churches, even the marble tablets which
were covering the walls, and brought them back to Persia. And on his way back he
came and crossed the Euphrates, and reached Callinicum; and he laid waste Balash
62and all the banks of the Euphrates and the region of Callinicum. And
he took away the bones of Mar Bacchus and the gold covering the reliquary of Mar
Sergius of Rusafa (Sergiopolis) and all the treasure of the church.
63(tr. M. Greatrex)

The northern front (540–62)
The invasion of Lazica (541)
One factor behind Khusro’s breaking of the Eternal Peace was the prevailing dissatisfaction
with Roman rule in Transcaucasia. The Armenians had previously complained to him of their
treatment at the hands of Justinian, while the Lazi now also came forward to urge Persian inter -
vention, having been forced to endure the quartering of Roman troops on their territory and
the rapacity of imperial officials (Proc. II.15.1–30).
1Accordingly Khusro made preparations to
invade Lazica, masking his intent by claiming that he was instead intending to campaign against
Huns who had attacked Iberia (II.15.31–5).
2Having entered Lazica, Khusro was met by the
Lazic king Gubazes, who formally handed over control of the kingdom to him. The Persians
next moved to seize control of the chief Roman base in the region, the newly founded city of
Petra. Despite an initial setback here, Khusro took the city, but did not destroy it. The Roman
defenders, having negotiated a truce with the king, joined his army (II
.17.1–28). The Roman
garrisons at Pityus and Sebastopolis also withdrew after destroying their bases. (VIII.4.4–5).
See Braund 1994: 292. When news reached the Persians of Belisarius’ capture of Sisauranon
and Harith’s attack on Assyria, Khusro installed a garrison at Petra and withdrew to his own
lands (II.19.47–9). See Bury 1923: II, 101–2, Stein 1949: 493–4, Braund 1994: 294–6. Else-
where Procopius sheds further light on the interrelationship between the campaigns of 54
Proc. Anecd. 2.29–31 (18.7–19.2): It happened also that Khusro had sent an army of
Huns against the Armenians subject to the Romans, in order that by reason of their
preoccupation with this (force) the Romans there might not notice what was going on
in Lazica. (30) Other messengers brought word that these barbarians (the Huns) had
come to battle with Valerian and the Romans, who had come forth (to meet them),
and had been heavily defeated by them, and as a consequence most of them had per -
ished. (31) When the Persians heard these things and, partly because of the miseries
which they had suffered in Lazica, and partly because they feared that they might
during the withdrawal chance upon some hostile forces among the cliffs and the
regions overgrown with thickets and all be destroyed in the confusion, they became
exceedingly anxious for the safety of their wives and children and native land. Then
whatever (section) of the Medic army that was fit began to heap abuses upon Khusro,
charging him with having violated his oaths and the common laws of all men by
invading Roman territory, (territory) to which he had no claim. (And they said that)

he was wronging a state which was ancient and worthy, above (all states), of the high-
est honour, (p.19) one which he could not possibly overcome in war; and they were on
the point of revolution.
4(tr. Dewing, revised)
Jac. Ede. Chr.p.321: Khusro again went and took Petra, a city in Lazica, and placed
an army there.(…)Andthenceforward it happened that the Romans fought against
it for a time; and seven years they overcame the Persians and took it f them>.
5(tr. M. Greatrex)
Return of the Armenians to allegiance to Rome (542)
Khusro’s second invasion of Euphratesia accomplished little, and was soon repelled by
Belisarius. Already those who had urged the Persians to begin the war we
re changing sides. 6
Proc. WarsII.21.34 (249.1–3): And the Armenians who had gone over to Khusro
received pledges from the Romans and came with Vasak to Byzantium. ( … ) 7(tr.
Dewing, revised)
The Roman invasion of Persarmenia (late summer 542)
During his invasion of Mesopotamia in 542, Khusro had entered into negotiations with
Justinian and had withdrawn after concluding a truce (see Chapter 9). Later the same year,
Khusro awaited the arrival of Roman emissaries in Atropatene,
8but they were delayed en
route. The Persian general in Armenia, Nabedes, therefore sent the Armenian patriarch at
Dvin to the magister militum per Armeniam Valerian, to hasten the progress of the envoys. But
the patriarch’s brother, who had accompanied him, informed Valerian that Khusro’s rule was
under threat; and when news of this reached Justinian, he immediately ordered all the
commanders in the East to unite and invade Persian territory (Proc. II.24.1–11). Khusro and
his army had meanwhile withdrawn to Assyria, worried by the arrival of the plague.
Numerous Roman commanders along the frontier readied their forces and set off into Persian
territory in haphazard fashion; 30,000 men are said by Procopius to have taken part in the
invasion. Only once the Romans were in Persian territory did the various leaders converge and
make for the chief city of Persarmenia, Dvin (II.24.12–21).
10Nabedes, whose force numbered
only 4000, therefore took refuge at a mountain fortress at Anglon and prepared an ambush for
the Romans there. The Roman army, attacking the Persian position in haste and poor order,
was forced back with heavy losses. The Persian victory was complete (II.25.1–35). See Stein
1949: 499–500.
The war resumes in Lazica, 547–9
The Lazi did not prosper under Persian domination. Their zealous Christianity was at odds
with their masters’ Zoroastrianism, while they also found themselves cut off from trading in the
Black Sea. Khusro, aware of the rising discontent in Lazica, decided to take measures to ensure
that his grip on the region was secure (Proc. II.28.15, 25–9). See Braund 1994: 296–7.

Proc.WarsII.28.18–24 (285.2–286.2): For it seemed to him that it would be a lucky
stroke and a really important achievement to win for himself the land of Colchis and
to have it in secure possession, reasoning that this would be advantageous to the Per -
sian empire in many ways. (19) In the first place they would have Iberia in security
thereafter, since the Iberians would not have anyone with whom, if they revolted, they
might be safe;
12(20) for since the most notable men of these barbarians together with
the king, Gourgenes, had looked towards revolt, as I have stated in the preceding
pages, the Persians from that time did not permit them to appoint a king for them -
selves, nor were the Iberians autonomous subjects of the Persians, but they (the two
peoples) bore much suspicion and distrust towards one another.
13(21) And it was evi -
dent that the Iberians were most thoroughly dissatisfied and that they would attempt
a revolution a little later, if they could only seize upon some favourable opportunity.
(22) Furthermore, the Persian empire would be forever free from plunder by the Huns
who lived next to Lazica, and he would send them against the Roman domains more
easily and readily, whenever he should so desire. (For he considered that), as regards
the barbarians dwelling in the Caucasus, Lazica was nothing else than a bulwark
against them. (23) But most of all he hoped that the subjugation of Lazica would
afford this advantage to the Persians, that starting from there they might overrun with
no trouble both by land and by sea the countries along the Euxine Sea, as it is called,
and thus win over the Cappadocians and the Galatians and Bithynians who adjoin
them, and capture Byzantium by a sudden assault with no one opposing them. 15(24)
For these reasons, (p.286) then, Khusro was anxious to gain possession of Lazica, but
in the Lazi he had not the least confidence. (tr. Dewing, revised)
Khusro’s solution, according to Procopius, was to assassinate Gubazes, the Lazic king, and to
deport all the Lazi. He is also said to have made preparations to build a navy on the Black Sea.
Neither venture was successful, and the Lazi switched their allegiance to the Roman camp once
again (Proc. II.29.1–9).
16Accordingly, in 548 Justinian despatched an army of 8000 men to
Lazica under the magister militum per Armeniam Dagisthaeus; in conjunction with Gubazes,
these forces laid siege to Petra, but were unable to take it (II.29.10–12). See Stein 1949: 505
n.2. In response, Khusro sent Mihr-Mihroe to relieve the garrison (II.29.13). While
Dagisthaeus prosecuted the siege of Petra, sending only 100 soldiers to defend the borders of
Lazica to the south, Gubazes sought to block the Persian advance further north. By offering the
Alans and the Sabir Huns three centenaria, he secured an agreement from them that they would
attack the Persians in strength in Iberia; Justinian, however, delayed in sending the money
(II.29.27–32). Meanwhile, Dagisthaeus proved unable to capture the city, despite the desperate
straits in which the garrison found itself (II.29.33–43). Mihr-Mihroe’s relief force, having with
difficulty pushed back the Roman force guarding the pass leading into Lazica, at last reached
Petra, which had continued to withstand Dagisthaeus’ efforts to storm it; the magister militum
quickly withdrew before the Persians arrived (II.30.1–14).
17Mihr-Mihroe restored the battered
defences of Petra and installed a garrison of 3000; but he and his forces were then ambushed
nearby by Dagisthaeus and some Lazi (II.30.15–22). Unable to sustain his army, over 30,000
strong, with adequate supplies, he withdrew with most of his forces to Persian territory, leaving
5000 men to ensure supplies to Petra (II.30.30–3). Gubazes remained on guard by the river
Phasis, protecting Lazica; and Justinian now at last paid the money promised to the Sabirs, as

well as providing Gubazes with a handsome reward for his defection. In addition, he sent
another army to the region, commanded by Rhecithancus (II.30.23–9). The 5000 Persians left
near Petra soon fell victim (probably in spring 549) to the combined forces of Gubazes and
Dagisthaeus, numbering 14,000; they were heavily defeated and their supplies captured. The
victors then penetrated into Iberia, leaving a substantial garrison of Lazi to guard the pass
leading into Lazica before returning westwards (II.30.34–48). See Stein 1949: 505, Braund
1994: 298–9 on this campaign (which was over by 1 April 549, Proc. I
The campaign of summer 549 19
Khusro’s response to the mixed Persian fortunes of the previous campaign was to send a large
army under the command of Khorianes, supplemented by Alan forces, to invade Lazica. They
penetrated as far as Mocheresis and there encamped in the region of the river Hippis (Proc.
20Dagisthaeus and Gubazes marched against the invaders, and prepared to attack
them; the Lazic and Roman cavalry were now operating separately. Initial cavalry skirmishes
proved indecisive (VIII.8.1–20). The arrival of the Roman and Lazic infantry turned the battle
to their advantage; the Roman cavalry dismounted, and the Persian cavalry was unable to break
the Roman formation. Once the Persian general Khorianes was killed in the engagement, the
Persians broke and fled. The Roman victory was complete (VIII.8.21–38). But in the mean -
time another Persian army had succeeded in delivering supplies to the garrison at Petra
The campaign of 550
The Lazi, unhappy with Dagisthaeus’ failure to retake Petra, accused him to Justinian of siding
with the Persians. He was recalled and replaced by the veteran commander Bessas; further rein-
forcements were sent at the same time (Proc. VIII.9.1–5). Bessas’ first task was to put down a
rebellion by the Abasgi, who had gone over to the Persians, unhappy with their treatment at the
hands of the Romans. A Persian army under Nabedes had even succeeded in reaching Abasgia
and there obtaining hostages (VIII.9.6–14). John Guzes and the Herul commander Uligagus
were therefore charged by Bessas with the mission of subduing the rebels. They sailed along the
Black Sea coast and took possession of the Abasgian fortress at Tracheia, which they razed to the
ground (VIII.9.15–30).
21The Persians also succeeded in gaining control of an important fort
among the Apsili, the neighbours of the Abasgi, thanks to the treachery of a Lazic noble named
Terdetes. The Persian garrison was soon ejected by the Apsili, who were unwilling then to
return to their allegiance to the Lazi. John Guzes, however, was able to effect a rapprochement
between the Apsili and the Lazi (VIII.10.1–7).
The Persian campaign of 551
Since the five-year truce had now expired, the two powers were once again formally at war.
While negotiations proceeded in Constantinople (see Chapter 9 below), Bessas attempted to
wrest Petra from Persian hands. He vigorously prosecuted the siege, assisted by some Sabir
allies; his forces amounted to 6000 men, matched against 2600 inside the city. After a bitter
struggle, the city fell, although a few Persians sought refuge in the acropolis (Proc.

VIII.11.1–64). When they refused to surrender, the acropolis was burnt. The Romans captured
large quantities of supplies and equipment, a testament to Khusro’s desire to maintain control
of the city (VIII.12.1–20). Bessas then razed the city’s walls to the ground, an action
commended by the emperor (VIII.12.28–9).
Mihr-Mihroe, meanwhile, who had been marching south of the Phasis to relieve Petra, turned
instead to cross the river close to the border with Iberia, seizing the forts of Sarapanis and Scanda.
With an army of Persian cavalry and 4000 Sabir Huns, he made for Archaeopolis, to the north of
the Phasis. The Romans had 3000 men stationed in the city, while a further 9000 were encamped at
the mouth of the river under Venilus, Uligagus and Varazes. Mihr-Mihroe passed by the fortress,
aiming to deal first with the larger force, but was thwarted when it withdrew across the river. He
therefore returned to lay siege to Archaeopolis (Proc. VIII.13.1–30).
23Despite the useful support of
Dilimnite troops and elephants, Mihr-Mihroe’s attack was driven back with heavy losses; many of
his forces died from lack of supplies rather than in the actual fighting (VIII.14.1–44).
24He therefore
proceeded to the more fertile region of Mocherisis, to the west, and set about rebuilding the fortress
which had earlier existed at Cotaeum; at the same time he invested the nearby Lazic fortress,
Uthimereos, which was guarded by a joint garrison of Romans and Lazi. Thus he was able not only
to blockade the Roman stronghold, but also to assure himself of adequate supplies and to bar the
Romans access to Suania, a strategic kingdom to the north of Lazica (VIII.14.45–54). See Braund
1994: 305 (noting, however, that other routes to Suania existed). Despite the peace negotiations taking place in Constantinople, both powers continued to
manoeuvre to secure the advantage in Transcaucasia. The Persians initially gained the upper
hand, seizing Uthimereos through the defection of a Lazic noble; they were now masters of
eastern Lazica, Suania and Scymnia (Proc. VIII.16.1–15).
25Further details on the Persian
annexation of Suania are provided by Menander, who is here reporting the words of a Roman
envoy, Peter the Patrician. Since he was seeking to prove that Suania had always been a depend-
ency of Lazica, and hence should be returned to the Romans along with Lazica, his words must
be treated with caution.
Menander frg.6.1.249–67: Actually, when Tzath was chief of the Suani, a certain
Deitatus was commander of the Roman troops there, and there were other Romans
also living among the Suani. When some ill-feeling between the king of the
Lazi and Martin, at that time the general of the Romans there, on account of this the
Colchian did not send to the Suani the usual supply of grain (for (it was) the custom
that grain be sent by the king of the Colchians). The Suani, therefore, angry that they
had been deprived of the customary items, made known to the Persians that if they
came there, they would hand over Suania to them. Meanwhile, they told Deitatus and
the other Roman commanders that ‘a large mass of Persians is reported to be advanc -
ing against the Suani, against which we do not have a battleworthy force. Your best
course of action is to retreat before the Median army with the Roman units here.’ By
means of so great a deception and by gifts the Suani convinced the army commanders
and rid themselves of the Roman garrison, and the Persians quickly arrived and took
over Suania. (tr. Blockley, revised)
Menander also reports the counter-claims of Khusro.

Menander frg.6.1.495–507:I (Khusro) subdued Lazica, but did not make an attack on
the Suani. I only heard (of them) when Mihr-Mihroe reported to us that it (Suania) was
a country not worthy of consideration, nor even worth fighting over, nor a (suitable)
object for a royal expedition, but (that) it was (inhabited by) one of the peoples around
the Caucasus, and that they have a petty king, and that the land is a thoroughfare for the
Scythians. Mihr-Mihroe died, and the nakhveragansucceeded to the command. He too
then wrote about them in no different fashion, (reporting) that they lived on the ridges
of the Caucasus, that they were actually thieves and plunderers and perpetrators of cruel
and impious deeds. I decided, therefore, to send an army against them, when they in
fear became Persians instead of Suani. As a result the land obviously belonged to me
from that time, and I do not refuse to keep it. (tr. Blockley, revised)
In winter 551–552 Mihr-Mihroe consolidated his grip on eastern Lazica, even threatening to
move against the Roman forces assembling at the mouth of the Phasis. They dispersed upon
hearing of his plan; the Romans perhaps withdrew to the south, while the Lazi took refuge in
the mountains. A diplomatic overture by Mihr-Mihroe to Gubazes met with no success (Proc.
VIII.16.16–33). See Braund 1994: 305–6.
The Persian campaign of 552
According to Procopius, Khusro used the money paid to him by Justinian by the terms of the
truce of 551 to hire a large force of Sabirs, which was to take part in a further campaign by Mihr-
Mihroe aimed at further entrenching the Persian position in Lazica. But the Persians had
arrived at the limits of their expansion: while the main Roman force, now under Martin,
remained in place at the mouth of the river Phasis, the garrisons at an unnamed fortress, Tzibile
and Archaeopolis were able to withstand all Mihr-Mihroe’s attacks (P
roc. VIII.17.11–19).
The Persian campaign of 554
There is no record of any significant military activity in 553, although an Armenian source, the
martyrdom of Yazdbozid, notes the arrival in Persarmenia of three high-ranking Persian nobles,
sent by Khusro to investigate complaints concerning the behaviour of Persian officials in the
region; evidently the Persians were determined to maintain their grip on Persarmenia, and were
prepared to conciliate the Armenians if necessary.
27Further Roman troops arrived in the
Transcaucasus, some of whom will have been transferred from the now quieter western theatre
of war. They were under the joint command of Bouzes, Martin and Bessas and accompanied by
Justin, the son of Germanus. See Stein 1949: 511 and Braund 1994: 306 with Agath. II.18.8.
In 554 Mihr-Mihroe decided to try to penetrate further into Lazica, although aware of the
strength of the Roman defences; their most advanced position lay at Telephis, where Martin
guarded a narrow pass (Agath. II.19.1–4). When Mihr-Mihroe captured it by a ruse, the
Roman forces stationed behind it withdrew westwards along the Phasis as far as Nesos, a forti -
fied island at the confluence of the Doconus and the Phasis (Agath. II.19.5–21.11). Mihr-
Mihroe did not follow up his victory, aware of the difficulties in supplying his forces so far west,
and, having reinforced a Persian fort at Onoguris, near Archaeopolis, returned to Mocheresis;
he died in Iberia from an illness the following summer (Agath. II.22.1–

Events of 555–6
Both sides then underwent a change of command. Mihr-Mihroe was succeeded by a general
whose name is given by Agathias as Nakhoragan, which probably in fact represents the Persian
titlenakhveragan. See Stein 1949: 514 n.2 and Christensen 1944: 21 and n.3. Following a
complaint from Gubazes about the incompetence of the Roman commanders, Bessas was recalled
and sent into exile; Martin and Rusticus, the two other individuals criticised by Gubazes,
remained in post, and formed a plot to rid themselves of the king (Agath. III.2.1–11).
Gubazes had been killed, in September or October 555, 30the Lazi refused to cooperate with the
Romans any further (Agath. III.3.1–4.8). Martin nonetheless undertook an assault against the
Persian fortress at Onoguris, while detaching a small force to ward off a Persian army said to be
coming from Mocheresis and Cotaeum to relieve the garrison. It proved unequal to the task, and
the besiegers were forced back in a disorderly retreat; the Roman camp outside Archaeopolis was
then plundered by the pursuing Persians (Agath. III.5.6–8.1). Although the loyalty of the Lazi
was wavering, they preferred to inform Justinian of what had happened and to demand justice,
rather than to defect to the Persians once again. Justinian accordingly despatched the senator
Athanasius to resolve the situation (Agath. III.8.4–14.3).
31See Stein 1949: 513–14 and Braund
1994: 308–9.
The campaigning of 556
The nakhveragan was in Mocheresis in spring 556. Gubazes’ younger brother and successor,
Tzath, arrived in Lazica from Constantinople at the same time, having been invested with the
regalia of office by Justinian. He was accompanied by the magister militumSoterichus (Agath.
Agath. III.15.6 (103.31–4): The general Soterichus set out immediately on the jour-
ney on which he had been despatched. He was bearing gold from the emperor for dis-
tribution among the neighbouring barbarians according to the treaty of alliance, since
this had been the custom for a very long time and occurred annually.
In this instance, however, Soterichus encountered hostility among the Misimians, a people
subject to the Lazi, according to Agathias; and having killed him, they felt they had little option
but to embrace the Persian cause (Agath. III.15.7–17.2). Meanwhile, the nakhveragan
advanced down the Phasis, towards Nesos, with (according to Agathias) 60,000 men
including Dilimnite allies.
33The Romans, on the other hand, enjoyed the support of a contin -
gent of 2000 Sabirs, serving as heavy infantry (Agath. III.17.3–5). An attempted ambush of the
Sabirs by the Dilimnites proved disastrous, and the Persian commander sought, without
success, to come to terms with Martin (Agath. III.17.6–19.7). The nakhveraganthen turned
against Phasis, which, he had learnt, was vulnerable to attack.
34He crossed the river, bypassing
the forces at Nesos, and marched on Phasis; most of the Roman forces, however, reached the
town before him and made preparations for a siege. There, a morale-boosting trick by Martin
routed the besiegers, who hastily fled eastwards (Agath. III.19.8–28.5). The nakhveragan
himself withdrew as far as Iberia, leaving a subordinate in command of the forces remaining in
Mocherisis (Agath. III.28.5–10). See Stein 1949: 514, Braund 1994:

While the formal trial, condemnation and execution of Rusticus and his brother John were
taking place, the Misimians implored the aid of thenakhveragan(Agath. IV.1.1–12.7). 35
The Romans counter-strike, summer 556
Thenakhveragan’s retreat gave the Romans the opportunity to recover lost ground. While
Bouzes and Justin remained on guard at Nesos, a Roman force of 4000, including 2000 Tzani,
set off to recover Misimia. Under two low-ranking leaders, it advanced no further than Apsilia,
while the Persians moved on Misimia. A stand-off ensued, punctuated only by a minor Roman
success against some Sabirs allied to the Persians (Agath. IV.13.1–14.5). A Roman cavalry
force, operating from Nesos, regained control of Rhodopolis without a struggle (Agath.
IV.15.1–3). As winter drew on, the Persians withdrew from Misimia, and Martin arrived at the
border fortress of Tzibile to take charge of the Roman force in Apsilia. Although he fell ill, his
subordinates proceeded against the Misimians. They successfully overran Misimia and laid
siege to Tzakher, the stronghold where the Misimians had assembled. When John Dacnas, sent
by Martin, arrived to take charge of the siege, the Romans succeeded in inflicting such damage
on it and the surrounding the region that the Misimians sued for peace. John agreed to their
petition and recovered the money which had been in Soterichus’ possession, which amounted
to 28,800 solidi(Agath. IV.15.4–20.10).
36See Stein 1949: 515 and Braund 1994: 310. While
Martin was then replaced as magister militum per Armeniam by Germanus’ son Justin, the
nakhveragan was cruelly executed for his failures. (Agath. IV.21.1–23.3).
Truce agreed (autumn 557)
See Chapter 9, below.
A Tzanic revolt subdued (558)
The Tzani, having returned to their previous habits of predatory raids on the surrounding prov -
inces, were with difficulty brought to heel by Theodore, a Tzan who had remained loyal to the
empire. They were henceforth obliged to pay an annual tax, a burden from which they had hith -
erto been exempt (Agath. V.1.1–2.3). See Stein 1949: 516 and n.2.
The dispute over Suania, 562
The peace negotiations which brought the war to a close are dealt with elsewhere (below,
Chapter 9). A particular point which exercised the negotiators was control of the kingdom of
Suania, which had at one stage been subject to the Lazi. In the end, when the treaty was agreed,
the question of Suania left open.

Diplomatic relations (545–62)
The continuing war between the Ghassanids and the Lakhmids (545–50/1)
Despite the truce of 545, the Arab allies of both sides remained at war. Nothing more is known
of the progress of their struggle in this period than what is recorded by Procopius. 1See Shahîd
1995: 237–9.
Proc. WarsII.28.12–14 (284.1–13): And a little later Harith and Mundhir, the rulers
of the Saracens, continued the war against each other by themselves, unaided either by
the Romans or the Persians. (13) And Mundhir captured one of the sons of Harith in a
sudden raid while he was pasturing horses, and straightaway sacrificed him to Aphro-
2and from this it was known that Harith was not squandering Roman chances (in
favour of) the Persians. (14) Later they both came together in battle with their whole
armies, and the forces of Harith were overwhelmingly victorious, and turning their
enemy to flight, they killed many (of them). And Harith came within a little of
capturing alive two of the sons of Mundhir; however, he did not actually capture
(them). Such then was the course of events among the Saracens. (tr. Dewing, revised)
The Persian attempt to seize Dara (547)
Despite the armistice in force since 545, Khusro remained eager to seize control of Dara, and
sought to do so by a ruse. He sent Yazdgushnasp (Isdigousnas) as an ambassador to Constanti -
nople; en route he would stay at Dara and, with the assistance of his large retinue, take posses -
sion of the city in conjunction with forces based at Nisibis. The plan was thwarted by George, a
former adviser of Belisarius, who insisted that Yazdgushnasp would be allowed in the city with
only twenty followers. He then continued to Constantinople, where he was cordially received
by Justinian, and received lavish gifts from the emperor (Proc. II.31–
Justinian seeks an end to hostilities (551)
After the five-year truce had ended, Justinian sent Peter the Patrician to the Persian court to
obtain a full-scale settlement. Khusro, who had just had to suppress a revolt by his eldest son

Anoshazadh (Anasozadus), 4refused to enter negotiations immediately, but sent Yazdgushnasp to
Constantinople soon afterwards. As in 547, he brought with him a large retinue, including several
relatives, and, to the annoyance of Constantinopolitans, was treated with great honour. But
Yazdgushnasp proceeded to accuse the Romans of violating the truce, on account of Harith’s recent
attack on Mundhir (Proc. VIII.11.1–10). See Shahîd 1995: 239–40. The negotiations continued,
however, and a limited armistice was agreed in late 551; in Lazica alone would hostilities continue.
Proc. WarsVIII.15.1–7 (566.5–567.5): In Byzantium, meanwhile, Khusro’s envoy
Yazdgushnasp, in conferring with the Emperor Justinian regarding the peace, wasted
a vast amount of time. (2) And after disputing many (points) thoroughly, they finally
reached an agreement by which there would be a truce in the realm of each sovereign
for five years, while envoys passed back and forth from each country to the other, fear -
lessly carrying on negotiations for peace during the period in order to settle their
differences regarding both Lazica and the Saracens. (3) It was further agreed that the
Persians receive from the Romans for this five-year truce twenty centenariaof gold,
and for (the) eighteen months which had elapsed between (the expiration of) the
former truce and this one – while they were sending embassies to one another – six
centenaria more. (4) For the Persians declared that (only) on this understanding had
they agreed that negotiations for the truce proceed. (5) Yazdgushnasp further
demanded that he should receive these twenty centenariaon the spot, but the emperor
wished to give four each year, for the reason that he might have surety that Khusro
would not violate the agreement. (p.567) (6) Later, however, the Romans gave the
Persians outright the entire amount of gold agreed upon, in order not to appear to be
paying them tribute each year. (7) For it is disgraceful names, and not deeds, which
men are wont as a general thing to be ashamed of. (tr. Dewing, revised)
A detailed account of a mission of Yazdgushnasp, perhaps this one, is provided by Peter the
Patrician, in an account embedded in a work compiled by Constantine VII Porphyrogenitus. It
is clear that an elaborate diplomatic protocol had emerged by this point; and, as others have
noted, the measures taken to protect Dara may have been inspired by the recent attempt of the
Persians to seize control of it under the guise of an embassy.
De Cer. 89–90 (398–410): (What should be closely observed upon the arrival of a great
ambassador of the Persians.) When a great ambassador is announced, the magistershould
send to the frontier one illustrismagistrate, or a silentiarius, or a tribune, or indeed one of
the nobles or magistriani, or whomever he sees fit to send for the honour of the
personage coming, so that he may receive and guide him safely from one place to
another. He who is sent comes (p.399) to Nisibis, and greets him, and, if he has a letter
from the emperor, he hands (it) over; if not, then (he hands over one) from the magister,
summoning him. Equally, the magistermay not write either, but the invitation is issued
only by mandata, (bidding him) come with good spirits and a retinue; and he (the
Roman envoy) goes out with him (the ambassador). The commanders of Dara should
meet him with (their) soldiers at the frontier, and receive the ambassador and his men.
And if there is anything that must be discussed at the frontier, it is discussed, while the
ruler of Nisibis has come with him (the Persian ambassador) as far as the frontier with a

force of Persians. If nothing is discussed, in this case too it is absolutely necessary for him
to come with a force, and while the Romans receive him and those with him, the rest of
the Persians must stay in Persian territory and he alone with his followers is to come to
Dara and be entertained. It is the duty of the commanders of Dara to display much
alertness and foresight, so that a force of Persians does not come along too, by a pretext
of the ambassador, and closely follow after (him) and gain control of the city by
cunning. The magistrates must pay great attention to this contingent (p.400) and be
discreetly vigilant and guard against this plan. Theducici, according to custom, pay the
cost of the journey here (Constantinople) for 103 days. This many (days) have been
determined from the start as sufficient for an ambassador going up (to Constantinople),
and as many for one going down. Sometimes he is slow on the road, and the emperor
issues a command, and an extra payment is made to him. The record of expenses for
him is conserved in the scrinium(office) of the barbarians.
Five post-horses and thirty mules are assigned to him, according to the agreements
(reached) when Constantine was praetorian prefect. The emperor, should he wish to
entertain him (the ambassador), orders a lot more to be given to him. And should he
wish to honour him as well, he should send for and receive him through men of good
repute in Galatia and Cappadocia and provide food (for him); similarly he should
send (a message) to Nicaea to look after him and entertain (him). The magistertoo,
when (the ambassador) reaches Antioch,
8should send a magistrianus, charged with
greeting and welcoming him, and to learn how he is being guided from place to place.
If the emperor wishes, he does this once, and then for a second time, writing (to him)
and greeting (him), and asking how he is being looked after. Both beasts and dromons
should be made ready for him at Helenopolis, so that, if he wishes (p.401) he may
depart on foot for Nicomedia, or, if he wishes, he may cross over in the dromons.Itis
vital that horses and beasts be made ready in Dakibyza so that they may receive him
and convey him to Chalcedon. In Chalcedon the magistershould prepare lodgings
both for him and his men; he should also send the optioof the barbarians and entrust
to him
9ready cash for expenses incurred in the day or days in Chalcedon. And he
sends gifts to him. It is the duty of the magisterto send (a message) to him to greet him
and to ask him how his journey was, whether he is weary or not, and simply to enter -
tain him in suitable fashion. His (the ambassador’s) lodging in Constantinople must be prepared in advance in
accordance with the man’s rank and the escort he brings with him; and there must be
made ready in it beds, bedclothes, ovens, fireplaces, tables, and buckets to carry water
and help with other hygienic services.
10But the comes rei privatae, that is the sacellariusof
the emperor – for now this duty has been transferred to him – provides the mattresses in
accordance with a token of the magister. Thepraefectus urbi (city prefect) provides the
beds and small cans and tables (p.402) and ovens and pots, again in accordance with a
tablet of the magister. The men of the arsenals provide the braziers. Workmen are also
allotted to him by the prefect from warehouses. And the bath of that house where he is
to stay must be made ready, or one near him, so that he himself and those with him may
bathe when they want; and for them alone is the bath available. Whenever he comes ashore, the magistershould send imperial horses (to him); the

spatharius(bodyguard) of the emperor provides them. They receive him off the
dromons and they carry him off to his house. And immediately the magistersends (a
message) to him and greets him and again asks him how he was looked after. He also
sends small gifts – the things he wishes – through the optio; generally the ambassador
too sends (a message) to the magisterand greets him in return. And the magistershould
receive kindly him who comes and should give a fitting reply to the greeting. The
magister through his own man informs him (the ambassador), ‘Refresh yourself, and
when you see fit, I (shall) greet you.’ And that man on the next day or the one after
makes an announcement in advance and comes and greets the magister, and the
magister receives him in person and (p.403) asks him before anything (else) about the
health of his king, and then about the children of his king, and about the magistrates
and about his own health and that of his household. (He also asks) how his journey
was, whether anything was amiss en route, if anything was passed over and says ‘We
have been charged by our pious master to do everything for your entertainment. If
therefore anything was remiss, that is our failure. We invite you not to be aggrieved or
silent, but to tell us, so that correction may be made.’ And all those with the ambas -
sador too do obeisance to the magister, throwing themselves to the floor.
And when the emperor decides to receive him, the magistersends for him and
informs him ‘The master has bid you come to (him).’ The master (emperor), when he
(the ambassador) is in Chalcedon, and when he has come here, should send a decurion
to greet him and to enquire concerning the health of the king, and how he himself was
looked after. When he (the emperor) receives him, the magistersends asubadiuva 12in
the evening and informs him that ‘The emperor has bid you come forward, and (so)
come forward.’ And the emperor similarly sends a decurion and greets him and
informs him that (p.404) ‘On the morrow we (shall) receive you, and (so) come
forward.’ Mandata
13are given in the evening for a silentium, 14and it receives the
ambassador of the Persians. The admissionalis 15should go and ask themagisterabout
the labaresioi (standard-bearers), and mandataare given, so that they too come and
meet (him), and they must stand in their positions with the chariot. All the magis -
trates who have worn silk garments go forward, and the ambassador enters through
the Regia (Royal Door), and the magisterreceives him in his schola
16and asks him
whether he has the gifts of the king; he must see everything before they go in and
receive a record of them. The magistergoes to the emperor and reports to him
concerning the gifts, and hands over to him the record. The ambassador waits in the
schola of the magister. Then after that the magistrates are received and advance to the
consistorium; the admissionalisand thechartularii of the barbarians and the inter -
preters must conduct the ambassador and must seat him in the anteconsistorium;
chartularii and interpreters must give a decree (citatorium)ofthemagister to the
admissionales, and the other (proceedings) take place as in a silentium.
The magister should (p.405) prepare armed candidati
18and handsome boys to
follow them. And the emperor comes out of the cubiculum, preceded by the patri-
cian, and takes his seat in the great consistorium; the magistrates come in, naturally
with dark brown (cloaks) according to custom.
19At this point the admissionalis
should lead the ambassador and bid him stand at the wall opposite the curtains of

the great summerconsistorium. The three doors of the consistoriumare opened, if he
(the ambassador) has horses among the gifts; and three all-silk curtains are hung.
After the magistrates are received, the magisterpronounces (as an example), ‘Let
Yezdekes, the ambassador of Khusro, the king of the Persians, and those coming
with him, be called.’ And he brings forward the guards.
20The (chartularii)ofthe
barbarians 21should give this pronouncement to the admissionales, as has been said;
and the admissionales make two tablets, one inscribed with large letters, and (this)
they give to the silentiarius, and he in turn to the ostiarius.
22And it is read to the
emperor by the chartulariusin thecubiculum. 23The other tablet they give to the
magister for his acknowledgement. The tertiocerius24receives a copy of the decree,
stands behind the magister, and informs him (of it). Therefore, according to the
pronouncement of (p.406) the magister, the decurion enters the small consistorium
and takes (with him) the armed candidati; and he moves them on, placing them on
the right and left in front of the magistrates, with the consulars. At that point he goes
out, and if he observes that the ambassador is ready, the decurion shouts, ‘Leva’.
When the curtain is raised, the ambassador outside throws himself to the floor,
where (there is) purple marble, does obeisance, and arises. And then he enters the
gate, (and) again he throws himself (down) and does obeisance on the floor and
arises. Again in the middle of the consistoriumhe does obeisance likewise, and then
he comes and adores the feet (of the emperor) and stands in the middle; he hands
over the letter and declares the greeting of his king. The emperor should then ask,
‘How fares our brother in God? We rejoice in his good health.’ And he says to the
ambassador any other words he wishes that occur to him. After this, the ambassador
says ‘Your brother has sent you gifts and I invite you to receive them.’ The emperor
permits this. The ambassador goes out, and, with his men, carries the gifts; and he
enters (again), himself carrying either a mantle or an ornament or whatever, if it is
valuable, and each of the other men carries one sample. They (the men) should
(p.407) be prepared by the interpreters in the anteconsistorium, and all go in carrying
(something). All of them stand at the wall opposite the chair outside the curtain; when
the curtain is raised, they throw themselves to the floor, and again go through the
door, and throw themselves down again, and do this for a third time. Then the
silentiarii accept all the gifts, and they have the task of conveying them to the vestosacra
according to the record of themagister, and of handing them over; and the evaluation
of the gifts takes place. The vestosacranimust immediately take the evaluation of the
gifts to the magister, so that he may know what has been brought, and, at the time of
the exchange of gifts, he may remind the emperor what he should send in return
through his own ambassadors. When therefore the gifts have been offered, the
emperor says to the ambassador, ‘Refresh yourself for a few days, and if we have some -
thing to discuss, we (shall) discuss it; and with good manners we (shall) dismiss you to
our brother.’ The ambassador thanks (him), and does obeisance, and again does obei -
sance at the same places, and withdraws. When the curtain has been released, the
decurion stands up, (and) the magisterpronounces (the command) ‘Transfer’,
the decurion takes the armed candidati, and moves (them) to the small consistorium.
And (p.408) then the emperor rises and the rest (of the proceedings) take place as

usual. The ambassador must remain below at thescholaof the magister, and the
magister must go down and bid him farewell and dismiss him.
(90): (What should be observed with regard to the ambassador during the other days.)
The emperor, once he has read the letter, permits, when he wishes, the magisterto
inform the ambassador that he may come to the palace on the following day. And he
(the magister), if he wishes, informs him
28through a silentiarius that he may come. A
silentium is held, the band of troops is held still, and the labaresioistand, and when he
comes forward, the magisterreceives him in his schola; and he leaves him seated and
goes up and reports to the emperor. He receives him (the ambassador) within, either
in the portico or in the Augusteus itself.
29If the ambassador has gifts of his own, he
asks the day before through the magisterthat they may be received; and if the emperor
allows, he (the ambassador) shows them to the magisterin theschola, and a record is
made for him. And the magistershould go first to the emperor and show the record of
the gifts to him. The ambassador, if he wishes his gifts to be received, goes in and asks
the emperor to receive them; and if the emperor allows, his (the ambassador’s) men go
in, carrying his gifts, following the same plan as for the royal gifts; and a conversation
is held. The emperor should again continuously recall the Persian king in pleasant
fashion, as well as (asking about) his health; and if there is peace, they talk of such
things, and the emperor dismisses him, and he awaits the magisteroutside. (Then) the
magister comes out and takes his leave of him, and also himself dismisses him.
On the other days, he (the emperor) summons him, and they discuss matters. And
if he sees fit, he allows the magisteror even other magistrates with him to hold discus-
sions with the ambassador outside. If there is complete peace between the states, the
emperor should send (messengers) to him to visit him continuously and to find out
how he is; (p.410) and he should send to him portions (of food) and gifts both on our
feast days and on his notable ones; and in every way he should look afte
r him.
We may take this opportunity to note also the recommendations in the treatise Peri Strat¢gias,
which is assigned by many (though not all) scholars to the sixth century ( Peri Strat¢gias43.1–13
[= Peri Presbeias]). See Lee and Shepard 1991: 32 (who assign the work to the tenth, rather than
the sixth, century). It stresses that while foreign embassies in Constantinople should be well
treated, they should also be kept under close watch, especially if they come from powerful states.
Procopius complains, however, that Yazdgushnasp, in addition to the generous gifts he received
from the emperor (see above), was also allowed to wander around the city freely (VIII.15.19–21).
See Tinnefeld 1993: 207–8.
The armistice, and in particular the payments to the Persians, were (according to Procopius)
highly unpopular in the imperial capital, partly because it allowed the Persians to consolidate
their grip on Lazica (VIII.15.13–15), and partly because it conceded regular payments to the
Persians, something which Roman emperors had always sought to avoid. This latter issue in
particular would continue to arise for some time to come.
Proc. WarsVIII.15.16–18 (568.9–569.4): They (the people of Constantinople in
551) (were also moved by the fact) that the very thing which the Persians had been
striving for from ancient times, but which had seemed impossible to achieve either

by war or by any other means – I mean, having the Romans subject to the payment
of tribute to them – this had been most firmly achieved at the present juncture in the
name of an armistice. (17) For Khusro imposed upon the Romans an annual tribute
of fourcentenaria, which he had clearly been aiming for from the start, and now
forty-six centenaria have been handed over for (a period of) eleven years and six
months on the pretext of the armistice, giving to the tribute the name of treaty,
although in the meantime he had, as stated, been carrying on a campaign of violence
and war in Lazica.
32(18) From this (plight) the Romans had not the least hope of
rescuing themselves in the future, (p.569) but they perceived that they had in no
hidden sense become tributary to the Persians. Thus were these things done. (tr.
Dewing, revised)
The Romans acquire the silk-worm (c.552)
Right from the start of his reign Justinian wished to free the Romans from the need to purchase
silk from the Persians, realising that it was an important source of revenue for the Sasanians; for
it was almost impossible for any silk to reach Roman territory without passing through their
hands. In the 540s, when it appears that the Persians increased the charge for raw silk, or
perhaps earlier, he had commissioned some monks, who had been to ‘Serinda’ (probably
Sogdiana), to bring back the eggs of the silk-worm to Roman territory; and this they accom-
plished in 552. Thus Justinian at last brought to an end the Persian monopoly in the provision
of the Roman empire with silk; but the dependence of the empire on imported silk continued
for some decades afterwards
33(Proc. VIII.17.1–8, Theoph. Byz., FHGIV.270, tr. Wilson
1994: no.64). See Stein 1949: 769–73 (773 n.2 for the Sogdian provenance of the silk-worms),
Pigulewskaja 1969: 158–9, Evans 1996a: 234–5.
The victory of Harith over Mundhir (June 554)
While in Mesopotamia and Armenia the armistice was untroubled, Mundhir evidently contin -
ued his raids deep into Roman territory, south of the Euphrates. The precise location of
Harith’s victory is uncertain. See Shahîd 1995: 240–4, Nölde
ke 1879: 170 n.1.
Mich. Syr. IX.33 (323–4a/269): In year 27 of Justinian, Mundhir (the son) of
Shaqiqa went up into the territory of the Romans and devastated many regions.
Harith (the son) of Jabala encountered him, fought against him, defeated and killed
him at the source (p.324) of the ‘Udaye in the region
34of Chalcis. The son of Harith,
called Jabala, died, killed in battle. His father buried him in a martyrionof this fort. 35
(tr. M. Greatrex)
According to the V. Sym. Styl. Iun., a Roman embassy went to Khusro in the early 550s to
resolve the ongoing conflict between Mundhir and Harith; the author claims, however, that the
attempt was foiled by the personal intervention of Mundhir, who pledged that he would immi -
nently launch another raid.

V. Sym. Styl. Iun.186–7 (164–6):After a short interval of time he (Mundhir) set up
camp and went up to the frontier (lands) of the Romans, exulting in (his) great army
and rulership. (p.165) And all the inhabitants of the East were greatly troubled. Then
Symeon, the servant of God, went into a trance and beheld a vision (which) he related
to us, saying, (ch. 187) ‘I saw myself today being snatched up by the Spirit and
arriving on some small hill, near the frontier (lands) of the Saracens, Persians and
Romans; and I was standing in the middle of an encampment
37of soldiers and Sara -
cens, 38where the phylarch Harith had his camp. And, lo, I saw on the opposite side
(that) a mass of many cavalrymen had come with the tyrant Mundhir (as numerous)
as the stars in the sky and the grains of sand on the shore; and fear descended on all
those with Harith and the sinews of their arms grew slack. And, lo, I saw the mêlée
begin and the formations clash with one another, and in (the battle) Mundhir had it
in his power to slay all of them; (but) the spirit of strength stood in my presence,
bearing a ball of fire, and, at my behest, he threw it at his (Mundhir’s) head and struck
him down. Be of good cheer therefore, children, for today the Lord has done a great
39to the entire East.’ When he had related this, we made a note among
ourselves of the day and the hour. Within one week a victory bulletin arrived in the
city of Antioch, announcing how, on that day and at that time at which the holy man
had related these things to us, the impious Mundhir had been struck down. And some
(men) from the Roman camp afterwards came to the holy man, who said that in that
battle which had taken place opposite the hill, about (p.166) which the holy man had
spoken publicly, they had been in dire straits and had called upon the servant of God;
and they knew well that that great service had been performed by the Lord through
him. And, in addition to this, they related all that had happened in the same battle.
Some of them even remained with the holy man until their death. We, along with all
those who had heard (Symeon’s vision), remembered (the words) prophetically
spoken by him and praised God for revealing to his holy servant events far off and
close at hand. Henceforth the entire East enjoyed peace and remained in great calm.
General armistice agreed (557)
As a result of the stalemate in the Transcaucasus (see Chapter 8), Khusro despatched his senior
ambassador Yazdgushnasp (here referred to as Zikh) to negotiate an arm
istice. 40
Agath. IV.30.7–10 (163.1–21): Khusro calculated that it was not possible for him to
deploy (his forces) against the Romans in the Colchian land (Lazica). (For) while they
had control of the sea and could easily send for all the things they needed from that
quarter, he was obliged to send even the smallest (quantities) of food to his troops by a
long and deserted road by means of porters and beasts of burden. Therefore, consider -
ing these things, he resolved to put aside the war, so that the peace which had been
limited for a long time to certain regions might not continue (to be) incomplete and
ineffective, but might be strengthened everywhere equally. (8) So he sent on an
embassy to Byzantium Zikh, (who was one) of the most distinguished men among

them (the Persians). (9) When this Zikh arrived and came into the presence of the
Emperor Justinian, he spoke much about the current situation and heard much
(about it). At length they came to an agreement by which the Romans and Persians
would keep possession of everything they had already happened to capture in the
country of the Lazi according to the custom of war, whether they were towns or for-
tresses. They would keep peace with each other and make ready for battle against one
another as little as possible until the authorities on each side should come to another,
more important and authoritative, agreement. Zikh, having therefore carried out the
aim of his embassy, went home to his own land. (10) When these matters were
announced to the generals, the forces for a long time remained completely at peace;
and what had previously occurred spontaneously was then reinforced by tr
Embassies continued to pass to and fro between Constantinople and Ctesiphon, resulting, after
a few years, in a comprehensive peace treaty. 42
The strains of life on the frontier (559/60)
Although the Mesopotamian front had been free of campaigning since the 540s, the inhabitants
of Amida in particular were afflicted by ‘an abominable and hideous affliction’ in 559/60.
Pseudo-Dionysius of Tel-Mahre, our chief source for the episode,
43describes a sort of mass-
panic which struck the city: a rumour arose that the Persian king was about to attack Amida,
and so the citizens fled in panic. The panic therefore spread in turn to the surrounding area.
Various other types of unusual behaviour are noted by Pseudo-Dionysius, who reports too that
Constantia and Edessa were similarly affected (Ps.-Dion. II.115–18/104–7, Spurious Life of
Jacob Baradaeus, PO19.259–62). See Harvey 1980: 3–4.
The conclusion of the war, 561–2
From late 561, high-level negotiations took place on the frontier between the two states, at
Dara. 44The Roman representative was the long-serving magister officiorumPeter the Patrician;
that of the Persians was Yazdgushnasp, also a veteran of many previous diplomatic missions.
The peace agreed in the end comprised three treaties, the first of which concerned the most
pressing issue – the return of Lazica to the Romans and Roman payments to the Persians. See
Verosta 1965: 153–4, Verosta 1966: 600–1, Antonopoulos 1990: 110–11, Winter and Dignas
2001: 164–77.
Menander frg.6.1.134–54: The Persians maintained that the treaty should be contin -
uous and that they should receive a fixed amount of gold every year from the Romans
in return for their not taking up arms. Moreover, they wished to lay down their arms
(only) after receiving in a lump sum forty, or at least thirty years’ instalments of the
money to be paid. The Romans for their part wanted the treaty to be a short one and,
in addition, to pay nothing for peace.
45There was a long dispute over this in which
many words were expended, but finally they decided to make peace for fifty years;
Lazica should be ceded to the Romans; the terms of the treaty should be firm and

strong, and prevail on both sides, not only in the East and Armenia, but also in Lazica
itself; the Persians would, however, receive 30,000 goldnomismataper year from the
Romans for the peace.
46It was also established that the Romans should provide in
advance the amount for ten years at one go in this way: that for seven years would (be
paid) immediately, and after the elapse of the seven years, the contribution for the
three remaining years would be paid without delay. Thereafter the Persians in this way
would receive annually the payment due. (tr. Blockley, revised)
The detailed treaty
Once the matter of Lazica and Roman payments had been dealt with, it remained to hammer
out detailed terms designed to ensure that the peace would not be broken by either side or its
47Two items, passed over in the treaty, provoked much debate; one was the question of the
ownership of Suania (whether it formed part of Lazica, and should thus be returned to the
Romans), which was left unsolved. The other concerned Roman payments to Khusro’s
Lakhmid ally ‘Amr, which the Persians, on behalf of the Lakhmids, asserted should be made on
a regular basis. Peter the Patrician denied that the payments had ever been regularised and it
seems that Yazdgushnasp did not succeed in extracting any subsidies for
the Lakhmids.
Menander frg.6.1.314–97: (1) Through the pass at the place called Tzon 49and
through the Caspian Gates the Persians shall not allow the Huns or Alans or other bar-
barians access to the Roman empire, nor shall the Romans either in this area or on any
other Persian borderlands send an army against the Persians. (2) The Saracen allies of both states shall themselves also abide by what has been
agreed and those of the Persians shall not take up arms against the Romans, nor those
of the Romans (against) the Persians.
(3) With regard to Roman and Persian merchants of all kinds of goods, they and
tradesmen of this kind shall conduct their business according to established practice
through the specified customs posts.
(4) Ambassadors and all others using the public post (to deliver) messages, when
they reach either Roman or Persian territory, shall be honoured each according to his
rank and (according to) what is appropriate, and shall receive the necessary attention.
They shall depart without delay, but shall be able to exchange the trade goods which
they have brought without hindrance or any impost.
(5) It is agreed that Saracen and barbarian merchants of whatever kind of either
state shall not travel by strange roads but shall go by Nisibis and Dara, and shall not
cross into foreign (territory) without official permission. But if they dare anything
contrary to the agreement – that is to say, if they engage in smuggling, so-called – they
shall be hunted down by the officers of the frontier and handed over for punishment
together with the merchandise which they are carrying, whether Assyrian or Roman.
(6) Should any persons during the period of hostilities have defected either from
the Romans to the Persians or from the Persians to the Romans, and if, having
defected, they wish to return home, there shall be no obstacle to them and they shall
meet with no hindrance. But those who in time of peace defected, or rather fled, from

one side to the other shall not be received, but every means shall be used to place them,
even against their will, in the hands of those from whom they have fled.
(7) Those who complain that they have suffered some hurt at the hands of subjects
of the other state shall settle the dispute equitably: either those who have suffered
harm themselves or their representatives shall meet on the frontier before the officials
of both states, and in this manner the aggressor shall make amends for the damage.
(8) Henceforth the Persians shall not complain to the Romans about the foundation of
Dara. But in future neither state shall fortify, i.e. protect with a wall, any place along the
frontier, so that no pretext for trouble shall arise from this and the treaty thus be broken. (9) The forces of one state shall not attack or make war upon a people or any other
territory subject to the other, but without inflicting damage or suffering any harm
whatever they shall remain where they are so that they too might enjoy t
he peace.
(10) A large force, beyond what is adequate for the defence of the town, shall not
occupy Dara, and the magister militum per Orientem shall not have his seat there, in order
that this not lead to incursions against or injury to the Persians. It was agreed that if some
such offence should happen, the commander at Dara should deal with the offence. (11) If one city damages another or in any way destroys its property not in accor -
dance with rules of war and with a regular military force but by guile and theft (for
there are such godless men who do these things so that there might be a pretext for
war), it was agreed that the judges stationed on the frontiers of both states
make a thorough investigation of such acts and remedy them. If these prove unable to
check the damage that neighbouring cities are inflicting on each other, it was agreed
that the case should be referred to the magister militum per Orientem
58on the under-
standing that if the matter of dispute were not settled within six months and the
injured party had not recovered what he had lost, the offender should be liable to the
victim of the injustice for a double indemnity. It was agreed that if the matter was not
terminated in this way, the injured party should send a deputation to the sovereign of
the offender. If within one year satisfaction is not forthcoming from the sovereign and
(the injured party) does not receive the double indemnity due to him, let the treaty be
broken in respect of this clause.
(12) Here you might find prayers to God and imprecations to the effect that may
God be gracious and ever an ally to him who abides by the peace, but if anyone with
deceit wishes to alter any of the agreements, may God be his adversary and enemy. (13) The treaty is for fifty years, and the terms of the peace shall be in force for fifty
years, the year being reckoned according to the old fashion as ending with the 365th day. It was also the practice, as I have said, that letters be sent by both rulers stating that
they, too, ratified everything upon which the envoys had agreed. When the terms had
finally been settled, the so-called ‘sacred letters’ were exchanged. (tr. Blockley, revised)
The status of Christians in Persia after 561
The reign of Khusro had seen a certain tendency towards persecution of Christians, at first
between 542 and 545 and then later in the 550s; during the most recent outbreak St Shirin had

suffered martyrdom in 559. 60The treatment of Christians in Persia had long been a cause for
concern to the Romans, and their omission from the detailed part of the treaty may be inter -
preted either as a concession to the Persians or, perhaps, an awareness of how easily any such
clause might prove to be a pretext for war. As it turned out, even the separate annex
61did not
succeed in settling the issue, and among the factors which prompted the resumption of war in
572 was the treatment of Armenian Christians.
Menander frg.6.1.398–407: When these matters had been agreed and ratified, they
turned to a separate consideration of the status of Christians in Persia. (It was
agreed) that they could build churches and worship without fear and sing their
hymns of praise unhindered, as has been our custom. Furthermore, they would not
be compelled to take part in Magian worship nor against their will to pray to the
gods revered by the Medes. For their part, the Christians would not by any means
venture to convert the Magians to our beliefs.
63It was also agreed that the Christians
would be permitted to bury (their) dead in graves, as has been our custom. (tr.
Blockley, revised)
Explicit confirmation of this agreement is provided by Tabari, I, 1000–1/314–15 (Nöldeke
288). See Bosworth 1999: 314 n.737.
Khusro’s encouragement of Arab Christians (c.565)
The Lifeof Mar Ahudemmeh, combined with the discovery of a church at Qasr Serij in Beth
‘Arabaye, sheds an interesting light on Persian willingness not only to tolerate Christianity, but
even to encourage it for political ends. Ahudemmeh was appointed (Monophysite) bishop of
Beth ‘Arabaye by Jacob Baradaeus in 559 and enjoyed great success in converting the Arabs to
64See Key Fowden 1999: 121–8.
V. Ahudemmeh, PO3.29.4–12: He (Ahudemmeh) built a great and beautiful house
of dressed stone in the middle of Beth ‘Arabaye, in a place called ‘Ain Qenoye; and 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 (Tayyaye) peoples were greatly devoted
to the name of Mar Sergis, and sought refuge in him more than in all other men. The
saint (Ahudemmeh) attempted by means of this house which he had built in the name
of Mar Sergis, to cut them off from the shrine of Mar Sergis of Beth Resafa
(Sergiopolis), on the other side of the Euphrates, since it was far distant from them.
He made it, as far as he was able, resemble the other, so that its appearance might hold
them back from going to the other. (tr. M. Greatrex)
It has been suggested, from the evidence of the Life, that the church was founded c.565. Its style
is close to that of churches in northern Syria, and there is no reason to doubt that it was
intended to be an alternative focus to Sergiopolis for devotees of St Sergius. Such a centre
would, as has been noted, reinforce Persian influence among Monophysites in this border
region, as well as reducing the need for cross-border movement – something both powers were
keen to keep to a minimum, as is clear from the treaty of 562.

From the Peace to the death of Justinian (562–5)
For the rest of Justinian’s reign the eastern frontier of the Roman empire was untroubled. There
was just one minor infringement of the terms of the treaty: the Lakhmid king ‘Amr attacked
Harith’s territory in 562 or 563, about which Harith complained during a visit to Constanti -
nople in November 563. Justinian apparently resolved the problem by offering a subsidy to
1No further progress was made on the possession of Suania. 2See Stein 1919: 6.
The opening of Justin II’s reign
Justin II, one of Justinian’s nephews, ascended the throne the day his uncle died, 14 November
565. He at once embarked on a foreign policy quite at odds with the more diplomatic approach
of his predecessor. See Stein 1919: 4–5, Whitby 1988: 250–1.
Soon after becoming emperor, Justin determined to resolve the Suanian question for good. He
despatched John, the son of Domnentiolus, to announce his accession to Khusro and to re-
open negotiations about Suania; he also empowered him to buy back the country, if this was
possible. On his way to the king, John restored the water supply of Dara. In July 567 the two
3and the king complained of Justin’s failure to supply the Arabs with subsidies; John
successfully rejected the charge, arguing that there was no reason why Justin should continue
Justinian’s generous policy. John later brought up the matter of Suania and negotiated the sale
of the country to the Romans; the Persians, however, attached conditions to the transaction,
which Peter was persuaded to accept (Menander frg.9.1). See Turtledove 1983: 293–5.
Another account of John’s mission is to be found in Michael the Syria
Mich. Syr. X.1 (331a/282): In this period 5the Persian king was Khusro. And in the
beginning they enjoyed a true peace. For this reason, in the second year of Justin,
according to the kings’ practice of sending one another gifts when they started to
reign, the patrician John of Callinicum was sent to bring gifts of honour to the Persian
king, and to make peace and unify the churches.
6(tr. M. Greatrex)
Theoph. Byz. 1 (FHG IV.270):In this book 7he (Theophanes) recounts in full how
the truce was ruptured when Justin, through the agency of Comentiolus, demanded
from Khusro the return of Suania. (Khusro) himself had suggested this but had not

restored it. He also records how the whole of Mesopotamia was shaken (by an earth-
quake), 8a prelude to the evils which followed. (tr. Donovan, revised)
Justin was incensed by John’s action, which had given the Suani the opportunity of rejecting
Roman rule, thus strengthening the Persian claim to the territory (Menander frg.9.2). The
emperor therefore informed Khusro and his envoy Yazdgushnasp, then en routeto Constanti -
nople, that he would not accept the terms agreed by Peter. Following the deaths of
Yazdgushnasp and John, Khusro despatched Mahbodh (Mebodes) to Justin in late 567, accom -
panied by a contingent of Arabs sent by ‘Amr. Neither obtained what they had hoped from
Justin: Suania, apparently, was not raised, and Mahbodh accepted Justin’s insistence on ceasing
payments to the Arabs. ‘Amr in turn retaliated, ordering his brother Qabus to attack the terri -
tory of Mundhir the Jafnid (Menander frg.9.3). See Shahîd 1995: 308–14, Stein 1919: 7,
Turtledove 1983: 295–6.
The conflict between Lakhmids and Ghassanids and Mundhir’s withdrawal from frontier defence (569–75)
The Lakhmids saw an opportunity to regain the initiative against their rivals after the death of
Harith in 569 or 570. The new Lakhmid ruler, Qabus, invaded the territory of Mundhir, son of
Harith, but sustained a heavy defeat (probably in 570). Mundhir followed up his victory by
taking the war into Lakhmid territory, where he gathered much plunder. A further attack by
Qabus in the following year was also decisively defeated; Mundhir informed Justin of these
developments, but was unable to secure any assistance from the emperor (Joh. Eph. HEVI.3).
See Shahîd 1995: 340–7. Justin’s reaction was rather to attempt to have Mundhir assassinated;
the attempt was never made, however, because Mundhir accidentally received the letter sent to
Marcian, ordering him to kill the Jafnid chief. He therefore withdrew from defending the
southern Roman provinces for two or three years (572–575) (Joh. Eph. HEVI.4; Chr. 1234,67
10The Romans had come to rely heavily on their allies here, so that Mundhir’s
action greatly weakened the defences of Syria and Euphratesia. See Liebeschuetz 1977: 494–9,
Whitby 1988: 210–11.
Unsuccessful attempts to reconcile Monophysites and Chalcedonians (566–8)
It would be inappropriate to enter into the details of the efforts made by Justin II and his wife
Sophia to bring about a rapprochementbetween the opponents of Chalcedon (themselves
divided) and its supporters. We may note, however, that the consequences of the continuing
division between the two parties was by this point leading to the emergence of a separate
Monophysite hierarchy; and this increasingly distinct church enjoyed the support of much of
the population of the East, on both the Roman and Persian sides of the border. These disputes
would be exploited to good effect by Khusro II in the early seventh century.
Roman negotiations with the Turks
In late 568 or early 569 an embassy from the Turks arrived in Constantinople. Its leader was a

Sogdian, a nation now subject to the Turks, called Maniakh, who had earlier attempted to enter
into friendly relations with the Persians in order to sell them raw silk. This had been rejected by
Khusro, and so the Turkish leader Sizabul had permitted Maniakh to establish contact with the
Romans. The Sogdian asserted that the Turks had crushed the Hephthalites and now wished to
form an alliance with the Romans; he also strongly hinted that they would be prepared to co-
operate against the Persians (Menander frg.10.1).
12In August 569 Justin sent off themagister
militum per Orientem Zemarchus on an embassy to the Turks in response (Menander frg.10.2).
He was well received and accompanied Sizabul on an expedition against the Persians, after
which he was sent back to Roman territory (Menander frg.10.3). The Roman envoys success -
fully regained Roman territory, avoiding a Persian ambush in Suania (Menander frg.10.4–5).
Zemarchus returned to Constantinople towards the end of 571. See Blockley 1985: 269 n.170
with PLRE III, Zemarchus 3. His report to the emperor was to have far-reaching consequences.
Menander frg.13.5: There were many other reasons for the war between the Romans and
the Persians, but the Turks were the nation which most encouraged Justin to open hostili -
ties against the Persians. For they attacked the land of the Medes and laid it waste,
sent an embassy to Justin to urge him to fight with them against the Persians. They asked
him to destroy, in concert with them, those hostile to both of them, and (so) to embrace
the cause of the Turks. For in this way, with the Romans attacking from one direction and
the Turks from another, the (state) of the Persians would be destroyed in the middle.
Aroused by these hopes, Justin thought that the power of the Persians would easily be
overthrown and annihilated. He therefore made every preparation to keep his friendship
with the Turks as firm as possible. (tr. Blockley, revised)
More immediate reasons also impelled the emperor to attack the Persians. First, there was the
Persian intervention in southern Arabia, where in 570 Khusro had given assistance to a
Himyarite prince in expelling the Ethiopians, who were the allies of the Romans. Cf. Tabari, I,
945–58/235–52 (Nöldeke 219–37) with Turtledove 1983: 298, Bosworth 1983: 606–7,
Rubin 1995: 284–6, Shahîd 1995: 364–72.
Theoph. Byz. 3 (FHG IV.270–1):Justin sent Zemarchus as an ambassador to the
Turks. (p.271) He entertained them 16magnificently and was treated with the greatest
kindness; he (then) returned to Byzantium. For this reason Khusro launched a cam -
paign against the Ethiopians, allies of the Romans, formerly known as Macrobioi but
now called Homerites. Through the agency of Miranus,
17a Persian general, he took
captive Sanatourkes, 18the king of the Himyarites, pillaged their city and made the
people subject to him. (tr. Donovan, revised)
The outbreak of the Armenian revolt (571–2)
Thus the arrival of the Turkish/Sogdian embassy of 568/9 had set in motion a chain of events
which eventually led to the outbreak of war. The evident hardening of the Roman stance
towards Persia appears to have had an impact on the Persarmenians, who had been suffering at
the hands of Persian marzbansfor some time, and particularly so under the marzbanChihor-

Vshnasp. 19A secret pact between the Romans and the Persarmenians was concluded in late 570,
and the Armenians openly revolted in late summer 571, supported by the Iberians; in February
572 the rebels killed Chihor-Vshnasp. See Stein 1919: 23–4, Turtledove 1983: 299; on the Ibe -
rians, see Toumanoff 1963: 379–80.
Stephen of Taron, 84.23–86.7 (59.24–60.24): Then in accordance with the
sequence of princes, after Mµeµ Gnuni Persianmarzbanscame to Armenia. First
Denshapuh, who contented himself in adultery and lit the Ormizdean fire in
Rshtunik‘; he forced the Christians to worship the flame, on account of which many
died. And after him one Varazdat, a Persian from the same family, in whose time a
grievous plague with sufferings (occurred). And then a terrible sign, sparkling and
blood-like, began to appear in the sky, which frequently flashed across the northern
district from the west to the east during the whole night in the shape of a column, and
went on for eight months. And then Khusro, king of Persia, raised to the office of hazarapetof Armenia one
Suren, his kinsman, whose name was Chihor-Vshnasp, who, when he came, occupied
our lands by greatly oppressing the Armenian nobles. For he committed adultery with
the wives of the nobles, not accepting the husband as lord of his wife. With him the
bdeashkh Vardan, son of Vasak, who was from the family of the Mamikoneans,
became enraged; having waited for a suitable occasion, he killed Suren the marzban
with the sword (and) threw him to the ground in the forty-first year of the kingship of
Khusro, which was the seventh year of the kingship of Justinian, in the month of Areg,
the 22nd of the month, which is of February, on a Tuesday.
20And the Armenian
princes all rebelled from the Persians; they received assistance from the Greek(s) and
opposed them in a violent struggle. Then Vardan, having taken his relations and other
nobles, took flight to the country of the Greeks, to the royal city Constantinople. And
he came before the king Justinian who built Saint Sophia. And he took the sacrament
with him. And he (Justin) named in his name the main door of Saint Sophia, which
up until today is called the Armenian Gate.
21(tr. Greenwood)
Narratio de rebus Armeniae 77–8:After a long time, a certain Vardan, a ruler, killed
the Persian tyrant Surenas, left the Armenians, and (78) ran off to Constantinople in
the fortieth year of Khusro and the thirtieth year of Justinian, who also built Hagia
Evagr. HEV.7 (203.3–26): While this man (Gregory) was holding the episcopacy in
his first year, 23the inhabitants of what was once called Great Armenia, but afterwards
Persarmenia – which was recently subject to the Romans, but when Philip, the succes -
sor of Gordian, had betrayed it to Shapur, what is called Lesser Armenia
24was pos -
sessed by the Romans, but all the remainder by the Persians 25– (whose people) revered
(the practices) of the Christians and were being cruelly treated by the Persians, espe -
cially with regard to their faith, sent an embassy to Justin in secret. They begged to
become subject to the Romans, in order that they might freely perform the honours
(due) to God without anyone hindering (them). When the emperor had admitted
(their overtures), and certain points had been agreed by the emperor in writing and
guaranteed by solemn oaths, the Armenians massacred their governors; and with their

whole army, bringing (with them) their neighbours, both of kindred and foreign race,
they united themselves to the Roman empire, Vardan having a precedence among
them by birth, dignity and experience in wars. In reply to the complaints of Khusro on
account of these transactions, Justin alleged that the terms of the peace had expired,
and that it was impossible to reject the advances of the Christians who defected to
(other) Christians. Such was his reply. Notwithstanding, he made no preparation for
war, but was caught up in his habitual luxury, regarding everything as secondary to his
personal enjoyments. (tr. Anon., revised)
An interesting sidelight on the events in Armenia is provided by Gregory of Tours. Although
his information is regarded with suspicion by some, his mention of the destruction of the
church of Saint Julian, passed over by other sources, is corroborated by
Chr. 724.
Greg. Tur.Hist.IV.40 (172.10–173.12): ( … ) After this, Antioch of Egypt and
Apamea of Syria, very great cities, were captured by the Persians, and the people car -
ried off as captives. At this time the church of the martyr Julian of Antioch was burnt
down by a serious fire. To the Emperor Justin (p.173) came Persarmenians with a
great mass of woven silk,
27seeking his friendship, and assuring him that they were hos -
tile to the Persian king. For his envoys had come to them, saying, ‘The emperor in his
concern inquires whether you are preserving the peace entered into with him.’ When
they replied that all that had been promised was being adhered to, the envoys said, ‘In
this (matter) it will be plain that you preserve his friendship – if you will worship fire,
as he does.’ When the people replied that they would never do this, the bishop, who
was present, said, ‘What divinity is there, which might be worshipped, in fire? God
created it for the use of men. It is kindled by touchwood, extinguished by water, burns
when attended to and grows warm if neglected.’ As the bishop continued with these
(words) and (others) like them, the envoys were inflamed with anger, and attacked
him with insults and beat him with clubs. Seeing their priest stained with his own
blood, the people rushed upon the envoys, came to blows with and killed them, and,
as we said, sought the friendship of this emperor.
John of Ephesus offers a detailed account of the circumstances leading up to the defection of the
Persarmenians, which he claims is based on the narrative of the Armenian catholicosand his fol -
lowers. John asserts that Khusro’s decision to force a fire-temple on the Armenians was based on
advice from the Zoroastrian clergy, who had heard of the coercion being exerted by Justin to
ensure all his subjects were orthodox (Chalcedonians).
Joh. Eph. HEII.20 (81.28–82.6, 83.15–84.3): After all this (had happened) – as the
catholicos and his companions proceeded to relate – he (Khusro) sent a marzbanto our
region, accompanied by (p.82) (a force of) 2000 cavalry, who came first of all to us, to
our city, delivering the directive that he should erect a fire-temple there for worship -
ping the king.
29When he, as the catholicosrelated and said, showed me and the people
of the city the directive, I burned with zeal and rose up against him, I and all the
people of the city, and we said … (The Christians fail to dissuade the marzbanfrom
proceeding with his task, although they point out to him that Shapur (III) had allowed

them their religious freedom and even show him a copy of the document.) 30But when he
saw their readiness and determination against him, and perceived moreover that they
outnumbered him, he retired with threats and protests against them and in great anger
went to inform the king of all that had taken place. And he, on learning it, was roused
in anger and burned with indignation. Vowing death against all the people of the
region, he sent against them, accompanying his marzban, 15,000 men to do battle
and ordered that they be exterminated if they would oppose his command and that he
should erect there a shrine of a fire-temple. But the people of the region, when they
had heard of these (matters), all assembled together, about 20,000 (in number) and
were prepared for battle (and) to fight to the death for the sake of their Christianity.
When these had arrived and had all drawn up in line against them (the Persians)to do
battle, they cried out ‘In the name of our Lord Jesus Christ’ and moved onward to
attack them. And Christ shattered them before the people of the region, and they
destroyed them all as one man (p.84); they slew the marzbanand took off his head,
and brought it to the patrician Justinian, who was stationed at that time at the border
at Theodosiopolis. ( … ) (tr. van Ginkel)
Mich. Syr. X.1 (331–2a/282–3): ( … ) Afterwards (p.332a) the Persians began to
oppress the Armenian people again, in order that they should worship fire, like the
Magi; for the Armenians were subject to the power of the Persians in this period. For
this reason, the Armenians revolted against the Persians and sought help from the
Romans. And when the Romans helped the Armenians, they defeated the Persians.
Khusro demanded that Justin return the Armenians to him, stating, ‘It is not right
that you give (succour) to a people which has revolted against its king. And if you do
not return the people to me, make the land tributary to me’. Justin replied, ‘I will not
deliver into your hands a Christian people which has forsaken the worship of demons
and has sought refuge with me’. Khusro responded and wrote to Justin for a second
time, saying, ‘If you do not give me the people and do not return the country to us,
give the gold which your kings gave by way of tribute for Armenia; and may peace
endure between us’. Justin replied again in harsh terms, saying, ‘I demand from you
even the tribute which you have formerly received. For it is for him who requests
peace to give tribute; and since you ask me for the country of the North, so we also
demand Nisibis; for it belonged to the Romans, and it was given to the Persians condi -
tionally, as is written in the archives.’
31(tr. M. Greatrex)
Joh. Bicl., a.567?.3 (211.18–21): When the Armenian and Iberian people, who had
received the faith from the preaching of the apostles of Christ, were forced by the Per -
sian king Khusro into the worship of idols, they refused such an impious order and
handed themselves, together with their provinces, over to the Romans. This affair
broke the treaties of peace between the Romans and the Persians.
Theoph. Byz. 3 (FHG IV.271):He (Theophanes) 33recounts in full also how the
Armenians, because of the hostile treatment which they had received at the hands of
Surena, particularly in matters concerning their religion, united and slew Surena
through (the actions) of Vardan, whose brother Manuel he (Surena) had killed, and of
a certain other man, Vardes. They then revolted from the Persians and went over to
the Romans. They left the town of Dvin in which they were living and entered Roman

territory. This action in particular was the cause of the Persians’ breaking of the treaty
with the Romans. Immediately the Iberians rebelled, and under their leader Gorgenes 34
went over to the Romans. At this time Tiflis was the capital city of the Iberians. (tr.
Donovan, revised)
Notwithstanding the events in Persarmenia, Khusro attempted to maintain the peace and in
early 572 sent a Christian ambassador, Sebokht, to Constantinople to receive the payment due
for that year and to confirm that the Romans were still prepared to abide by the treaty. Justin
made clear, however, that he had given his backing to the Persarmenians, as fellow Christians,
and that he was prepared to invade Persia if necessary (Menander frg.16.1).
35As is noted by
John of Epiphania (below), Justin had no intention of submitting to the indignity of making
yearly payments to the Persians.
36A summary of the factors which induced Justin to go to war is
offered by several sources.
Joh. Eph. HEVI.23 (321.26–322.2): The first cause why the peace was broken was
that the Persarmenians surrendered themselves (and their country) to the Romans.
The second cause of bitter enmity was the fact that the emperor of the Romans had
sent ambassadors to the barbarous tribes who live more inland
37from the lands of the
Persians, whom they call (p.322) Turks (Turkioi), together with many other causes
( … ) (tr. van Ginkel)
Joh. Epiph. 2 (FHG IV.273–4), cf. Th. Sim. III.9.3–10:
38When the Emperor Justin-
ian had ruled the Roman empire for 39 years in all, he died. At the end of (that) time,
he lived in peace with everyone including the Medes also. Upon his death, Justin the
younger, who was his nephew, succeeded to the Roman empire, and the treaty which
had previously been made between the Romans and the Persians was broken. (Justin-
ian had made (the treaty) with Khusro, the son of Kavadh, and it was for a duration of
55 years.)
39They fought a great war against each other, which began in the seventh
year of Justin’s reign and lasted for twenty (p.274) years. It ended in the ninth year of
the reign of the Roman Emperor Maurice. (Each nation) attributed to the other the
causes for the strife between them. The Romans were angry because the Persians
intended to bring about the revolt of the Himyarites – an Indian tribe allied and sub -
ject to them – to rebel. When they (the Himyarites) refused to take their (the Per -
sians’) side, they made an attack on them at a time when the treaty was still valid.
Besides this, when the Turks had sent an embassy to the Romans, the Emperor Justin
had even admitted it into his presence. He had also sent out Zemarchus, a man of sen -
atorial rank, with them, (and so) the Persians wanted to bribe the Alans, through
whose territory they (the Romans) intended to travel, in their eagerness to murder
Zemarchus and the Romans and Turks with him. In similar fashion the Medes made
counter-accusations, blaming the Romans for the outbreak of war, since when the
Armenians, their tributaries, were looking to rebel, having killed the governor, Surena
by name, and went over to join the Roman empire, they had received them with hos -
pitable treaty terms and accorded them a close alliance. The strife between them was greatly exacerbated, if anyone should wish to know
the least obvious, but nonetheless the true cause (of the war), because the Emperor

Justin no longer deigned to make an annual payment to the Medes of the 500 lbs of
gold, on which condition the treaty had previously been agreed, or to make the
Roman state tributary to the Persians in perpetuity.
41(tr. Donovan, revised)
Menander frg.13.5: See above, p.137.
The opening of war, 572–3
In the wake of Sebokht’s embassy, Justin made the fateful decision to launch a full-scale war on
Persia. As is clear from John’s account, the newly appointed magister militum per Orientem
therefore had little opportunity to go on the offensive in this year. See Whitby 1988: 254 and
PLRE III, Marcianus 7.
Joh. Epiph. 3 (FHG IV.274), cf. Th. Sim. III.10.1–3: Therefore when the period
had come to an end for which Khusro, the Persian king, had received the money in
accordance with the terms agreed (for it was decided to advance the total amount for
ten years at one and the same time),
42and since in addition no one had attempted to
come to an agreement, Justin, the Roman Emperor, speedily despatched the general
Marcian to the east. He was among the patricians of the senate and related to Justin.
He was not without experience of the dangers of war and besides was brave in appear-
ance and in reality. Marcian crossed the Euphrates and reached Osrhoene. When the
height of summer had already passed, and the barbarians had not so far foreseen any
hostility, he dispatched 3000 infantry to the area called Arzanene. He put in com-
mand of them Theodore and Sergius, whose family was from Rhabdion,
Juventinus, who was in command of the units in Chalcis. They made a sudden attack
and laid waste the Persian territory, and having captured sufficient booty, they
retreated as quickly as possible. ( … ) (tr. Donovan, revised)
The failure to take Nisibis, 573
Only in spring 573 was Marcian in a position to undertake a campaign in Mesopotamia; and by
then the Persians had made ready their defences.
Evagr. HEV.8–9 (203.27–205.28): The emperor sent out his relative Marcian,
appointing him commander of the forces in the East, but did not give him a
battleworthy army or other equipment for war. He occupied Mesopotamia amid
obvious danger and the (possible) ruin of everything, stringing along with him very
few troops, and these without weapons, and with some diggers and cattle-drivers
pressed into service from among the provincials.
44There were (some) small engage -
ments with the Persians near (p.204) Nisibis, although the Persians were not ready for
war. Having gained the upper hand, he laid siege to the city, though the Persians did
not think it necessary to close the gates, and mocked the Roman army quite disgrace -
45And many strange things were observed, indicating the troubles to come, and
among them we saw a newly born ox at the start of the war, from whose neck hung

two heads. (V.9) Khusro, when sufficient preparations for war had been made,
accompanied Adarmahan for a certain distance, and after transporting him across the
Euphrates within their own country, sent him into Roman territory through (the
place) called Circesium. Circesium is a most important town to the Romans, lying at
the outer limits of the state. Not only do the walls, rising up to a limitless height, make
it strong, but also the rivers Euphrates and Khabur which surround it and make the
city a sort of island.
46He (Khusro) himself, with his own force, crossed the river Tigris
and made his expedition towards Nisibis. These events were unknown to the Romans
to a great extent,
47and so Justin, convinced by a rumour which said that Khusro had
either died or was at his very last gasp, was vexed at the supposed slowness of the siege
of Nisibis; and he sent men to urge Marcian on, and to bring him the keys of the gates
as soon as possible. To Gregory, the bishop of Theopolis (Antioch), it was first
announced that the matter was making absolutely no progress, and that he (Marcian)
was bringing disgrace (on the Romans), seeking the impossible against such a great
and notable city with such a contemptible army. For the bishop of Nisibis
48was a great
friend (p.205) of Gregory, having been honoured by him with great gifts, and was
especially discontented with the senseless violence of the Persians towards the Chris -
tians, which they constantly suffered at their hands.
49He desired that his own city be
subject to the Romans, and provided information to Gregory of all that took place in
enemy territory, supplying everything at the appropriate time. These things he (Greg-
ory) immediately reported to Justin, informing (him) of the swift advance of Khusro.
But he (Justin), wallowing in his usual pleasures, paid no heed to what had been
written, and did not wish to believe (it), thinking what he wanted. For meanness and
confidence towards outcomes are the norm for dissolute men, and disbelief if they
turn out contrary to their wishes. He therefore wrote to him (Gregory) at last, passing
over these matters as if they were entirely false, or if they were true, as if the Persians
would not arrive in time (to lift) the siege, or if they did, as if they would be badly
beaten. He therefore sent a certain wretched and arrogant man, Acacius, to Marcian,
ordering him to remove him from office, even if Marcian should have put one foot in
the city before (he came). This he accomplished punctiliously, serving the commands
of the emperor to no advantage. For having reached the camp, he dismissed Marcian
from office in enemy territory, having made no announcement to the army. The cap -
tains and unit commanders, when, having passed the night, they learnt that the com -
mander had been dismissed, no longer came before the multitude, but withdrew in
scattered flight, breaking the absurd siege. ( … )
Chr. 1234, 65 (202.7–203.20): And with the envoy
51, the emperor Justinian sent his
aunt’s son Marcian 52with an army to lay siege to Nisibis. But when they set out and
reached the city of Dara, he (Marcian) sent a part of the army to invade the borders of
the Persians; they took captives and pillaged and returned to Dara. And when the Per -
sians who were at Nisibis saw (this), they were greatly afraid. But the marzbanof
Nisibis, since he was a shrewd man, schemed to go out to Marcian, the leader of the
army, (went) up to the border and, talking with him, asked for a period of four
months while he apprised the king (of the situation).
53But Marcian, because he was a
gentle and simple man, granted (this) to him, and returned to Edessa. But the Persians

brought provisions to Nisibis, and cut down the parks and gardens which were around
the city, as far as a missile can be thrown. They expelled the Christians who were in the
city and apprised King Khusro (of the situation). But the emperor Justinian, when he
heard that Marcian had returned to Edessa, wrote to him that he should swiftly and
without delay go down to lay siege to Nisibis. And Marcian set out when the
emperor’s letter reached him, and reached Dara on the Saturday of Holy Week; he
celebrated the Sunday of the Resurrection (i.e. Easter) in the city of Dara,
54and on the
Tuesday went down to lay siege to Nisibis. And he surrounded it on all sides and put
up a strong fight against it; and the city was close to being crushed. But the emperor
Justinian was extremely angry that Marcian had turned back and arrived at Edessa,
(thus) allowing Nisibis a time of respite. After he wrote to him, so that he would go
and lay siege to Nisibis, and when Marcian had arrived and put up a strong fight
against it, (p.203) in those days all thought that Nisibis had been taken; but Khusro
was by the river Tigris, and ready to come against the Romans, and when the city was
close to being taken Acacius Archelaus, a savage man, came, who had been sent by the
emperor Justinian to remove Marcian from (his) command, without consideration
for rank, and to take his place. And when Marcian heard the emperor’s order, he asked
Archelaus to allow him two days until the city was taken. But he was angered and did
not agree. And at night, when no one was paying attention, Marcian and those of the
officers who were with him left and came to the city of Dara. And at daybreak the
Roman army saw the tent of Marcian
55overturned, and his standard which was taken
from its place, they thought that they had fled before the Persian army coming up
against them; and a commotion and disturbance fell on the camp; they turned to
flight and abandoned (their) tents, laying down their weapons and engines and provi-
sions, and fled to Dara. But when the Persian army, which was at Nisibis, had opened
the gates, it went out and pillaged the tents and everything that had been left behind.
And this evil end happened to the Roman forces in the year 880 (568/9).
56(tr. M.
Joh. Epiph. 3 (FHG IV.274), cf. Th. Sim. III.10.4–5: ( … ) When the winter season
had passed, Marcian assembled his forces again and set out from Dara. The Persians,
under the generalship of Baramanes,
57who had been appointed to command the units
there, came out to meet (Marcian) in front of the city of Nisibis. A hotly contested
battle ensued in the vicinity of a Persian place called Sargathon,
58in which the
Romans vigorously put the barbarians to flight and slaughtered many of them in the
battle. They then made an attempt on the fort at Thebetha and spent ten whole days
there. Since they could not capture it by any means, they returned again to the fortress
of Dara. While the spring was still at its height, they invaded the enemy’s territory for
a second time, and, by the order of the Emperor Justin, they decided to besiege
Nisibis. (tr. Donovan, revised)
Joh. Eph. HEVI.2 (278.8–280.5): Of the events which are now happening in the
eastern territories, we will present some short accounts. First we will start the exposi -
tion of our stories with the illustrious patrician Marcian, a relative of the Emperor
Justin, one of the army commanders who at that time were sent to the East. Burning
with zeal for the nation of Christians, he gathered his army and descended and laid

siege to the city of Nisibis, which until now was held by the Persians. He attacked
vigorously in order to take it. He built against her ramparts all around, and because
he also had engineers with him, he raised against her siege engines, namely lofty
towers and strong bulwarks. The city was in distress. And its inhabitants together
with the Persian garrison despaired of their lives, seeing that at that moment it was
on the verge of being captured by the Romans. When those inside despaired, (and)
when those outside prepared themselves to take the city and destroy it, (and) when
all the Romans thought to be ready to enter it, behold, a violent-tempered man
arrived, named Acacius Archelaus, sent by Emperor Justin against Marcian for no
just reason to deprive him of his command
59and cut his girdle and send him away
from the regions of the East. As soon as he had arrived and shown his order, when
Marcian with his army was about to take Nisibis by storm and they expected to enter
it (p.279) the next day and take possession of the city, the entire army was perplexed
and their arms became weak. The illustrious Marcian, who had been assiduously
making his preparations and was on the verge of capturing Nisibis, when he heard the
order, said to Acacius, ‘Behold, you see what great labour we have taken upon us for
the purpose of capturing this city. Right now, be patient, and grant us a delay of two
days only, and then do what you have been ordered. It is lawful that what the emperor
orders will be done.’ But he became angry with him and insulted him and was furious
and in front of all his troops he grabbed him and pulled him about, and threw him
down and cut his girdle, insulting him and even, as was said, struck him on the cheek.
The whole army was indignant and the entire corps lowered their arms. While
sneering because they had seen an unfair act, they took his banner, lowered it and
turned it upside down. Immediately the entire army fled and turned around and
retreated from the city, lamenting and grieving about what had happened to their
commander, a good man and a believer
60, and also because at the very moment that
they were expecting to enter and take the city, they had turned their backs, while they
were not pursued by their enemies, and they had become (an object of) laughter and
scorn to their haters. But when the Persian army, which was in the city, saw this, the breaking up and
sudden retreat of the Romans, and the overturning of the banner of Marcian, they
were astonished and became encouraged and armed themselves and pursued after
them and fell upon a body of infantry which had remained behind, and defeated and
slew many of them. Thus they returned to their city laughing and mocking at what had happened to the
Romans of their own doing. Forthwith, too, they wrote and informed their king of all
these (things), saying, ‘Come immediately, and let us cross over (p.280) into Roman
territory, because our noble gods, the sun and the fire, have thrown them down,
namely thus, in that they have fallen upon one another because of an order from their
emperor and they have dismissed Marcian with scorn, and have all fled away from the
city.’ (tr. van Ginkel)

The Persian counter-strike (mid-573)
Khusro’s response to the Roman invasion was two-fold. He led an army up the Euphrates,
which divided near Circesium. While he continued northwards, up the Khabur, ready to drive
off the Romans from Nisibis and strike at Dara, a detachment under Adarmahan continued up
the Euphrates into Euphratesia and Syria.
61See Stein 1919: 44, Downey 1961: 561–2, Whitby
1988: 257.
Evagr. HEV.9 (205.29–206.16): ( … ) Adarmahan, therefore, with a considerable
force of Persians and Scenite barbarians, having passed by Circesium, inflicted every
possible injury (p.206) with fire and sword on Roman territory, setting no limits to his
intentions or actions. He also captured fortresses and many villages. No one stood in
his way, in the first place because there was no one in command, and secondly,
because, since the Roman troops were shut up in Dara by Khusro, his foragings and
incursions were made in perfect security. He also directed an advance upon
Theoupolis (Antioch) through his subordinate, but did not go there himself. These
(troops) were repulsed most unexpectedly; for scarcely anyone, or indeed very few
persons, remained in the city; and the bishop had fled, taking with him the sacred
treasures, because both the greater part of the walls had fallen to ruins,
62and the popu-
lace had revolted, wishing to take control by means of a revolution, as often occurs,
especially at times like this. They themselves (the populace) also fled, leaving the city
deserted; and no one at all took thought for (any) preparation or counter-attack
(against the Persians). (tr. Anon., revised)
Joh. Eph. HEVI.6 (292.12–293.13): When the Persian king was still besieging the
city of Dara and saw that there was nobody trying to stop him, he immediately sent a
marzban named Adarmahan against the city of Apamea with a large army. En routehe
seized many fortresses and destroyed and burnt important and strong villages until he
arrived at Apamea. After the Persian (i.e. Khusro) had conquered Antioch (in 540) he
had besieged Apamea and had caused great hardship, and Apamea had also surren-
dered to him after it had received his word (of honour). When the king had entered it
(Apamea) and had been a spectator at the horse-races, he had not destroyed nor
burned anything in it. For this reason they (the citizens of Apamea) were full of confi -
dence that he (Adarmahan) who had come (now) would also do them no harm.
Because of this trust the leaders of the city and the bishop went out to him and paid
homage. He said cunningly to them, ‘Because the city is ours, open the gates for me,
that I may enter and see it.’ They, trusting him, and not expecting that he would do
anything evil, opened the gates for him and he entered (p.293) the city. Immediately
he seized the gates and started seizing and binding men and women and plundering
the city. They brought the entire booty and all the people outside and placed them
outside the city. After they had plundered the entire city of Apamea, which was full of
riches, (accumulated) over a long period, and far more prosperous than most cities,
and had led its entire population and its entire booty outside and also the bishop with
them, they then set it on fire and burned it down completely. Thus they led away all
the captives and all the booty from it and from elsewhere and went down to the king,

while he was still encamped against Dara. And the captives, who were led down from
it and from elsewhere, were counted before the king (and they numbered) 292,000. 64
They were divided (up) and crossed over to Persia. 65( … ) (tr. van Ginkel)
Joh. Epiph. 4 (FHG IV.275), cf. Th. Sim. III.10.6–11.1: While they (the Romans)
were encamped near the city, King Khusro set out from Babylon with the Median
army. Having crossed the Tigris, he made his way through the desert so that his
movement should not be detected by the Romans. When he drew near to Ambar, a
Persian fort (which was situated a five-day journey’s distance from Circesium) 67he
speedily dispatched the general, Adarmahan by name, who crossed the Euphrates at
this very spot (Ambar), to lay waste territory. Khusro had provided him
with 6000 Persians and nomad barbarians.
68Khusro himself
Khabur and hastened on the Romans besieging
Nisibis. Adarmahan, then, crossed the Euphrates in the vicinity of Circesium, and
went out plundering the Roman possessions. For as a result of the previous peace and
the ease which they had enjoyed in abundance during the reign of Justinian, the prac -
tice of war had been neglected by them, and their courage thoroughly corrupted.
Since no one dared to join battle with the barbarians, Adarmahan got very close to the
city of Antioch and after ravaging the districts and fields lying close to the city, he
pushed on into Coele Syria. He made camp not far from the large city of Apamea.
When the inhabitants sent out an embassy and deceived the men (sent)> by announcing that he would preserve the town unharmed.
Once, however, he was inside the town, the Medes plundered its possessions, enslaved
its inhabitants, and consigned it to the flames. Then as quickly as possible they
returned to their own territory.
69After these events the Emperor Justin sent out
Acacius (whom the Romans were accustomed to call (the son) of Archelaus) and
relieved Marcian of his command while he was still besieging Nisibis, suspecting him
of being slack in his duty, since nothing had been done to bring about the capture of
the city. (tr. Donovan, revised)
Mich. Syr. X.9 (349a/312): At once Khusro despatched the marzbanAdarmahan,
and he plundered and laid waste Barbalissus, Qasrin, Beth Dama, Gabbulon, the
(region) surrounding Chalcis, Gazara, the mountain, and the territory of Antioch.
(tr. M. Greatrex)
See also Greg. Tur. p.139 above, and Chr. 724p.150 below.
The siege and fall of Dara (summer–autumn 573) 71
Chr. 1234, 66 (203.20–205.7): Concerning the siege of the city of Dara by the Per -
sians. But when King Khusro, who had been encamped on the river Tigris with his
army, heard what had happened to the Roman army and about their flight from the
city of Nisibis, he set out and came to Nisibis and found all the weaponry of the
Romans; he took everything and came to smite the city of Dara. Khusro had with him
23,000 cavalry, 40,000 infantry and 120,000 farmers to help in the work.
72And when
he had reached Dara and laid siege to (it), he sent stonecutters to cut through the

eastern mountain, and he cut off the aqueducts from it; and with great labour he
diverted the river, which flowed in a deep valley and entered it (p.204). Afterwards he
forced the workmen to build a wall of bricks outside the surrounding wall (of the city).
But when the mountain rock turned out to be hard, they kindled a fire above it, and
afterwards poured vinegar on it, and thus they cracked it. He fought against the city,
night and day, for a period of six months, and he built two mounds against it. And
after six months, when he had been unable to crush the city, he decided to go. He sent
(a message) to the Romans, asking for fivecentenariaof gold from them – as a fine
which he had imposed on the city – so that he might leave them.
73But there was no
reply to him from the citizens. For it is said that that the envoy who had been sent for
the gold did not report the matter to them (the citizens of Dara) because he thought
that the city could not hold out. Therefore Khusro was angry and began to fight
fiercely. They (the Persians) tied pieces of wood from the mounds to the wall, and
crossed over the wall, and overran it. But the Romans who were in the city fled, so that
they might enter the citadel;
74but the leaders did not open the gates, and so the Per -
sians who had entered the city and the Romans fell upon one another, and began to
slaughter and be slaughtered for seven days. Other men were not able to enter until
the city was foul on account of the dead. Then a battle raged against the city; and
against a high tower, which was called Hercules’, they (the Persians) set up three
mounds and ballistaewhich hurled huge stones. And the citizens were much enfee-
bled, and could not get near the tower; and the besieged relented. So now Dara was
taken on account of our sins.
75When the king saw the dead in the middle of the city,
he forbade a slaughter of the leaders of the city; (rather,) he carefully bound their
hands. And he gathered up the gold and silver of the city and amassed a weight of 200
centenaria. When Khusro saw (this), he said to the leaders, (p.205) ‘God will seek
from you all the blood which has been shed; when you possessed all this gold, why did
you not give one hundredth of it to me, and I would have left you ?’ And they swore
that the envoy, who had been sent, had said nothing to them about any gold.
the number of captives, beyond those who died by the sword, was 98,000. 77From the
building of Dara until this capture (there were) 72 years. (tr. M. Gr
John of Ephesus (HE VI.5) offers a lengthier version than that of Chr. 1234, adding a few
details. 78According to John, Khusro’s initial camp was on the northern side of the city; he
commends the competence of the Roman commander John, the son of Timostratus, in
contrast to Evagrius ( HEV.10), who states that John either did little to save the city or actually
betrayed it.
79He further states that the Romans slackened in their defence on account of their
increasing confidence and the cold season. According to John, at the end of the siege, after the
mêlée in the city, the Romans were tricked by the Persians into dropping their guard and
allowing the besiegers to enter the city unhindered. In this way they gained possession of Dara.
Joh. Epiph. 5 (FHG IV.275), cf. Th. Sim. III.10.2: After the Romans had retreated
and reached a fortress situated on a mountain (it was called Mardin by those living
there) King Khusro suddenly applied himself the siege township for six months>. He di the water-(supply) of the city 148

surrounded (it) with ramparts>; he also erected great unds alongside the wall
and employed ramming devices as siege engines. 80When no help came for the inhabit-
ants from outside, he took the city after the Medes had ascended the walls by force. He
plundered the entire city and enslaved the population including John, the son of
Timostratus, who surpassed all the others in power and rank and had assumed the
control and administration of the city. He (Khusro) left behind a sufficiently strong
force and retired to his home territory. Meanwhile the Romans were still tarrying at
the fortress of Mardin with Magnus, the keeper of the imperial purse,
81who had been
entrusted with overall command. ( … ) (tr. Donovan, revised)
The war in Armenia, 572–3
In the wake of the assassination of Chihor-Vshnasp, the Armenians had seized control of Dvin,
but it was soon retaken by the Persians. Later in 572 it changed hands again, as Vardan, in
conjunction with Roman forces under Justinian, successfully expelled the Persians. In the
process, however, the Romans burned down the church of St Gregory outside Dvin, which the
Persians had been using to store supplies (Sebeos 68/7).
82See Whitby 1988: 254–6 and Stein
1919: 38. The extent of Roman gains in the Transcaucasus should not be underestimated: the
defection of the Persarmenians and Iberians gave them a huge advantage in the region, and
immediately rendered the dispute over Suania irrelevant. Henceforth, as will be seen in the
following chapter, Roman armies were able to penetrate deep into Azerbaijan and even to pass the
winter there.
Sebeos, 67.27–68.8/6–7: And it happened in the 41st year of the kingship of Khusro,
son of Kavadh, (that) Vardan rebelled and withdrew from submission to the Persian
kingdom, together with all the Armenians in unison. They killed the marzbanSuren
by surprise in the city of Dvin, and took much plunder and went into submission to
the Greeks. At a time before this, one named Vahan, prince of the country of Siwnik‘,
rebelled and withdrew from the Armenians. He requested from Khusro the Persian
king that they should transfer the chancery of the country of Siwnik‘ from Dvin to the
P‘aytakaran city,
84and that he should assign the city to the census (p.68) of Atropatene
so that the name of Armenians would not be applied to them. And the order was
fulfilled. Then the Greek king made a vow with the Armenians and confirmed the
same pact which had existed between the two kings, the blessed Tiridates and
Constantine. He gave them an imperial army in support. And when they received the
army, they attacked the city of Dvin and, having besieged it, destroyed it from top to
bottom, and expelled the Persian force which was located in it. (tr. Gr
Joh. Bicl. a.571?.1 (212.20–1): The Emperor Justin made Armenia and Iberia
Roman provinces after the Persians had been driven back, 86and the Persian king pre -
pared for war through his commanders.
In the following year, Justinian, perhaps as a consequence of complaints over the destruction of
the church of St Gregory, was replaced by John, while the ineffectual Vardan Vshnasp (who had
replaced Chihor Vshnasp as marzban) was succeeded by Go³ on Mihran, the father of Bahram

Chobin. As is clear from Theophanes of Byzantium (below), John had among his forces Lazi,
Abasgi and Alans. Whether he also had any Roman troops is uncertain, since it is possible that
they had been sent south to help in the Mesopotamian campaign. See Stein 1919: 38–9 and 49
n.2 on the tribes listed in Theoph. Byz. In August 573 Khusro despatched Go³on Mihran
against the rebels, but only gradually did he restore some Persian control to the region (Sebeos,
68/7, 70/10–11).
Composite account of events, 572–3
Chr. 724, 145.12–19: And again in the year 884 (572/3) Khusro and his troops went
up and laid siege to Dara. He sent Adarmahan 88his marzban and went up as far as
Antioch. He set on fire ?Amos and the house of Mar Julian. And he went to Seleucia
and subdued it.
89He went to Apamea and it surrendered to him of its (own) free will.
He set fire to it, captured its inhabitants and went away. And when he went down to
his lord (Khusro), he subdued Dara and captured its inhabitants and emptied it; and
he made some of the people of the Persians dwell in it. (tr. M. Greatre
Theoph. Byz. 4 (FHG IV.271):(He reports) that Marcian the nephew of the emperor
Justin was appointed magister militum per Orientem and in the eighth year of the reign
of Justin was sent out to fight against Khusro. John, the magister militum per
Armeniam and Miranes, the general of the Persians, also known as Baramaanes,
mustered their forces.
90The Colchians, the Abasgi and Saroes, 91the king of the Alans,
allied themselves with the Armenians, while the Sabirs, the Daganes and the
Dilimnite people
92(joined) Miranes. Marcian fought against Miranes near the city of
Nisibis and put him to flight, killing 1200 men in the battle and taking captive 70
men; seven Romans were killed. Already he was laying siege to the walls of Nisibis.
When Khusro learned of this, he assembled 40,000 cavalry and over 100,000
infantry, and hastened to bring help and to combat the Romans. Meanwhile Marcian
was slandered to the emperor as one who was desirous of usurping (the throne).
emperor was persuaded (of the charge) and dismissed him from his command. He
appointed in his place Theodore, surnamed Tzirus, the son of Justinian.
94As a result
of this, disorder ensued and the Romans lifted their siege. Khusro besieged the city of
Dara and captured (it). (tr. Donovan, revised)
It is also worth noting that Firdausi (tr. Mohl 1868: 509–21) offers a brief overview of this war,
unlike Tabari, who omits it altogether. Firdausi reports that war broke out when the new
Roman emperor (i.e. Justin II) no longer wished to be a vassal of Persia. Khusro therefore
attacked the Romans, capturing Beroea (Halab) and Sakila (? Seleucia), despite some problems
in furnishing his troops with pay. The war was brought to an end when Roman envoys brought
the required payments to Khusro in return for a cessation of hostilities

Continuation of negotiations (574–5)
The fall of Dara affected the Emperor Justin severely; he became insane, although he retained
occasional moments of lucidity. Henceforth effective control of affairs was in the hands of
Tiberius, assisted by Justin’s wife Sophia. Early in 574 they received an embassy from Khusro,
seeking to renew negotiations; Sophia replied by despatching Zachariah as an ambassador
(Menander frg.18.1). At the same time Tiberius undertook an extensive recruitment drive,
mainly in the Balkans, to bolster the Roman forces in the East, while Justinian was appointed
magister militum per Orientem and given overall command of the whole eastern front (Th. Sim.
III.12.3–4, Evagr. HEV.14 [209.27–210.2]).
Joh. Epiph. 5 (FHG IV.275–6), cf. Th. Sim. III.11.3–4: ( … ) Not many days
later 2the Emperor Justin, when an illness of the body suddenly afflicted (him),
grew fearful of everything, and made a truce for the present year with the Persians.
Since the ailment (continued to) trouble him, he proclaimed Tiberius, the
commander of the imperial guard – the Romans call this man comes excubitorum–
as his son, and chose him to share control of the empire. He declared him Caesar
and handed over to him responsibility for ruling. (p.276) Of all the things done by
Justin during the course of his rule this decision was the best and the one most
promising of future security; it was the cause of very many benefits in the affairs of
the Romans. As he deliberated on the present situation, Tiberius decided to take
matters vigorously in hand, so that nothing else intolerable would occur … .
(tr. Donovan, revised)
Menander frg.18.2: The Empress Sophia, the wife of Justin, sent , who
ranked among the court physicians, as envoy to the Persian king Khusro. When he
arrived there offering 45,000 nomismata,
4he made a one-year agreement in the East,
so that there might be a truce. 5He said that during this period a major embassy 6would
be sent by the empress, which would discuss everything in greater depth, and which
would in addition put an end to the war, if it should also happen that the Roman
emperor in the meantime returned to health. Thus Zachariah made a one-year truce
in the Roman dominions in the East, but not for those in Armenia, paid over the
45,000 gold nomismatafor this (concession) alone, and departed. When these matters

had been settled, the general Eusebius was recalled to Byzantium. 7(tr. Blockley,
In late 574 8a further embassy was sent from Constantinople, in which served the quaestor sacri
palatii Trajan and Zachariah. Their mission was to secure a general truce for three years to allow
further time for negotiations to continue; their top priority was to gain respite for the
Mesopotamian frontier. The Persians, however, wished for a more enduring period of peace,
and the Romans agreed to a five-year pact, for which the Romans would pay 30,000 nomismata
annually, subject to ratification by the emperor (Menander frg.18.3).
Menander frg.18.4: When affairs of state had already encompassed Tiberius, 9Trajan
and Zachariah the envoys wrote to him that the Persians were unwilling to make a
three-year agreement, but wanted it to be for five years. And he did not assent (to this),
since he did not wish to make a longer truce, and he told them preferably to make the
agreement for two years, and, if this were impossible, not to accept (one) for more
than three years. When, therefore, the letter to this effect had been sent to the envoys
and they had read what the letter revealed, Mahbodh, who had come to the border
near Dara for this purpose, when he learned that the Romans were not satisfied with
the terms approved by Trajan and Zachariah, sent Tamkhusro against the Roman
10He immediately overran and burned the territory close to Dara until
Mahbodh was persuaded to accept the 30,000 gold nomismataeach year for the three-
year peace on the understanding that during this period the high officials of both
states would meet to discuss how arms might be put aside completely. The Caesar
accomplished these things aiming, with some forethought, for his own advantage,
since he knew that within three years his forces would be sufficient for him and would
be clearly battleworthy (to face) the Persians. The Persians, too, were aware of the
Caesar’s forethought, realising that the respite had been (agreed to) by him for no
other reason than that he was looking to the future so that forces for what was neces-
sary might be prepared in advance by him. Nevertheless, they were contemptuous of
the Romans, thinking that they would not be able to retrieve their defeat, even if they
were given more time. When a truce had been made in the East,
11all the tumult (of war) was transferred to
Armenia, which was partitioned between the two sides, and at the beginning of spring
the war began. (tr. Blockley, revised)
Joh. Epiph. 5 (FHG IV.276), cf. Th. Sim. III.12.6–9: ( … ) He (Tiberius) sent Theo-
12who had been in charge of affairs in Armenia and had held many other distin -
guished offices and was adequately endowed with reason and especially well able to
recognise at a glance what was necessary, on an embassy to the barbarians. According
to the usual practice therefore he explained what had taken place concerning him
(Tiberius) and urged Khusro to enter into negotiations. A little later he (Tiberius) in
the same way speedily despatched to the East Justinian, the son of Germanus, who
had been enrolled among the patricians in the senate, entrusting the whole conduct of
the war to him. He was inured to the dangers of war, having reached that (point) in
life where he was tripped up neither through the rashness of youth nor by the

weakness of old age. Justinian therefore arrived as quickly as possible in the East and
took charge of the discipline and order of the soldiers, while Tiberius Caesar speedily
sent out a not inconsiderable force and continually worked hard at making prepara-
tions for war. He distributed an unlimited amount of money and assembled the stron -
gest and most warlike soldiers from among the nations, giving very careful consider -
ation to everything to do with the war which was upon him.
13While these matters
were progressing in this way, and it was a short time into the period of the truce, the
Persians, after assembling opposite Dara, arrived in the vicinity of Constantina, 490
stades to the west of Dara.
14(tr. Donovan, revised)
At some point in 575 the Ghassanid ruler Mundhir met Justinian at Sergiopolis, and there a
reconciliation was effected between the two parties. In the wake of their meeting, Mundhir
launched a raid deep into Lakhmid territory, seizing and plundering their capital at Hira (Joh.
Eph. HEVI.4).
15See Stein 1919: 61–2, Whitby 1988: 264, Shahîd 1995: 373–83 (placing the
raid in spring 575, before the three-year truce was agreed).
Continued fighting in the Transcaucasus (574–5)
While negotiations continued in Mesopotamia, the war was vigorously pursued to the north. In
575 Roman commanders took hostages from the Sabirs, the Alans and other tribes, who were
brought to Constantinople and welcomed by Tiberius. The Sabirs, however, rejoined the Per-
sians soon afterwards (Menander frg.18.5). See Stein 1919: 62–3, Turtledove 1977: 231–2,
Whitby 1988: 264. Help from the Turks, in whom Justin had placed such hopes, never materi-
alised; rather, following the unsuccessful mission of Valentinus in 576–577 to Turxanthus,
they attacked and seized the Roman city of Bosporus in the Crimea (Menander frg.19.1–2). See
Blockley 1985b: 274–8, Sinor 1990: 304. Persian efforts to regain control of Armenia were
only partially successful. See Chapter 10, p.150 with Toumanoff 1963: 380.
The campaign of Melitene (576)
In 575 or early 576 Theodore son of Bacchus 16was sent on a minor embassy to Khusro, partly to
thank the king for his good treatment of Zachariah and Trajan’s mission, and partly to indicate
the emperor’s willingness to resume high-level negotiations on the border (Menander
frg.18.6.1–12). Khusro, however, decided to accept the ambassador and simultaneously to
attack the Romans in Armenia, where the truce did not apply. Arriving from the south earlier in
the year than usual, he took the Romans completely by surprise: the generals Curs and Theo -
dore were still operating around Albania, ensuring the loyalty of the tribes there, while the
magister militum per Orientem, Justinian, had yet to move his forces north from Mesopotamia
to the front (Menander frg.18.6.13–47).
17Khusro passed through Persarmenia unopposed,
bringing Theodore with him; in late spring he arrived outside Theodosiopolis, determined to
take it in order to restore his grip on Persarmenia and Iberia. He was opposed by a Roman force
to the south of the city, and preferred therefore to open negotiations rather than risk an assault
on the city. Theodore was released to go to Tiberius and enjoined to return within thirty days, if
an invasion was to be avoided (Menander frg.18.6.48–116; Joh. Eph. HEVI.8).

Khusro’s despatch of Theodore was only a ruse. Passing by Theodosiopolis, he continued
Joh. Eph.HEVI.8 (298.1–300.9): ( … ) When they 19had blocked him in and were
preparing to do battle with him, he (Khusro) became afraid and turned away from
them to go against another city. They also hastened there against him and opposed
him and repulsed him also from there.
20After they had tested his army, they took
more courage (in their struggle) against him and they despised him. After these
(events), when he saw that it (the campaign) did not go according to his wish, he
marched to the northern mountains and focused his eyes on Cappadocia in order to
enter and conquer Caesarea.
21When the armies of the Romans saw (it), they marched
against him and in front of him also to Cappadocia. They positioned themselves
against him and blocked him in the mountains of Cappadocia and stopped him and
did not let him pass. For many days they encamped opposite one another, while he
did not dare to fight an open battle with them. When he saw that they were more
numerous and powerful than him and that he was not able to pass them and march on
Caesarea, he was alarmed and very troubled. Therefore he began to plan whether he
could pass (them) and make his escape to his own country. His Magi reproached him
and quarrelled with him. Then he turned around and headed towards Cappadocia
against Sebastea, while all were afraid of the Roman armies. Out of shame for being
mocked that he had not been able to achieve anything of what he had planned, he
attacked and burned Sebastea with fire, for he could not find booty and prisoners
because the entire territory had fled from before him.
22When he marched on from
there, he made clear that he was retreating and therefore marched eastward in order to
make his escape, if possible, to his own country. Therefore the Roman armies treated
him with contempt when they came into contact with his army, and despised him and
took courage against him. When he saw that they had surrounded him and taken up
positions against him on all sides, he was forced to flee hastily to the mountain, leaving
behind his entire camp and his pavilion, (p.299) which is his tent, and his entire
equipment and his baggage-train of gold and silver and pearls, and all his majestic
state garments, departing empty-handed. The Romans hastened and entered the
camp and took possession of it, slaying whomever they encountered there. And they
seized his entire equipment and that of his nobles and also the place of worship of his
fire, in which he used to worship, and the horses, which carried and transported it.
There some became rich, as some of the Romans found and took royal goods, banded
together and took what they had found and fled. And they were never found or seen
again. Those who fled and escaped from the camp of the Persians went to their king,
crying and saying, ‘My lord, the Romans have fallen upon us and have slain many of
your servants and have plundered our entire camp and have looted it.’ When he had
heard these (matters), he replied to them, ‘Let them.’ For he had ordered that his
entire army should surround him and should erect a wall of shields for him. He posi -
tioned them in lines while riding between them and beseeched them showing his
white hair to them saying, ‘My brothers and sons, take pity on my grey hair, attack
and fight for the empire of the Persians, lest it be mocked and ridiculed. Behold, I as a

horseman, as one of you, will fight with you.’ For his nobles were constantly quarrel-
ling with him, saying, ‘Whether we live or die, in us the Persian has gained a bad repu -
tation, because none of the kings of the Persians has ever done what you have done,
that is, you have brought us to die in these mountains.’ From these people themselves
the Romans learned what was said amongst them. The Persians had made a plan and
marched (down) on the other side in order to flee (p.300) to a city
23and moved away
against Melitene. Had there not been hatred and dissension between the commanders
of the Romans, had they but agreed with each other, they would have destroyed him
and his entire army there. After having pledged their word to one another, they split
up and surrounded him. They
24faced the Patrician Justinian, son of Germanus, and
Justinian became scared and fled from before him, and his fellow (commanders) did
not follow him closely and help him. When he (Khusro) and his army saw (this), they
became encouraged and took heart and attacked Melitene and set fire to it. (tr. van
Eustratius, V. Eutychii, 1719–32: We all know about the incursion into our state by
the godless Persians, when Khusro, the new Nebuchadnezzar, came to Sebastea and
Melitene. There was therefore at that time much suffering and violence: nearly all the
neighbouring peoples – both those around Nicopolis and Neocaesarea, and Comana
as well as Zela, and (those) of other cities nearby – sought refuge in Amasea, as in the
strongest city, encouraged not so much by the city as by the prayers of the holy man
(Eutychius). For both all the local inhabitants and the foreign peoples – most of them
were Iberians – as well as the citizens of other cities placed their hopes, after God, in
him; and paying heed to his mouth and to what was said by him, they were
Evagr. HEV.14 (210.20–211.16): ( … ) But while he (Khusro) was lingering and
whiling away the time, 26and evading battle, Curs, the Scythian, who was in command
of the right wing, advanced upon him; and since the Persians were unable to stand his
charge, and were most obviously abandoning their formation, he made an extensive
slaughter of his opponents. He also attacked the rear, where both Khusro and the
whole army had placed their baggage, and captured all the royal stores and indeed the
entire baggage. Khusro beheld (this) and endured (it), deeming it more bearable than
a charge of Curs against himself.
27He (Curs), having together with his troops made
himself master of a great amount of money (p.211) and spoils, and carrying off the
beasts of burden with their loads as well, among which was the sacred fire of Khusro
which was established as a god, made a circuit of the Persian camp, singing songs of
triumph. Around nightfall he arrived at his own army, which had already broken up
from its position. Neither Khusro nor they (the Romans) had begun battle; there had
only been some skirmishing, or just even one man from each camp had joined battle,
such as usually happens. During the night Khusro, having lit many fires, made prepa -
rations for a night assault; and since the Romans had formed two camps, he attacked
those in the northern section at the dead of night. On their giving way under this
sudden and unexpected (onset), he advanced upon the neighbouring town of
Melitene, which was undefended and deserted by its inhabitants, and having fired the
whole place, prepared to cross the Euphrates. ( … ) (tr. Anon., re
Sebeos, 68.18–69.8/7–8: This is that Vardan, against whom the Persian king, called

Khusro Anushirwan, came in person with a multitude of armed men and many
elephants. Having advanced through the province of Artaz, he crossed Bagrevand and
passed beside the city of Karin.
28And he advanced and came to Melitene and camped
opposite it. On the morning of the following day, with great speed they drew up,
contingent facing contingent and line facing line, and they engaged one another in
battle. The battle intensified over the face of the earth and the battle was fought
fiercely. And the Lord delivered defeat to the Persian king and all his forces. They were
crushed before the enemies by the edge of the sword and fled from their faces in
extreme anxiety. Not knowing the roads of their flight, they went and threw them -
selves into the great river which is called Euphrates. The swollen river carried away the
multitude of fugitives like a swarm of locusts, and not many were able to save them -
selves on that day. But the king escaped by a hair with a few others, taking refuge in
the elephants and cavalry. He fled through A³ znik‘ (Arzanene) and arrived back at his
own residence. And they seized the whole camp with the royal treasures. (p.69) And
they took the queen and the women and they took possession of the whole pavilion,
the golden carriage of great weight which had been adorned with precious stones and
pearls, which having been (so) named was called by them the carriage of glory. The
Fire was seized which the king used to take about with him continually for his assis -
tance, which was considered greater than all fires, (and) which was called by them
At‘ash. It was drowned in the river with the mobadhan mobadh
29and a further multi-
tude of the most senior people. At all times God is blessed. 30(tr. Greenwood)
Th. Sim. III.14.1–9 also recounts the battle at Melitene and Khusro’s retreat, but adds nothing
to the sources already cited. See Schreiner 1985: 285 n.426, Whitby and Whitby 1986: 95
Following this defeat west of Melitene, Khusro was forced to retreat eastwards in haste, sacking
the city en route. Much of his army was destroyed as it attempted to ford the Euphrates; such an
impression did the expedition leave on Khusro that he passed a law forbidding future Persian
kings from undertaking similar campaigns in person.
Joh. Eph. HEVI.9 (300.12–302.10): After the Persian had attacked the city of
Melitene, he immediately gave orders to set fire to it completely. But when he left in
order to cross the Euphrates, heading towards his own country, the commanders of the
Roman armies sent him the following message: ‘This, what you have done, namely
attacking and burning a city, is not in accordance with the stature of a king, namely to
create ruins and run away. Even for us ourselves, servants of the emperor, it would be
very disgraceful if we were to do what you have done. How much more (is it disgraceful)
for you, because you not only think of yourself as a king, but even as the king of kings.
For it is not proper for a king to do such deeds, to come with a band of robbers, to
plunder, flee, set fire and burn. But it befits a king authoritatively, confidently and
regally to take up position openly in battle, and whenever he wins, let him subsequently
triumph as a king, and let him not enter as a thief, cause damage, steal and run away. But
prepare yourself and we will do battle against one another in the open in order that both

victory and defeat will clearly be known to the others.’ After he had heard these (things),
he ordered a battle on the next day on the plain 32to the east of the city at some distance
from it. (p.301) Before dawn, both parties were lined up against each other a short
distance apart. They were lined up, arrayed and facing each other from the dawn until
the sixth hour. No man moved from his position and the king was standing behind his
army and they looked at each other (wondering) who would start. Certain people have
told us these (things) under oath and have narrated to us all these (events). They were
the interpreters
33of the Persians and Romans. (They said): ‘Finally we, the three of us,
kicked our horses and left the ranks of the Romans for the area in between both parties
and we came at full speed up to the edge of the Persian ranks and turned around at full
speed without making contact, three times, and we were fast because of the pace of our
horses. Both parties were looking intently and staring at us. For we went out in order
to provoke them to do battle. Even then none of them moved from his position and
went out against us, while they were lined up and arrayed as a wall. They did not even
utter a word to each other.’ Finally he (Khusro) sent word: ‘At the moment there
cannot be a battle today, because too much time has passed.’ Therefore the armies
turned away from each other. At night, before dawn reddened, the king and his army
were found near the river Euphrates, making plans and labouring about how he would
cross the river, six miles from Melitene, for the Romans pursued him in order to vex
and destroy him. This they did achieve. The Persians were pressed together and when
the Persian army saw the Roman army in pursuit, they threw themselves on their
horses into the river and more than half of his army submerged and drowned. He and
the rest, on horses and by swimming, saved themselves with difficulty and got across
(the river) and hastily drove on into Roman Armenia. While they marched on
quickly, he ordered all villages which they came across (p.302) to be burnt down with
fire. In this manner he bent his course to the high mountains of Qurha,
35where there
never had been any road, and he was forced to position his army in front of him in
order that they would make a road for him. Therefore they cut timber from the woods
and at some places levelled mountains and cut through them and made a road. Thus,
confused and worn out, he barely escaped out of the hands of the Romans and entered
his own territory. For this reason he made a ruling and made a law that a king should
not go out to war, if he does not go out against (another) king. (tr.
van Ginkel)
Evagr. HEV.14 (211.17–25): When the Roman army had assembled and followed him
(Khusro), however, he feared for his own safety, mounted an elephant, and crossed the
river. A great multitude of his army found a grave in the waters of the river; realising these
men had drowned, he went off and left. Having paid this extreme penalty for his great
insolence towards the Roman power, Khusro, with the survivors, regained the eastern
(lands), where he had a truce that provided that no one should attack him.
36( … ) (tr.
Anon., revised)
Joh. Bicl. a.575?.1 (214.7–16): The Persian king Khusro, with an overwhelming multi -
tude, moved his armies forward to lay waste the Roman borderlands. Against him
Justinian, appointed by Tiberius as leader of the Roman army and magister militum per
Orientem, prepared for war; he had with him extremely brave peoples, who are called
37in the barbarian language. In the plains which lie between Dara and Nisibis he

(Justinian) met him (Khusro) in a fierce engagement and overcame the aforementioned
king in battle. 38When he (Khusro) turned in flight with his army, he (Justinian)
overran his camp. The victorious Justinian laid waste the borders of the province of
Persia and sent (back) the weapons (taken) from them to Constantinople for the
triumph. Among other things (sent) were 24 elephants,
39which provided a great spec -
tacle for the Romans in the royal city. To the huge benefit of the public purse a multi -
tude of the Persians – the booty from the spoils of the Romans – w
as put on sale.
Th. Sim. III.14.10–11 (140.5–19): And so the Babylonians (i.e. the Persians), having
been defeated, fled as fast as they could, while the Romans, keeping on the offensive,
gave the Parthians an experience of evils. Furthermore, in addition to this they also
plundered the Persian camp, pillaged the king’s tent, and gloriously carried off all their
equipment. They captured the elephants and despatched them to the Caesar together
with the Persian spoils. (11) When the king of the Persians had been defeated and made
his retreat homewards in shame, on coming to Melitene he set fire to the beauty of the
city, since he found that it was undefended
40and enjoying complete quiet. After
crossing the Euphrates and making his withdrawal through Arzanene, he inscribed the
disgrace of the failure in a law: for he decreed that in future it did not befit the Persian
king to travel on expeditions of war.
41(tr. Whitby and Whitby, revised)
The Roman counter-attack (576–7)
In the wake of Khusro’s withdrawal, the Romans went on the offensive; their counter-attack
lasted until mid-577. See Whitby 1988: 267.
Evagr. HEV.14 (211.25–30): ( … ) Nevertheless Justinian made an irruption into
Persian territory with his entire force, and passed the whole winter season there with-
out anyone troubling (him) at all. He withdrew about the summer solstice without
having sustained any loss whatever, and passed the summer near the border, amidst
much comfort and renown. (tr. Anon., revised)
Th. Sim. III.15.1–2 (140.20–9): And so when the king of the Persians had thus
paraded his misfortune in the law, he was at a loss as to what he should do. But the
Roman (army) seized on the Persian disasters and marched towards the interior of
Babylonia, ravaged and removed everything in their path, and what they encountered
became a victim of destruction. (2) Then they became marines on the Hyrcanian
(Caspian) Sea and, after great achievements and the infliction of misfortunes on the
Parthians, they did not return to their own (territory); for the winter season inter -
vened on their actions, and disaster waxed fat in Persia. With the arrival of spring the
Romans retired, carrying off bravery too as travelling-companion.
42(tr. Whitby and
Whitby, revised)
Joh. Bicl. a.576?.2 (214.25–7): Romanus, themagister militum, son of the patrician
Anagast, captured the king of the tribe of the Suani alive; he brought him, together
with his treasury, his wife and his children, to Constantinople, and subsumed his
province within the dominion of the Romans.

The resumption of negotiations (577)
Roman fortunes were at their height in late 576, following Khusro’s inglorious retreat. 44The
Persian king therefore sent off Nadoes to reopen negotiations; in response Tiberius despatched
Theodore, son of Peter the Patrician, John, Peter and Zachariah to Constantia. The Persian
envoy for these talks was the sarnakhorganMahbodh. Little progress was made at first, as discus -
sion centered around the attribution of blame for the outbreak of war (Menander frg.20.1; Joh.
Eph. HEVI.12). Probably in 577 Mahbodh proposed that the Romans could obtain peace by
resuming their annual payment of 30,000 solidi, ceding Persarmenia and Iberia, and handing
over those who had rebelled there (Menander frg.20.2.1–15).
Menander frg.20.2.15–69: When Mahbodh made these (proposals), the Roman
envoys, as they had been enjoined by the Caesar, immediately said that this could not
be called a peace if the Persians hoped to keep the Romans (liable) for a certain sum, as
if for the payment of tribute.
46The Caesar would not accept to pay any of this money
nor would he buy peace like something openly for sale. For if this was the case, (the
peace) would be neither enduring nor stable. The Roman envoys said that he
(Mahbodh) should first therefore renounce this (demand) and thus explore on what
terms the (cause) of peace might progress. And again, for this reason, much verbal sophistry was expended by both sides.
Initially the Persians did not give way, but argued that they (the Romans) should
either pay the 30,000 nomismata(mentioned) above or make one large payment, if
needs be. Finally, however, Mahbodh decided to show them a letter which had
recently been sent to him by his king, in which he insisted that, on account of (his)
friendship for the Caesar, he would offer to renew the peace on equal terms
without any money (involved). When news of this spread through the capital, all
became excited, and both the authorities and the rest of the populace thought that
swords would at once become idle and peace firmly established, since the Caesar
too was very ready to hand over Persarmenia and Iberia to the Persians. For he saw
very clearly that if they were deprived of this considerable tract of territory, they
would never give up, even if Persian fortunes should completely decline and grow
weak. (He declared), however, that he would surrender neither the princes of the
Persarmenians nor their relatives, nor indeed anyone at all who had willingly
deserted to the Romans, and, moreover, that he would make peace only on condi -
tion that there should be no impediment to those of the Persarmenians and
Iberians who wished to leave their country and migrate to the Roman empire. For
the Caesar laid great store by the oaths of the Emperor Justin to the Persarmenians
and those of the Iberians who had come over (to him). The emperor had sworn
that as far as he was able he would use every means to make their native land
subject (to him), but if he proved unable to sustain the war to the end, he would
never hand over those who had raised the revolt, their blood relatives and, in short,
all those who wished to have a share in the Roman empire. The Persian king decided to be content with these terms: that the Romans evacuate
Persian Armenia and Iberia and that he allow that the inhabitants here should go

wherever they wished – not unreasonably, I think – for he knew that, with the excep-
tion of a very few of those in office who had initiated the revolt, none of the
Persarmenians and Iberians, out of love for their native land, which resides and is
lodged in men by nature, would migrate to foreign parts. Furthermore, he also hoped
that when the war was ended, he would soon settle the situation in Persarmenia and
Iberia. For these lands were highly productive and brought him a very large income.
For these reasons therefore the Persian king was willing to end the war on these terms.
(tr. Blockley, revised)
Tiberius also sought to regain Dara, offering to buy it back from the Persians (Menander
frg.20.2.68–78). A deal was at the point of being struck when a surprise victory of the Persians
in Armenia (in summer or autumn 577) brought all the negotiations to nothing. See Whitby
1988: 219 and PLREIII, Tamchosroes.
Sebeos, 71.1–4/11: Then came Tamkhusro, and he had two campaigns, one to
Basean, at Bolorapahak, where the Murts‘ and Araxes join, and one to Bagrevand, at
47And in both he achieved a very great victory. He stayed for two years and he
went.48(tr. Greenwood)
Th. Sim. III.15.8–9 (141.21–142.4): At that time then, a fierce battle was joined
between Romans and Parthians for Armenia, with Tamkhusro commanding the
Babylonian force and Justinian leading the Roman throng; the Romans fell short of
(their) (p.142) former glory. (9) It was for this reason that the Medes gave up on the
peace treaty and their love of war was rekindled again, since they were incapable of
moderation because of their recent successes.
49(tr. Whitby and Whitby, revised)
These Roman defeats, combined with their ill-treatment of the Armenians reported by John of
Ephesus (HE VI.10, II.24) and alluded to by Menander (frg.23.4), persuaded many to return to
their former allegiance (Joh. Eph. HEVI.11).
50See Stein 1919: 70, Turtledove 1977: 265–6.
A raid into Osrhoene, ravaging the area around Constantia, Resaina and Tell Beshme, was
undertaken by Adarmahan in 577; the attackers drew off, however, when the people of
Constantia warned of the imminent arrival of more forces under Justinian (Joh. Eph. HE
The resumption of the war in Mesopotamia (578)
Late in 577 Tiberius appointed Maurice magister militum per Orientemto succeed Justinian,
bestowing on him in addition overall command in the East in an attempt to eliminate the
internal dissensions which were plaguing the eastern army.
52On his way to the frontier, he
recruited fresh forces in Cappadocia, and proceeded to Citharizon, where he met up with his
fellow commanders (Joh. Eph. HEVI.14, 27).
53Forewarned of the arrival of more Roman
troops, Khusro decided to launch a pre-emptive strike in Mesopotamia (probably in June
578); his chief negotiator, Mahbodh, was put in charge of the operation, a calculated insult to
the Romans. Mahbodh’s army, consisting of 12,000 Persians and 8,000 Sabir and Arab allies,
laid waste the region around Resaina and Constantia; it then retreated to Persian territory by
way of Thannuris, a Roman fortress now in Persian hands. Simultaneously the Persian

commander in Armenia, Tamkhusro, having tricked Maurice into expecting an attack on
Theodosiopolis, instead skirted Citharizon and headed south. In a quick raid, lasting under
three weeks, he despoiled the country around Martyropolis and Amida (Menander
frg.23.1.16–43, 23.6, Joh. Eph.HEVI.14, 27, Th. Sim. III.15.11–12).
54See Honigmann
1935: 22 and Szádeczky-Kardoss 1976: 109–10.
Neither raid brought the Persians any significant advantage (Menander frg.23.5), and Maurice
was quick to respond. See Stein 1919: 74. Despite having a fever, he set off eastwards into
Arzanene, deporting much of the (Christian) population. He laid siege to the fortress of
Chlomaron, but was unable to capture it either by force or by persuasion; he did, however, take
Aphumon, and installed a Roman garrison there (Joh. Eph. HEVI.15, 34; Menander frg.23.7).
A passage from Agathias clearly shows that Maurice penetrated well into Arzanene, reaching
almost as far as Corduene.
Agath. IV.29.7–8 (161.7–15): ( … ) For he (Khusro) happened at that time 56to be
moving towards the Carduchian mountains, to the village of Thamnon, to take up
residence (there) for the summer season because of the mildness of the region. (8)
Maurice the son of Paul, appointed magister militum per Orientemby the Roman
emperor Tiberius (II) Constantine, suddenly invaded the land of Arzanene, which is
adjacent and close to the territory surrounding the village,
57and indeed did not cease
ravaging and plundering the whole (country) without mercy. After crossing the waters
of the river Zirma, he moved up further still, despoiling and burning (whatever) lay
close by.
Th. Sim. III.16.1–2 (143.13–21): Accordingly, after Arzanene had thus suffered
harm from the Roman spear, the general (Maurice) changed course and invaded the
lands of Arabia situated not far from Nisibis in the most direct fashion.
59(2) Next,
after laying waste as far as the river Tigris, he dispatched Curs and Romanus across to
the other bank to ravage the entire enemy (territory); but, after he himself had laid
waste the fort of Singara, since the winter season was peeping in,
60he collected his
forces and arrived among the Romans. (tr. Whitby and Whitby, revised)
Tiberius followed up Maurice’s remarkably successful first campaign with an invitation to the
Persians to come to terms. In late 578, around the time of the Emperor Justin’s death, he again
sent Zachariah, on this occasion accompanied by another Theodore,
61to Khusro, simulta -
neously releasing many Persian captives taken during the war 62(Menander frg.23.8.1–12).
Menander frg.23.8.13–24: ‘I too want peace and welcome (it) because it is God-given,
just as within me there is a natural friendship with you. Therefore I am prepared to give up
all Persarmenia and Iberia, save indeed those of the Persarmenians and Iberians who wish
to be subject to us. I both give up the fortress of Aphumon and shall concede Arzanene
you, recovering from you only Dara in exchange for so many things.’ These things the
Emperor Tiberius indicated to Khusro, and he sent off Zachariah and Theodore with the
powers of major envoys
64to arrange a peace as best they could. (tr. Blockley, revised)
In the meantime, the inhabitants of the Roman frontier provinces prepare
d for the worst.

Chr. 1234, 73 (209.4–8):(…)Atthis time, 65there was anxiety in the cities beyond
the Euphrates; and they were steadfast, by night and day, on Sundays and feast-days,
in building the walls of the cities; and all were in distress and fear.
66(tr. M. Greatrex)
Continued negotiations (579)
Khusro had also been sufficiently impressed by Maurice’s campaign to despatch an envoy of his
own, Ferogdath, in an effort to re-open negotiations on the frontier. He arrived in Constanti -
nople while Zachariah and Theodore were still on their way to Persia, and was offered the same
terms as Tiberius had outlined to his own emissaries (Menander frg.23.8.25–57). According to
Menander, peace would have been agreed at this point, had not Khusro died before the Roman
ambassadors reached Persia (in February or March 579). They proceeded with their mission
nevertheless, armed with a further letter from Tiberius, stating that he was willing to make
peace with the new king, Hormizd IV, on the same terms as with his father. At the same time
Maurice returned to the East from Constantinople, to await developments there (Menander
frg.23.9.1–23). But Hormizd proved to be uncompromising and refused to accept Tiberius’
terms; he insisted upon the payment of the yearly sums agreed by Justinian, thus undoing all the
progress made in negotiations over the previous years. The Roman ambassadors were treated
with contempt, and their mission returned to Roman territory after a long and unpleasant stay
in Persia (Menander frg.23.9.24–125).
67The length of time taken up by the fruitless embassy
meant that no campaigning was undertaken on the Mesopotamian front in this year. 68Fighting
continued in Armenia meanwhile, where Tamkhusro was succeeded by Varaz Vzur. But the
Romans continued to have the upper hand; a notable victory of the commanders Curs and John
Mystacon is reported, although so too is a Roman defeat (Sebeos 71/11, Joh. Eph. HEVI.28).
Hormizd and the Christians of Persia
Roman sources display a hostile attitude to Hormizd, during whose reign there was a continual
state of war with Rome. His son and successor, Khusro II, did nothing to improve his father’s
image while he was a refugee on Roman soil. The Persian tradition is more varied, and Tabari in
particular preserves some material quite favourable to the king.
70See Nöldeke 1879: 264 n.5,
Christensen 1944: 442–3, Bosworth 1999: 295 n.696.
Tabari, I, 991/298 (Nöldeke 268): 71‘Just as our royal throne cannot stand on its two
front legs without the two back ones, our kingdom cannot stand or endure firmly if we
cause the Christians and adherents of other faiths, who differ in belief from ourselves, to
become hostile to us. So renounce this desire to persecute the Christians and become
assiduous in good works, so that the Christians and the adherents of other faiths may see
this, praise you for it, and feel themselves drawn to your religion.’ (tr. Bosworth)
The Roman offensive of 580
Early in 580, following a visit to Constantinople, Mundhir scored another victory against the
Lakhmids, who had sought to exploit his absence from the front (Joh. Eph. HEIV.42). See
Shahîd 1995: 398–406, Stein 1919: 92, PLREIII, Alamundarus.

Only one source, Theophylact, reports the events of 580. 72
Th. Sim. III.17.3–4 (145.21–146.3):When Hormizd’s boastfulness was made plain
to the Emperor Tiberius and summer had again arrived, Maurice collected his forces
and moved into Persia, after sending Romanus, Theoderic, and furthermore Martin
to the far side of the Tigris to lay waste the interior of Media. (4) And so they invaded
with the mass of the army, and pillaged the fertile and most fruitful (areas) of the Per -
sians; after spending the whole of the summer season in the slaughter of Persians, they
ravaged Media, and wrought destruction with great firmness. (tr. Whitby and
Whitby, revised)
The Sasanian diplomatic offensive in the Transcaucasus (c.580)
Soon after the accession of Hormizd, the Persians regained the initiative in the Transcaucasus.
As the excerpt below attests, Hormizd’s despatch of his son Khusro to Albania successfully loos -
ened the Roman grip on the region. In the light of this manoeuvre, Hormizd’s intransigence
with Roman peace efforts becomes more understandable. See Toumanoff 1963
: 380–2.
HVG 217/228–9: Then Hormizd, the king of the Persians, gave Ran (Albania) and
Movak‘an to his son, who was called Kasre Ambarvez (Khusro II). He came and took
up residence at Partaw, and he began to hold talks with the erist‘avebiof K‘art‘li
(Iberia). He promised them great benefit and put in writing for them the patrimony of
their eristavates from child to child, and thus did he win them over by blandishment.
The erist‘avebi apostasised and began to give individual tribute to Kasre Ambarvez.
The children of Bakur remained in the mountainous part of Kakhet‘i; and as for the
descendants of Mirdat, son of Vakhtang, who ruled in Klarjet‘i (Cholarzene) and
Javakhet‘i, they remained amongst the rocks of Klarjet‘i. The Persians took all the rest
of K‘art‘li, Somkhiti (Armenia) and Asp‘uragan (Vaspurakan) and began to wage war
on the Greeks. (tr. G. Hewitt)
The campaigns of 581
Theophylact also provides the most credible account of the campaign of t
he following year. 73
Th. Sim. III.17.5–11 (146.3–28): With the arrival of winter (580), the Roman leader
came to Caesarea in Cappadocia, but as the summer (581) came round again, he
arrived in the east with the whole army at Circesium, a city of the Romans. (6) Next,
he subsequently hastened through the desert of Arabia to reach the land of Babylonia
and then to steal a victory by the shrewdness of the enterprise. (7) In this he was
accompanied by the leader of the nomadic barbarians (his name was Alamoundarus
[Mundhir]) who, they say, revealed the Roman position to the Persian king; for the
Saracen tribe is known to be most unreliable and fickle, their mind is not steadfast,
and their judgement is not grounded in prudence. (8) Therefore, as a result of this, the
ˆ ˆ

king of the Persians transplanted the war to the city of Callinicum, after appointing
Adarmahan as a not untalented custodian of the expedition. (9) Then, after Mundhir
had like a drone destroyed the beehives (full) of honey, or in other words had ruined
Maurice’s enterprise, the manoeuvres of the expedition against the Medes became
unprofitable for the Romans: for they returned to quench the disasters at home. (10)
And indeed the general consigned to burning flames the grain ships which had accom-
panied him down the river Euphrates; he himself, with the pick of the army, came
with all speed to the city of Callinicum. (11) When the Parthian contingents came to
grips, the Roman spear won supremacy; then flight came upon the Persians, and their
insolence received a check. (tr. Whitby and Whitby, revised)
Joh. Eph. HEVI.16–17 (312.15–314.4): Then Maurice and Mundhir, the son of
Harith, king of the Arabs, together gathered their armies as one and crossed over into
the territories of the Persians by the route through the desert. They penetrated many
day-marches into the territories of the Persians, as far as Beth Aramaye. When they
had reached the great bridge of Beth Aramaye, full of confidence to cross over it and to
capture the important cities of the Persian Empire, it was found that the bridge had
been broken down. When the Persians had heard (of their plan), they had broken it
down. After they (Maurice and Mundhir) and their troops had seen much hardship,
especially the Romans, they came to an argument with one another and turned back
without being able to achieve anything to their advantage. Worn out, they could but
barely escape and leave for the territory of the Romans. They wrote heavy accusations
against one another. Maurice was of the opinion about Mundhir that he had given
notice in advance and had informed the Persians (of their plan) and that they had
(therefore) broken down the bridge lest they should cross. This was a lie. (p.313) The
emperor was deeply concerned, sending messengers to both commanders, and with
difficulty he reconciled them with one another. Eventually Maurice went up to the
emperor and it is not known whether he accused Mundhir (or not).
74(17) When the
Persians saw that Maurice and Mundhir were descending upon their land and saw
that the (Roman) territory was without an army, the marzbanof the Persians, named
Adarmahan, crossed over into the territory of the Romans with many troops and
arrived in the territory of Constantia and Resaina. He destroyed and burnt the rest of
what they had left the first time. They crossed over into the territory of Edessa, a
wealthy region, and burnt and destroyed and ruined the entire region of Osrhoene.
Full of confidence and without fear he moved through the region for many days as if
he was staying in his own country, while he did not leave one house standing every -
where he passed by, sneering at the entire army of the Romans because it was inca -
pable of guarding it (i.e. the region). Eventually, when Maurice and Mundhir came
up from the territory of the Persians, worn out, and he (the marzban) heard that they
intended to march against him, he sent them the (following) message: ‘Since I have
heard that you are coming against me to do battle, do not exhaust yourselves by
coming, because you are tired from the exertions of the journey, but rest and I will
come to you.’ Eventually, after the destruction and plunder and the capturing of
people which he had accomplished, and after he had led away the captives and had
done whatever he wanted, and after he had heard that they intended to march against

him, he took with him his loot and all the captives which he had made, and moved to
his own territory (p.314), while nobody of the 200,000 Romans who were eating the
emperor’s meal had attacked him. When he had made up his mind to go, they
marched out to go up against him. When they did not catch up with him, they said
that he had fled. (tr. van Ginkel)
According to John, Mundhir himself then proceeded to exact revenge on both his Lakhmid and
Persian opponents; following a victory in the field, he captured and plundered the Lakhmid
camp (Joh. Eph.HEVI.18).
Chr. 1234, 74 (209.21–210.15): And Mundhir took forces of the Romans and Sara -
cens and entered Persian territory. They laid siege to a certain fort which was called
‘Anat (Anatha). They fought greatly against it; they armed themselves and got on
small boats in the waters of the Euphrates, and made war from the boats. But the
Persians fought them from the wall with the stones of ballistae, and many were
drowned and perished. And they withdrew from there, while taking captives and
pillaging the territory. But the leader of the Persians, whose name was Adarmahan,
who had gone to Apamea and destroyed it, (p.210) had been left at Nisibis by Khusro.
And while the Romans were laying siege to Anatha, he came to Edessa and laid it waste
and set fire to it, took captives, slew many, overturned many churches, and came
against Callinicum.
76But the forces who were with Mundhir assembled there; and
when they fell upon each other, many of the Persians were slaughtered. 77And at dawn
on the following day, when they were ready to join battle, Adarmahan, knowing that
he could not fight with Mundhir, cleverly sent (to him), saying, ‘(Since) tomorrow is
Sunday, let us not go into battle, but on the next day, Monday, let us prepare for
battle.’ And the Romans agreed. When it was dark the Persians ate and drank, and lit a
fire in front of the gates of the tents, as was customary, so that they should be thought
to be there; but in the evening they set off and fled. And in the morning none of them
was found. But they crossed over to Constantia; they laid waste the monastery of
Qartmin and the region of the Tur Abdin and came to Nisibis.
78(tr. M. Greatrex)
Chr. 819 (10.13–14): And in the year 891 (579/80) the Persians went up again and
set fire to the monastery of Qartmin, and Constantia, Carrhae and Edessa.
79(tr. M.
Evagrius (HE V.20 [215.27–216.5]) claims that Maurice defeated Adarmahan, and that
Theoderic, who was probably serving as comes foederatorum, and his men behaved in a cowardly
fashion at the battle.
At the same time as Maurice invaded Lower Mesopotamia, Roman forces were involved in a
campaign in Persarmenia, which also met with failure 81(Menander frg.23.11).
The Persian offensive of 582
As a result of the recriminations which followed the failure of the previous year’s campaign,

Mundhir was arrested at Huwwarin (Evaria), south-east of Emesa, and taken to Constanti-
nople. Henceforth the Ghassanids turned their fire against the Romans under the leadership of
Mundhir’s son Nu‘man; Arabia, Syria and Palestine suffered at his hands, and he even laid siege
to Bostra. A Roman attempt to appoint another ruler for their allies proved fruitless, and
although Nu‘man soon voluntarily came to Constantinople, the Ghassanid federation was
never again united under the leadership of one chief (Joh. Eph. HEIII.40–3, 54–6 [chapter-
headings only], VI.41–2, Evagr. HEVI.2).
82See Shahîd 1995: 455–78, Stein 1919: 93–5,
Sartre 1982a: 190–2, Allen 1981: 246–7.
Tiberius sought to revive negotiations after the campaigning of 581, and Zachariah was
despatched to the frontier once again. His interlocutor was Andigan.
83While the talks were
underway, Maurice occupied his troops with the fortification of Shemkhart in eastern
84Andigan and Zachariah met on the frontier – in tents made ready for the occasion
– and went over the same ground as before. Andigan tried to intimidate the Romans through
his knowledge of Roman difficulties elsewhere in their empire, but was rebuffed by Zachariah.
After Andigan then attempted to put pressure on Zachariah by drawing attention to the army of
Tamkhusro mobilised nearby, just outside Nisibis, negotiations were abandoned in favour of
conflict. Tamkhusro, accompanied by Adarmahan, advanced westwards towards Constantia,
where Maurice, from his base at Monocarton, had been preparing for just such an attack
(Menander frg.26.1).
85Outside Constantia the Romans met Tamkhusro’s forces in June 582;
in the ensuing battle the Persian commander was killed and his forces defeated. The Persians
withdrew, but remained in Roman territory for a further three months (Joh. Eph. HEVI.26,
Menander frg.26.5, Th. Sim. III.18.1–2).
86See Whitby 1988: 274, Turtledove 1977: 330.

After ten years of largely inconclusive conflict, neither side ventured on any large-scale
campaigns in the following few years. While the Romans tightened their grip on Arzanene, the
Persians maintained pressure south of the Tur Abdin, but without any significant gains. Almost
the only source available for much of the campaigning is Theophylact Simocatta, who was
himself reliant on earlier accounts.
1See Olajos 1988 and Whitby 1988: 222–42.
The Roman offensive in Arzanene (late 582)
Maurice’s successor as magister militum per Orientem, John Mystacon, led an army into
Arzanene, and was met at the border, at the junction of the rivers Nymphius and Tigris, by the
Persian general Kardarigan. The battle which ensued initially favoured the Romans, but dissen-
sion among the commanders led to a Roman defeat (Th. Sim. I.9.4–11).
2See Whitby 1988:
Further campaigning in Arzanene (583)
Arzanene continued to be the focus of military activity in the following year. While the Persians
attempted to recapture Aphumon, a Persian fort taken by Maurice in 578, the Roman general
Aulus laid siege to Akbas, a recently constructed Persian fortress just east of the river Nymphius
and thus right on the Roman border. The Persians besieging Aphumon were able to join the
defenders of Akbas in repelling the Romans initially, but by the end of the year the fortress had
fallen into Roman hands. Rather than occupy it, the Romans chose to destroy it (Th. Sim.
I.12.1–7, Joh. Eph. HEVI.36).
3See Whitby 1988: 277–8, PLREIII, Aulus.
The Roman offensive into Beth Arabaye (584)
Much of this year was taken up in unsuccessful negotiations, known only from chapter head -
ings of lost parts of John of Ephesus’ Ecclesiastical History(VI.37–9) and from a short entry in
Michael the Syrian (X.21 [379b/361]).
4While the talks proceeded, the new magister militum
per Orientem, Maurice’s brother-in-law Philippicus, strengthened the fortifications of
Monocarton (Th. Sim. I.14.6). In the autumn came a new Roman strategy: having assem -
bled his forces at Monocarton, Philippicus proceeded to the Tigris. After a lengthy march, he

arrived at a place called Carcharoman, from where he proceeded to lay waste the region around
Nisibis. Kardarigan, who was about to undertake an offensive into the Tur Abdin, was forced to
return eastwards to deal with the incursion. Philippicus withdrew his forces before meeting
Kardarigan; some, it appears, went north, while others returned to Roman territory south-west-
wards to Resaina (Th. Sim. I.13.1–12, Evagrius, HEVI.3 [224.2–6]).
Joh. Eph. HEVI.43 (277.3–5): Concerning famous princes among the Persian
marzbans, who were taken prisoner and went up in chains to the royal city (i.e. Con -
6(tr. M. Greatrex)
In this year too the power of the Jafnids was divided following the arrest and deportation of
Joh. Eph. HEVI.41–2 (276.32–277.3): Concerning (p.277) the elevations and
abasement of the chiefdom of the Tayyaye of the Romans. (42) Concerning those of
the leaders of the Tayyaye who went off (and) surrendered to the Persians. (tr. M.
Mich. Syr. X.19 (375a/350–1): The kingdom of the Tayyaye was divided between
fifteen princes, and many of them inclined towards the Persians; and the empire of the
Christian Tayyaye came to an end and ceased to exist because of the treachery of the
7(tr. M. Greatrex)
The campaigns of 585
Philippicus resumed campaigning in Arzanene in this year, but then fell ill and entrusted his
forces to Stephen and Apsich. The Persians, meanwhile, went on the offensive. Kardarigan
moved westwards from Nisibis and laid siege to Monocarton. After failing to capture the fort,
he moved northwards across the Tur Abdin and Sophanene towards Martyropolis, where
Philippicus was; but having sacked a monastery to the west of the city, he swiftly returned to
Persian territory (Th. Sim. I.14.1–10).
8See Whitby 1988: 279–80.
The campaign of Solachon (586) 9
When Philippicus returned to the front from Constantinople in spring 586, he was met by the
Persian envoy Mahbodh at Amida. The Persians again demanded money in return for peace;
and the bishop of Nisibis, Simon, was also sent to convey the same proposals. But Maurice
rejected the Persian demands, and the magister militumtherefore moved southwards, first to
Mambrathon, then to Bibas (Th. Sim. I.15.1–15).
10The Persian commander, Kardarigan,
advanced westwards to meet the Romans, but Philippicus had so positioned the Roman forces
that the Persians would have to traverse a waterless plain before engaging them (Th. Sim.
The Persians, taking with them adequate water supplies, moved towards Constantia and
Monocarton. Philippicus, apprised from captured Persians of Kardarigan’s movements,
arranged his forces for battle in the plain of Solachon, a few miles east of the Arzamon.

Theophylact provides a very full account of this battle, perhaps derived from the campaign
journal of Philippicus and/or a source associated with Heraclius, the father of the Emperor
Heraclius (Th. Sim. II.2.1–7).
Evagr.HEVI.3 (224.6–11): Battle was also joined with the Persians, and, after a hard
battle had taken place and many of the most noteworthy Persians had fallen, he
(Philippicus) took many prisoners and let go unharmed a detachment which had
retreated to one of the opportune hillocks (there), (and) which he could have
destroyed; they promised that they would persuade their king to send (an embassy) for
peace as quickly as possible. (tr. Anon., revised)
Joh. Eph. HEVI.44 (277.5–7): Concerning another war in the third year, and the
victory which was given by God to the Romans.
13(tr. M. Greatrex)
Th. Sim. II.3.1–4.14 offers a detailed account of the battle, drawing attention to Philippicus’
use of the ‘image of God Incarnate’ to inspire his troops. 14Although the Roman right wing
under Vitalius was initially victorious, the desire of the Roman troops to loot the enemy camp
might have led to disaster for the Romans in the centre and on the left. Philippicus was able to
draw Vitalius’ troops back into the fray, however, and the Romans eventually emerged trium -
phant. Kardarigan escaped to a hill, where he remained for several days. See Whitby 1988: 281.
The defeated Persians were refused admission to Dara as they fled. Philippicus, once he had
ascertained that no Persian reinforcements were in the vicinity, rewarded soldiers who had dis-
tinguished themselves in the battle. Kardarigan and his men escaped from the hill, but not with-
out further losses (Th. Sim. II.5.1–6.13).
Philippicus followed up his victory by an incursion into Arzanene. Although some of the inhab-
itants of the region offered to collaborate with the Romans, a Roman attempt to capture
Chlomaron was thwarted by the arrival of Kardarigan and the union of his forces with those
stationed in the city (Th. Sim. II.7.1–8.12). Philippicus was therefore obliged to withdraw his
forces to Aphumon, and from there he proceeded to Amida. He then attended to the strength-
ening of fortifications in the Tur Abdin (Th. Sim. II.9.1–17).
15When Philippicus fell ill again,
it was left to Heraclius to conduct a further invasion of Persian territory. He proceeded across
the Tigris, penetrating as far as Thamanon in Corduene before turning southwards, crossing
the Tigris once more, and despoiling Beth Arabaye. From there he returned to Amida by way of
Resaina (Th. Sim. II.10.1–5).
The campaigns of 587
Philippicus, still unwell, entrusted most of his army to Heraclius, while giving a third of his
forces to Theodore, a native of the Tur Abdin, and Andrew 17(Th. Sim. II.10.6–7). Heraclius
undertook a further expedition into Persian territory and succeeded in capturing a well-
defended unnamed Persian fortress. Theodore and Andrew for their part, having received intel -
ligence from a local, recaptured the fort of Beiuades (Sina), north-east of Dara, from the Per -
sians and restored that of Matzaron (to the north of Dara). Philippicus then moved to
Constantia for the winter (Th. Sim. II.18.1–26).

Mutiny on the eastern front (588–9)
Philippicus, as he returned to Constantinople for the winter (587–588), learnt that he had been
replaced asmagister militum per Orientem by Priscus. He therefore made public Maurice’s
decree reducing the pay of soldiers by a quarter. Priscus meanwhile advanced to Monocarton,
where he had ordered the troops to gather, by way of Antioch and Edessa. In April 588 he
arrived at Monocarton, but his demeanour angered the soldiers, and such was his unpopularity
that he was forced to withdraw to Constantia (Th. Sim. III.1.3–15, Evagr. HEVI.3–4).
soldiers elected Germanus, the duxof Phoenice Libanensis, 20as their leader; efforts at mediation
by Priscus through the bishops of Constantia and Edessa met with failure. Maurice therefore
restored Philippicus to his command (Th. Sim. III.2.1–11, Evagr. HEVI.5–6). The mutiny
persisted nonetheless, and 5000 soldiers set out against Edessa. The arrival of Philippicus at
Monocarton did nothing to calm the situation; and the Persians exploited it by marching on
21The mutineers, however, repelled the incursion, and proceeded to launch an
attack into Persian territory. Tensions within the Roman camp were at last eased by the arrival
of Maurice’s envoy Aristobulus (Th. Sim. III.3.1–11). The Roman army then moved north to
Martyropolis, from where a force was sent out to attack Arzanene; it may have been this force
which was defeated by the new Persian marzbanin Armenia, Aphrahat (Armenian Hrahat), to
the west of Lake Van at Tsalkajur (Sebeos 71/12). See Higgins 1939: 32–3. At any rate it was
also opposed by the Persian general Maruzas and withdrew towards Martyropolis; near the city
a battle was fought, in which the Romans were overwhelmingly victorious. Much booty was
captured and sent on to Maurice in Constantinople (Th. Sim. III.4.1–5, Evagr. HEVI.9).
October (588) Antioch was struck by a severe earthquake, which, according to Evagrius, killed
as many as 60,000 people, and thus inflicted a further financial and demographic blow to the
largest city of the East (Evagr. HEVI.8). See Allen 1981: 251, Downey 1961: 568–9. In spring
589 the emperor paid the soldiers promptly, while Germanus and other leaders of the rebellion
were summoned to Constantinople; there they were tried, convicted, but pardoned (Th. Sim.
III.4.6, Evagr. HEVI.10).
If Theophylact is to be believed, a remarkable Roman feat took place around the time of the
mutiny: Romans captured at Dara in 573, together with imprisoned Qadishaye, succeeded in
effecting an escape from the ‘Prison of Oblivion’ in Khuzistan and fighting their way back to
Roman territory (Th. Sim. III.5.1–7, Theoph. A.M. 6080, 261.29–26
The renewal of war (589)
Perhaps prompted by the need to deploy more troops in the Balkans to face the growing Avar
threat, Maurice set in motion a co-ordinated Roman offensive to bring about an end to the east -
ern conflict. Four strands will be noted below: a Roman-sponsored raid by the Iberians and
their allies into Albania, the appointment of Romanus to a command in Lazica, an offensive
into Persarmenia by the magister militum per Armeniam John Mystacon, and an incursion into
Beth Arabaye by the Roman forces in Mesopotamia. This last was disrupted by the Persian cap -
ture of Martyropolis, but took place later in the year.
24See Higgins 1939: 40 for this plan.
In April a reconciliation between Philippicus and the troops in the East was brought about at
Litarba by Gregory, the bishop of Antioch. The rebels and Philippicus then met at Antioch, and
an amnesty was granted (Th. Sim. III.5.10, Evagr. HEVI.10–13).
25Despite their setback near

Martyropolis of the preceding year, the Persians now gained control of the city through the
defection of a Roman officer, Sittas. Although Philippicus at once laid siege to the city, he was
unable to take it or to prevent a Persian army under Mahbodh and Aphrahat from joining the
defenders. Maurice therefore replaced Philippicus with Comentiolus (Th. Sim. III.5.11–16,
Evagr.HEVI.14, Chr. 1234, 78 [214/113–14]).
26In autumn 589 the new magister militum per
Orientem, having captured the Persian fort of Akbas and thus tightened the Roman blockade of
Martyropolis, undertook an invasion of Beth Arabaye and confronted the Persians at
Sisauranon; the outcome was a decisive Roman victory, attributed largely to Heraclius by some
and to Comentiolus by others (Th. Sim. III.6.1–5, Evagr. HEVI.15).
Events in the Transcaucasus (589)
Since the late 570s the chief theatre of war between the two powers had shifted southwards to
the former Armenian satrapies (Justinian’s Armenia IV), Mesopotamia and Beth Arabaye. The
Persians had in the meantime regained control of Persarmenia, and Hormizd sought to tighten
his grip on the region by appointing his son Khusro as the ruler of Albania and Iberia. See p.163
above and Toumanoff 1963: 380–1.
28Trouble even broke out in Roman Armenia at this time:
a plot was formed to murder the commander John (Mystacon), which was foiled by the arrival
of the imperial agent Domentziolus. The leader of the rebels, Smbat, was taken to Constantino -
ple but pardoned for his sedition (Th. Sim. III.8.4–8). See Whitby 1988: 291, PLREIII,
Symbatius 1 (dating the uprising to 589), Higgins 1939: 38–9.
In April 588 the Persian general Bahram Chobin gained a decisive victory over the Turks
threatening Persia’s north-eastern frontier. From there he proceeded westwards to the
Transcaucasus, where he arrived in the following year to repel the Iberians and their allies, who
were just beginning to embark on a raid of Albania (Th. Sim. III.6.9–14, Sebeos 73–4/14–15).
See Higgins 1939: 38, Thomson and Howard-Johnston 1999: 168.
HVG 217–18/229: After a few years, great troubles arose among the Iranians, for the
king of the Turks entered into Iran. The Greeks came and harassed the Persians in
Mesopotamia and, penetrating into Persia, began to devastate (it). Then Kasre
Ambarvez (Khusro II) left Ran (Albania) and K‘art‘li (Iberia) and went to the aid of his
father. While the Iranians were thus occupied, all the erist‘avebiof Iberia, of the Upper
and Lower (country), conferred and dispatched an envoy to the emperor of the Greeks,
asking him to place a king over them from the Iberian royal family, but leaving them as
erist‘avebi, one and all, undisturbed in their principalities. (p.218) The Caesar fulfilled
their wish and gave to them as king the son of the brother of Mirdat, Vakhtang’s son by
(his) Greek spouse, who was named Guaram and who ruled over Klarjet‘i (Cholarzene)
and Javakhet‘i.
29The Caesar conferred upon this Guaram the dignity of curopalatesand
sent him to (the city of) Mts‘khet‘a. 30(Toumanoff, rev. Rapp)
Having received financial support from the emperor, Guaram enlisted some allies and invaded
Atropatene in spring 589 (HVG 219–20/230). But when news came that Bahram Chobin had
defeated the Turks decisively, many of the invaders withdrew, including the Iberians them -
selves. As anticipated, Bahram then turned his attention to Roman (and allied) territory, under -
taking a plundering expedition into Suania. A Roman force under Romanus, operating in
ˆ ˆ

Albania, was forced to withdraw to Lazica after a minor defeat; but in a subsequent engagement
by the river Araxes in Albania Romanus gained a notable victory over the Persians. Bahram was
then dismissed and insulted by Hormizd, and therefore determined to rebel (Th. Sim.
III.6.15–8.3, 8.10–11).
31See Whitby 1988: 291, Thomson and Howard-Johnston 1999:
168–9, PLREIII, 1091.
The rebellion of Bahram Chobin and its impact on the frontier
Among the first supporters of Bahram’s revolt were the soldiers of Nisibis, recently defeated by
Comentiolus, who feared punishment from Hormizd. Supporters of Hormizd were killed and
preparations made to march southwards to the heartlands of the Persian kingdom (Th. Sim.
IV.1.1–9). Comentiolus naturally profited from the divisions in the Persian camp, and at the
end of the campaigning season of 589 captured Akbas (Th. Sim. IV.2.1). John Mystacon for his
part, who had been besieging Dvin, proceeded further into Persian territory, overrunning
Atropatene (Sebeos 74/16). See Higgins 1939: 39, Thomson and Howard-Johnston 1999:
169. In February 590 Hormizd was deposed before Bahram could oust him, and his son Khusro
crowned in his stead. Bahram refused to come to terms with Khusro and, having defeated him
in battle on 28 February, seized the throne for himself. Khusro fled to Roman territory and
sought Maurice’s help in regaining his position (Th. Sim. IV.2.2–12.8, Evagr. HEVI.17, Chr.
1234, 80 [215–16/115–16]).
32See Whitby 1988: 292–6 and Higgins 1939: 27–30 (for precise
Naturally Khusro attempted to prove his goodwill to the Romans, sending a message to the
garrison of Martyropolis to surrender to the Romans, initially to no effect (Th. Sim.
33In a second petition to the emperor, sent probably in summer 590, Khusro
proposed an alliance so that he could be restored to the throne (Th. Sim. IV.13.2–26). 34On the
territories conceded by Khusro to Maurice see Garsoïan 1999: 264–7
Th. Sim. IV.13.24 (177.23–7): In exchange we give back Martyropolis, we shall
forthwith freely offer up Dara, and we shall lay war in the tomb without payment,
having established peace by bidding farewell to Armenia, on whose account war ill-
fatedly gained free rein among men. (tr. Whitby and Whitby, revised)
Sebeos 76.8–18/18–19: Then king Khusro sent to king Maurice prominent men
with gifts and wrote as follows: ‘Give me the throne and royal place of my fathers and
ancestors and send me an army in support through which I shall be able to defeat my
enemy and restore my kingdom, and I shall be your son. I shall give you the regions of
Syria – all Aruastan
35as far as the city of Mcbin (Nisibis) – and out of the land of
Armenia the area of Tanuter authority 36as far as Ayrarat and the city of Dvin, and up
to the shore of the lake of Bzunik‘ and to Arestawan; 37and a great part of the land of
Iberia, as far as the city of Tpklis (Tiflis). Let us observe a pact of peace between us
until the death of us both; and let this oath be secure between us and between our sons
who will reign after us.’
38(tr. Greenwood)
Narratio de rebus Armeniae 93–6:The general Mushe³ the Mamikonean, called the
Taronite, went away to Persia and, once victorious, confirmed Khusro in his rule.

(94) He reigned for 38 years. This man gave all of Armenia, as far as Dvin, to the
Emperor Maurice, 40in return for which he was established as king beneath him. (95)
And he received many great gifts and honours from him – he himself and his armies.
(96) In the same year the Persian king Khusro sent Mushe³ to Maurice.
Thus Maurice was presented with the chance not only to obtain peace without any payments
but also to make territorial gains unparalleled since 298. See Goubert 1951: 169–70. Despite
some opposition,
41he decided to back Khusro and sent Domitian, bishop of Melitene and a rel -
ative of his, to the king at Hierapolis (Th. Sim. IV.14.1–6, Evagr. HEVI.18). 42Bahram mean -
while, having crowned himself king in March, outbid the offer of his rival and proposed to hand
over Nisibis and Beth Arabaye to the Romans – to no avail, however (Th. Sim. IV.14.8).
of the emperor’s backing for Khusro only emerged late in the year, 44but as it did, defections to
his camp multiplied: his supporters gained control of Atropatene, while the forces of Nisibis
also rallied to his side. In late 590 (or early 591) Martyropolis was duly surrendered to the
Romans, and Sittas arrested and executed (Th. Sim. IV.15.1–17, Evagr. HEVI.19).
Whitby 1988: 298–300, Higgins 1939: 46–7 (suggesting that the handover may not have
occurred until early 591). The restoration of the city to Roman control was apparently com -
memorated by an inscription set up on Khusro’s instructions, only fragments of which survived
until the nineteenth century. See Mango 1985: 91–104.
Section 1: mention of the Euphrates.
Section 2: … which they had from ancient ?times … to our army it was handed over
… to us and …
Section 3: (?In this) manner [we] wrote to our slaves … your state came back to life
and waged war against … happened you know. But it is not necessary that what was
seized … lest in any way a greater injury follow … those who accomplished such things.
Section 4: mention of the Romans and of Nekra and of today.
Section 5: mention of ‘the god, king of kings’, which is interpreted as a reference to
Khusro himself. 47
Sections 6–7: … your state … cities by the providence of the gods and fortune …
through the army and the might of … which the Romans had at Nekra and now …
because we appeared considerate to … I confirmed as we said and becau
se …
Section 8: I wrote to the Romans and … [so that] he might better know (that) we
ordered … to be affixed on the gate and just as on the tablet … it was inscribed and
affixed in the same manner and to the … of the cities of which by the providence and
help … we received for (?) one and each … to be inscribed on a t
The tide continued to turn in Khusro’s favour in 591. The Persian king implored the aid of St
Sergius, and was careful throughout his stay in Roman territory to appear favourable to Chris -
48and soon after his prayer to St Sergius (7 January 591), Zatsparham, one of Bahram’s
commanders, was killed by forces loyal to Khusro while en routeto Nisibis. Maurice replaced
Comentiolus with Narses at Khusro’s request, as well as furnishing him with a loan and troops
(Th. Sim. V.1–2, Chr. 1234, 81 [216/116]).
49In February Khusro was acclaimed king at
Mardin, close to the Persian border, an event of sufficient magnitude to cause many Syriac
sources to date his reign from this point (rather than the previous year). See Higgins 1939:
47–9. The Roman army crossed the frontier in spring 591 and was admitted into Dara, which

Khusro then handed back to Roman control. In response to this gesture, Maurice called Khusro
his son – a fateful gesture, as it turned out (Th. Sim. V.3). 50While the king and the Roman
forces of Mesopotamia advanced to the Tigris, probably in June, Mahbodh was sent south -
wards to seize control of Ctesiphon (Th. Sim. V.6). The main field of conflict was Chnaitha, a
region west of Atropatene. Here Bahram unsuccessfully sought to prevent the Roman–Persian
army of Khusro and Narses from linking up with the Roman army under John Mystacon
advancing from Armenia. The final battle at Ganzak, by the river Blarathos, in late summer was
a decisive victory for Khusro and his allies. The bulk of the Roman forces then returned to
Roman territory (Th. Sim. V.7–11; Sebeos 79/23).
51See Higgins 1939: 53–4, with Whitby
1988: 302–3. Bahram himself, having escaped his defeat, continued to oppose Khusro, but to
little effect (Th. Sim. V.15.1).
The conclusion of the war (591)
Th. Sim. V.15.2 (216.10–13): The treaty between Romans and Persians proceeded
on equal terms, and thus indeed that great Persian war was gloriously brought to an
end for the Romans.
53(tr. Whitby and Whitby, revised)
Sebeos 84.20–32/28–9: Then king Khusro gave gifts to them all according to each
one’s status and dismissed them from him. He himself set out from Atropatene and
reached Asorestan, his own royal residence. He was confirmed on the throne of the
kingdom, and he carried out his promise of gifts for the emperor. He transferred to
them all Aruastan as far as Nisibis; and the land of Armenia which was under his
authority, the Tanuter tun as far as the river Hurazdan, the province of Koteik‘ as far
as the town of Garni, and up to the shore of the lake of Bzunik‘ and up to Arestawan,
and the province of Gogovit as far as Hats‘iwn and Maku. The region of the
Vaspurakan gund was subject to the Persian king. Out of the Armenian nobles, many
were in the Greek sector, and a few in the Persian. But the king summoned Mushe³ to
the palace, and he saw his country no more.
54(tr. Greenwood)
On the territories ceded by the Persians to the Romans, see also p.172 above. Iberia was proba -
bly divided between the two powers, each acquiring a zone of influence in a theoretically inde -
pendent kingdom (HVG 221/230–1). See Toumanoff 1963: 384–6, Martin-Hisard 1998b:
1178. Maurice reorganised the whole of Roman Armenia following these gains. Because these
new arrangements were only fully operational for at most a decade, their impact on frontier
defence and organisation is unclear.
The optimism which attended the peace is well brought out in a Lifeof St Golinduch, a Persian
convert to Christianity who came to Roman territory in the 580s, which was compiled in 602
by Eustratius.
Eustratius, V. Golinduch 23 (170.1–26): Nor were these things alone in this condi -
tion. 57And through the help of God, the aforementioned most holy man (i.e.
Domitian) well directed the movement of affairs; and ‘he broke down the wall that
separated them’ (Eph. 2:14) and ‘he abolished the law of the commandments in
decrees’ (Eph. 2:15). Christ our God, again in person through those worthy of

himself, that is, through the priest and the king, 58pacified the states. Both (nations),
belonging by nature to God and granted by him to those worthy (of him) came to an
agreement, both according to the kinship of the union in the flesh and according to
the purpose of the gifts of the Holy Spirit (cf. 1 Cor. 12:1–4), that is, of the high
priesthood and the kingship.
Hence Christ is both our high priest and king and is named according to
Melchizedek (cf. Heb. 5:5–10 and elsewhere), and the angel which told the good news
to the mother of God and ever-virgin Mary said: ‘Behold, you will conceive in your
womb and bear a son, and you shall call his name Jesus. He will be great and will be
called the Son of the Most High; and the Lord God will give to him the throne of his
father David, and he will reign over the house of Jacob for ever; and of his kingdom
there will be no end.’ (Luke 1:31–3) When therefore through the assistance of the true
and natural high-priest and king Christ our God and through both his servants, both we
and the Babylonians gained a peaceful end (to the war), the cities – both that of the Holy
Martyrs (i.e. Martyropolis) and Dara – were given back to our Christ-loving state. (The
peace was accomplished) through their embassy and the recently slain martyr
60also known as Mary; (and) the aforementioned most holy man (Maurice)
invited the victorious woman to present herself in the ruling city of the Christians … . 61
The years of peace (591–602)
The situation of both powers in the last decade of the sixth century favoured peace. Maurice was
occupied in combatting the serious Avar threat in the Balkans, while Khusro was faced with a
rebellion by Bestam, whose brother Bindoes he had executed.
62Just as in the opening years of
the fifth century, the two powers were in constant contact. See Sako 198
6: 106.
Between late 591 and early 592 Khusro despatched gifts to the shrine of St Sergius at
Sergiopolis. The text which accompanied the gifts is preserved by both Theophylact and
Evagr. HEVI.21 (235.10–236.16), cf. Th. Sim. V.14.4–6 (212.21–213.20):
Khusro, having become master of his own kingdom, sent to Gregory a cross, adorned
with much gold and with precious stones, in honour of the victorious martyr Sergius;
this (cross) Theodora, the wife of Justinian, had dedicated, and Khusro had carried off
with the other treasures, as has already been mentioned by me. He also sent another
golden cross, and Khusro engraved these things on the cross in the letters of the
Greeks: ‘This cross I, Khusro, king of kings, son of Khusro (have sent).
64Because 65of
the diabolic action and evil-doing of the most ill-fated Bahram Gushnasp and the
cavalrymen with him, we fled to Roman territory;
66and because the ill-fated
Zatsparham came with an army to Nisibis to incite the cavalry of the district of
Nisibis, we too sent cavalrymen
67with a commander to Charcha. And through the
fortune of the all-holy and famous saint Sergius, since we had heard that he was the
granter of petitions,
68in the first year of our reign, 69on 7th January, we petitioned
that, if (p.236) our cavalrymen should slay or capture Zatsparham, we would send a

jewelled cross to his home. And on 9th February they brought the head of Zatsparham
to us. Therefore, having gained our request, because each element was unambiguous,
we sent to his all-holy name this cross,
70which had come into our possession, together
with the cross sent by Justinian, emperor of the Romans, to his house, and which, in
the time of the estrangement of the two states, had been brought here by Khusro, king
of kings, son of Kavadh, our father, and was found among our treasures. (These
crosses we have sent) to the house of the all-holy saint Sergius.’ Gregory, having
received these crosses, with the approval of the Emperor Maurice, dedicated them
with much ceremony in the sanctuary of the martyr.
Persecution of Monophysites in the Roman East (590s)
Gregory then seized the opportunity to strengthen the Chalcedonian position in the frontier
region, where opposition to the council had always been strong. Khusro’s patronage of St
Sergius may have been loosening the natural bonds between Christians (especially Monophysites)
and Romans.
Evagr. HEVI.22 (238.22–8): Gregory too, after Khusro had given the crosses, visited
the wastelands of the (areas) called limites(i.e. frontiers), where the doctrines of
Severus extensively prevailed, with the approval of the government. He expounded
the doctrines of the church and brought into union with the Church of God many
garrisons, villages, monasteries, and entire tribes.
72(tr. Anon., revised)
Soon afterwards, in 598/9, stronger measures were taken by Domitian at Edessa and throughout
Mesopotamia to enforce adherence to the Council of Chalcedon; Monophysite sources allege
that the persecutions inflicted were brutal (Chr. 724, 145/17,
AG 910, Chr. 819, 10/76, AG
910,Chr. 1234, 82 [217–18/117–18], Mich. Syr. X.23 [386–7b/118 n.270]). 73Such policies
can hardly have endeared the imperial government to the inhabitants of the region, the majority
of whom were opposed to Chalcedon. As will be seen, when the Persians took over Osrhoene
and Mesopotamia, they were able to restore Monophysite bishops to their sees and thus gain
useful support.
Khusro’s continuing patronage of the shrine of St Sergius (593/4)
Even once secure on his throne, Khusro continued to honour the shrine of St Sergius. Between
spring 593 and the following spring he sent a further collection of gifts, including a gold paten,
to Sergiopolis, to thank the saint for the conception of his Christian wife, Shirin (Th. Sim.
V.14.1–12, Evagr. HEVI.21 [236.20–238.12]).
75Both Sebeos (85/29–30) and Tabari (1000/
314) emphasise Khusro’s toleration of Christianity, leading to legends according to which he
was even converted.
76The king also built three churches, which were consecrated by Anastasius,
the patriarch of Antioch (Chr. 1234, 81 [217/117], Mich. Syr. X.23 [387a/117 n.267], Chr.
Seert 58,PO 13.466–7, cf. Agap., PO8.447). See Labourt 1904: 209, Flusin 1992: II, 101–2.
At some point after the conclusion of peace, probably in 592, tension rose again between the
two powers. The cause was a raid by Arabs allied with the Romans on Pers
ian territory.

Th. Sim. VIII.1.3–8 (283.19–284.19):For this reason the Emperor Maurice des -
patched to Persia as ambassador George, who held charge of the tax collection of the
eastern cities; Romans call this man praetorian prefect. (4) And so Khusro, complain -
ing at what had happened, made trouble for the ambassador and brought about for
him a long delay in the barbarian country; (p.284) George therefore was harshly
treated for many days in Persia without gaining any access to the king. (5) Then, since
affairs for Khusro were still in confusion,
78the notion occurred to the barbarian,
(quite) reasonably, that he should not begin a war against the Romans for the
moment; then the Babylonian king admitted George into the palace. (6) And so
George, gaining the moment as his ally,
79persuaded the barbarian not to dissolve the
peace treaty. And so, in this way, Khusro ‘willingly yet with unwilling heart’, accord -
ing to the poem,
80welcomed quiet. (7) And so the ambassador reported in full to the
emperor all that had occurred, but his exposition of the conversation did not gain a
good conclusion. For in addressing the emperor George said: ‘The king of the Persians
stated these things in the hearing of the satraps: “On account of the ambassador’s
excellence, I grant deliverance from war.”’ (8) On hearing this, the emperor grew
angry with the ambassador, and the success of his mission became perilous for George.
For, in truth, speech that is not regulated by moderate (words) knows how to provide
great misfortunes for its practitioners. (tr. Whitby and Whitby, revise
Joh. Epiph. 1 (FHG IV.273):Book 1 of the Histories of John of Epiphania,
scholasticus and former prefect, concerning the attack by Khusro (II) on Maurice, the
Roman Emperor. The losses, which the Romans and the Persians bore and inflicted when they were
fighting each other during the reign of Justinian, the Roman Emperor, have been
related by Agathias of Myrina, a man who was outstanding amongst the lawyers in
Byzantium. Writing later than Procopius of Caesarea, he related what had been
achieved against the barbarians. Since this is the most momentous event of which we
have heard, I shall relate how the Persian king, after he had been expelled from his own
kingdom, became a refugee from his native country and approached the Roman state
and how he requested the Emperor Maurice to grant him an alliance and to restore
him to his kingdom. (I shall do this) not relying on any superiority of style or indeed as
a result of (my) earlier , but in order that so momentous an event should
not remain unknown to posterity, since even the greatest deeds, if they are not in some
way or other preserved in writing and handed down by memory, are extinguished in
the darkness of silence. For written accounts give life to events that h
ave passed. I did not consider it strange therefore to give, as far as I was able, an account of what
took place to those who did not know about these matters, since I had a part in some
of the events which took place and I had been in talks with Khusro the younger him -
self as well as with other very distinguished Medes; previously, as an assessorto Greg -
ory, the archbishop of Antioch, I happened frequently to have taken part with him in
conversations with them (the Persians) and later, after the end of the war, I also hap -
pened to go to Persia, accompanying George, who was making a concord for what had
81It is necessary for the understanding of (generations) to come to have an
accurate knowledge also of the main events that had happened previously, not least of

those relating to the rising against Hormizd, the father of Khusro. Therefore I think
that first I should recall briefly previous events, and then come to the remaining narra-
tive, so that those who (already) know should call to mind the disturbances, and that
those who have not heard (about them) at all should be able to obtain a clear know -
ledge of the origins from which the subsequent events arose. (tr. Donov
an, revised)
Embassies continued to journey between Constantinople and Ctesiphon, the majority of them
undertaken by churchmen. Thus in c.596/7 Probus, bishop of Chalcedon, visited the Persian
court and was well received by Khusro; the Persian king responded in kind, despatching Milas,
bishop of Senna, to Maurice (Th. Sim. V.15.8–11, Chr. Seert67,PO 13.494, 78, PO13.514).
Concrete political ends were served by some of these missions, for the Persian catholicos, who
was in regular correspondence with Maurice, asked the emperor to release captives he had taken
in Arzanene, Bezabde, Beth Arabaye and Singara in order to elicit a reciprocal gesture from
Khusro and to strengthen the good relations between the two rulers (Chr. Seert. 67,PO
83See Sako 1986: 113.
The increasingly prominent role of Christians in Persia may have intensified the disputes
which had always plagued them. After the catholicosSabrisho gave his support to Henana of
Nisibis against the city’s bishop, Gregory of Kashgar, the latter left the city. Around the year
599, the city rose in revolt and slew the Persian governor; how far this was connected to
disputes within the Nestorian Church is uncertain. Khusro soon quelled the uprising,
despatching a nakhveraganwith an army to retake the city. Promises of clemency made by
bishops sent with the army proved to be empty, and a massacre ensued ( Chr. Khuz.18–19,
Chr. Seert 75,PO 13.514).
Armenia and Iberia between the two powers (591–602) 85
The Armenian nobles, under Mushe³ Mamikonian, supported Khusro in his campaign against
Bahram. Although the king soon fell out with Mushe³ , who therefore left Armenia to serve the
Romans in the Balkans (Sebeos 81–4/25–9, 90–1/36),
86Khusro made great efforts to ingratiate
himself with the Armenians who remained in Persian territory. He was assisted in this by
Maurice’s wish to move large numbers of Armenians to the Balkans, a policy fiercely resisted by
most Armenians (Sebeos 86–7/31–2). See Goubert 1951: 204–6, Thomson and Howard-
Johnston 1999: 176–7. But Khusro’s attempt to exploit this anti-Roman feeling was thwarted
by the Armenians themselves, who robbed an official sent to win them over, and a joint
Roman–Persian force soon crushed this rebellion (Sebeos 87–8/32–4). See Grousset 1947:
256–7, Thomson and Howard-Johnston 1999: 176, dating the episode to 594–595. A further
rebellion in Roman Armenia was also put down by the Roman commander Heraclius after the
rebels had sought refuge in the mountains of Corduene (Sebeos 89–90/34–5).
87While Smbat
Bagratuni proved to be a loyal servant of Persian interests (on the north-east frontier of the
Sasanian kingdom),
88Atat Khorkhoruni, who was summoned to Constantinople in 601 to
serve in the Balkans, proved less obedient. He returned to Armenia while on his way to the
capital, and withdrew as far as Nakhchawan, in Persian territory. A Roman army was sent
against him, and even advanced to Nakhchawan, thus crossing the frontier; it retreated,
however, upon the approach of a Persian army (Sebeos 104–5/55–6). See Goubert 1951:
204–6, Thomson and Howard-Johnston 1999: 189.
89Maurice’s last effort to effect a large-scale

transplantation of the Armenians to the Balkans was foiled both by Atat’s flight and by his own
sudden death in the following year (Sebeos 105/56). 90
As we have seen, Maurice had restored Roman influence in Iberia through the appointment of
Guaram ascuropalatesin 588. But his descendants did not prove loyal, and in the early seventh
century Stephen I aligned himself with Khusro II in his attacks on the Roman empire (HVG
223/232). Iberian Christianity remained unaffected, however. Coins were issued in the late
sixth century, modelled on the silver drachms of Hormizd IV, but making use of Christian
images. See Toumanoff 1963: 389–90, 428–34, Lang 1983: 523. It was also in this period that
the Iberian Church broke from the Armenian, but remained in communion with the Church of
Constantinople. See Martin-Hisard 1998b: 1222–33, Garsoïan 1999: 3
The Roman view of the Persians and their army (late sixth century)
TheStrat¢gikon attributed to the Emperor Maurice offers a glimpse of Roman intelligence and
the methods devised to counter Persian tactics. Although the account is clearly hostile, the
description given reflects an awareness of Persian customs and methods. See Dagron 1987:
209–13, Lee 1993a: 103–4.
91Many of the observations made here correspond to what is known
of the Persian campaigns outlined in this work. 92
(Maurice), Strat gikonXI.1 (354–60): The Persian nation is wicked, dissembling,
and servile, but at the same time patriotic and obedient. The Persians obey their rulers
out of fear, and the result is that they are steadfast in enduring hard work and warfare
on behalf of the fatherland. For the most part they prefer to achieve their results by
planning and generalship; they stress an orderly approach rather than a brave and
impulsive one. Since they have been brought up in a hot climate, they easily bear the
hardships of heat, thirst and lack of food. They are formidable when laying siege, but
even more formidable when besieged. They are extremely skilful in concealing their
injuries and coping bravely with adverse circumstances, even turning them to their
own advantage. They are so intractable in negotiations that they will not initiate any
proposal, even one they regard as vitally important for themselves, but will wait until
the proposal is made by their opponents. They wear body armour and mail, and are armed with bows and swords. They are
more practised in rapid, although not powerful, archery than all other warlike nations.
Going to war, they encamp within fortifications. When the time of battle draws near,
they surround themselves with a ditch and a sharpened palisade. They do not leave the
baggage train within, but make a ditch for the purpose of refuge in case of a reverse in
battle. They do not allow their horses to graze, but gather the forage b
y hand. They draw up for battle in three equal bodies, centre, right, left, with the centre
having up to four or five hundred additional picked troops. The depth of the forma -
tion is not uniform, but they try to draw up the cavalrymen (p.356) in each company
in the first and second line or phalanx and keep the front of the formation even and
dense. The spare horses and the baggage train are stationed a short distance behind the
main line. In fighting against lancers they hasten to form their battle line in the

roughest terrain, and to use their bows, so that the charge of the lancers against them
will be dissipated and broken up by the rough ground. Before the day of battle a
favourite ploy of theirs is to camp in rugged country and to postpone the fighting,
especially when they know their opponents are well prepared and ready for combat.
When it does come to battle, moreover, especially during the summer, they make
their attack at the hottest hour of the day. They hope that the heat of the sun and the
delay in beginning the action will dampen the courage and spirit of their adversaries.
They join the battle with calmness and determination, marching step by step in even
and dense formation.They are really bothered by cold weather, rain, and the south wind, all of which
ruin the effectiveness of their bows. They are also disturbed by a very carefully drawn-
up formation of infantry, by an even field with no obstacles to the charge of lancers, by
hand-to-hand combat and fighting because volleys of arrows are ineffective at close
quarters, and because they themselves do not make use of lances and shields. Charging
against them is effective because they are prompted to rapid flight and do not know
how to wheel about suddenly against their attackers, as do the Scythian nations. They
are vulnerable to attacks and encroachments from an outflanking position against the
flanks and rear of their formation because they do not station sufficient guards in their
battle line to withstand a major flank attack. (p.358) Often, too, unexpected attacks at
night against their camp are effective because they pitch their tents indiscriminately
and without order inside their fortifications. To do battle against them our forces should be drawn up as prescribed in the book
on formations. Select open, smooth and level terrain, if you can do so, without any
swamps, ditches or brush which could break up the formation. When the army is pre-
pared and lined up for battle, do not delay the attack if you have really decided to fight
a pitched battle on that day. Once you get within bowshot make the attack or charge
in even, dense, regular order, and do it quickly, for any delay in closing with the
enemy means that their steady rate of fire will enable them to discharge more missiles
against our soldiers and horses. If it is necessary to fight on very rough ground, it is better not to have the whole
battle line on horseback in such places, but to draw some up in infantry formation
while others remain mounted. When lancers attack archers, as we have said, unless
they maintain an even, unbroken front, they sustain serious damage from the arrows
and fail to come to close quarters. Because of this they require more even ground for
such fighting. If the army is not really ready for combat, it must not engage in a
pitched battle. Instead, employ it safely in skirmishes and raids against the enemy,
which can be done smoothly on favourable terrain. Neither the enemy nor our own
troops should be allowed to discover the reason for putting off a pitched battle, since it
would embolden the one and make cowards out of the other. Wheeling or turning
around in withdrawals should not be directed against the enemy’s front, but to turn
up their flanks and take their rear. For as a result of their efforts not to break up their
formation while in pursuit, (p.360) the Persians easily expose their rear to forces
wheeling around against them. By the same token, if a force withdrawing before them
wants to turn about and attack the front lines of the pursuing Persians, it will suffer

injury on running into their well-ordered ranks. For the Persians do not attack in a
disorderly fashion as the Scythians do in pursuing, but cautiously and in good order.
For this reason, as has been said, forces wheeling about should not attack their front,
but should be sure to go by the flanks against their rear. (tr. Dennis,
rev. Burgess)181

Unrest in the Roman army in the Balkans led to Maurice’s downfall in 602. Phocas seized the
imperial throne in November that year, executing Maurice and all his family (with the possible
exception of his eldest son, Theodosius). As protocol had established, Phocas informed the
Persians soon afterwards of the change of ruler.
Th. Sim. VIII.15.2–7 (313.16–314.16): In the fifth month, 2having set out his own
proclamation in writing, the tyrant dispatched Lilius 3to Khusro, selecting him as the
messenger of his tyrannical election: for it was customary for Romans and Persians to
do this whenever they ascended to the royal might. (3) Lilius came to Dara, carrying
regal gifts. Germanus, who was adorned with the rank of the consuls and had received
charge of the army arrayed there, received him with exceptional splendour. (4) For
shortly before the time of the tyranny, when the Persian king Khusro had been roused
to hatred of
4the commander Narses, the Emperor Maurice had relieved the leader
Narses 5of custody of Dara (p.314) and had appointed Germanus instead; Maurice
wished thereby to calm the Babylonian ill-temper. (5) Now while Germanus and
Lilius were out riding, forming a pair, during the third day a military man struck the
mounted Germanus with a sword. (6) And so Germanus dismounted from his horse
and arrived home carried in a litter; but since the consequences of the sword-thrust
did not gain influence in a vital place, Germanus was cured of the blow in a few days.
Then, after he had been made better by medicines, he gave an exceptionally notable
feast for Lilius, and sent him on to Khusro. (7) And so Khusro exploited the tyranny as
a pretext for war, and mobilised that world-destroying trumpet: for this (tyranny)
became the undoing of the prosperity of the Romans and Persians. For Khusro
decided to pretend to uphold the hallowed memory of the Emperor Maurice. And so
in this way the Persian war was allotted its birth, and Lilius remained among the
Persians in great hardship. (tr. Whitby and Whitby, revised)
Sources from 603
With the end of Theophylact’s Historythe evidence available to us on the eastern front dimin -
ishes appreciably. The chronology of the following twenty years at least is difficult to determine
with any precision; nearly all dates must be regarded as approximate. A few contemporary

sources may be mentioned here, the most important of which are in Syriac. TheChronicle of
724 (also known as that of Thomas the Presbyter and referred to by Palmer [1993] as ‘a chron
icle composed about
AD 640’) contains some useful notices. See Hoyland 1997: 118–20. The
Chronicle of Guidi, also known as the Khuzistan Chronicle, composed slightly later in the
seventh century, furnishes a different perspective – that of a Nestorian Christian, living and
writing in what was, at the time of these events, Sasanian territory. Cf. Hoyland 1997: 182–5,
arguing for a date of composition not later than the 660s. The Chronicon Paschale, composed in
Greek in Constantinople around 630, contains some useful entries, particularly for events later
in the reign of Heraclius. See Whitby and Whitby 1989: xi, 190–1. But the source most
commonly used to construct an account of this period is the early ninth-century Byzantine
chronicler, Theophanes, supplemented by material in still later Syriac sources, such as the
Chronicle of 1234 and Michael the Syrian. Recent studies have shown, however, that all these
three sources, as well as the Christian Arabic chronicler Agapius of Manbij (Hierapolis), were
drawing much of their information on this period from a common source, identified convinc -
ingly as Theophilus of Edessa, who composed a chronicle around the year 750.
6The tenth-
century Nestorian work, the Chronicle of Seert, may well also have drawn on this common
For the history of this period, 8two obvious conclusions impose themselves. On the one hand,
insofar as by a comparison of these later sources we can arrive at some sort of reconstruction of
Theophilus’ chronicle, we can gain access to an earlier (and clearly quite well informed) source.
On the other hand, these various sources cannot now be seen as independent witnesses: if
Michael the Syrian and Theophanes agree on a certain event, this merely shows that they were
almost certainly here both deriving their information from a common source, Theophilus of
Edessa. No greater weight should be given to a date or event below simply because it is reported
by several of these sources.
We should also point out that we have offered few translations of the relevant Syriac sources,
since they are for the most part readily accessible in Palmer 1993. In Chapter 15 we have
offered, however, a translation of the whole first part of the Khuzistan Chronicle, of which none
(in English) is yet available.
Khusro’s invasion and the rebellion of Narses
The usurpation of Phocas gave Khusro an ideal opportunity to turn the tables on his former
allies and helpers. 10Having recently suppressed the rebellion of Bestam and quelled the uprising
in Nisibis, he was further assisted by the relative quiet of the north-eastern front since Bahram
Chobin’s successes. The defence of the Roman east was complicated by several factors: first,
opposition to Phocas. Narses, unhappy at his removal from Dara, at some point rebelled openly
against Phocas and seized control of Edessa. Second, Khusro sent someone purporting to be
Maurice’s son Theodosius with his invading armies, thus appealing to those of the population
who looked back with favour on Phocas’ predecessor. See Thomson and Howard-Johnston
1999: xxii.
Narses’ revolt broke out in late 603. See Stratos 1968: 59, PLREIII, Narses 10. It seems highly
unlikely (despite the report of Pseudo-Dionysius below) that he actively collaborated with
Khusro: when the Persians advanced westwards, he withdrew to Hierapolis (rather than unite
with the Persian forces). See Olster 1993a: 93.

Chr. 819(10.23–4), cf. Chr. 846(230.16–17): And in the following year ( AG 913 =
601/2) Narses entered Edessa, occupied it and there stoned the bishop Severus.12
Ps. Dion. 148.9–12: In the year 914 (602/3) Narses, commanding the army of the
Persians, crushed Edessa, entered it and captured Severus, the bishop of the city; and
he stoned him, and he (Severus) died. (tr. M. Greatrex)
Theoph. A.M. 6095 (291.27–292.1): Narses the Roman general rose up against the
tyrant and seized Edessa. And so Phocas wrote to the general Germanus to lay siege to
Edessa; Narses wrote to Khusro, the Persian king, to collect his forces (p.292) and
make war on the Romans. (tr. Mango and Scott, revised)
Khusro probably invaded Roman territory towards the end of 603. Whether Germanus was
sent to besiege Narses in Edessa is uncertain, since Chr. 1234(85, 220/121) and Michael the
Syrian (X.25, 390b/379) state that it was a certain John (Iwannis) who attacked and took the
city. See Olster 1993a: 93–4.
The siege and capture of Dara (604)
After a siege of at least nine months, Dara fell to Khusro in summer 604; Khusro must therefore
have begun his attack in November 603 at the latest 13(Chr. 724, 145/16, AG 915, Mich. Syr.
X.25 [390a/122 n.279],
AG 915, Chr. Khuz., 21).
The fullest account of the events of 604 is provided by Theophanes. The provenance of his
material is uncertain: it is not to be found in any of the Syriac source
Theoph. A.M. 6096 (292.6–25): ( … ) And Khusro, the Persian king, collected a
great force and sent it against the Romans. When Germanus heard of this, he was
afraid, but began the war under compulsion. When Germanus was wounded in
battle, his bodyguards brought him safely to Constantina; and the Romans were
14And on the eleventh day Germanus died. Phocas conveyed the forces
from Europe to Asia after increasing the tribute to the Khagan, thinking the Avar
nation to be at rest. Dividing the armies, he sent against the Persians and
the other part to the siege of Edessa, against Narses, under the command of the
eunuch Leontius, who was one of his magnates.
15Khusro collected his forces and
came against Dara. Narses departed from Edessa and took refuge at Hierapolis.
Khusro met the Romans at Arxamoun
16and, putting together a fort from (his)
elephants, began the war 17and won a great victory. He captured many of the
Romans and beheaded them. When these things had been done, Khusro returned to
his own land entrusting his forces to Zongoes.
18When Phocas learnt of this, he was
angered at Leontius and brought him ignominiously to Byzantium in iron fetters;
and he appointed Domentziolus, his own nephew, as general, and made him
19(tr. Mango and Scott, revised)
Further information on Phocas’ attempts to deal with the eastern frontier are provided by a con -
temporary source, the Life of Theodore of Sykeon composed by the saint’s disciple George; it is a

work of immense value as an account of life in Asia Minor in the period immediately preceding
the Persian invasions in the 610s. Theodore’s monastery lay close to the main road between
Constantinople and Antioch, not far from Ancyra, and he therefore received visits from many
important figures travelling to and from the East.
V. Theod.–11 (96): And not many days later the Emperor Maurice was
slain, and Phocas took over the kingdom. When his nephew Domitziolus became a
patrician and curopalateshe was sent by the emperor to the East to take over the army
and to marshal his forces against the Persian people, which was attacking and pillaging
our territories. The aforementioned illustrious man, when he had come as far as
21learnt of the raid of the Lazi as far as Cappadocia and of the plot of the
patrician Sergius, the father-in-law of the emperor, and was in much grief and fear,
not daring to complete the remainder of the journey.
Following the capture of Dara, Khusro took no further part in the campaigning. His place was
taken by Rasmiozan, a Persian general more commonly known by his title o
f Shahrvaraz. 23
Between 604 and 610 all the remaining Roman cities east of the Euphrates were taken. It is
impossible to arrive at any firm chronology of this conquest, and the following dates should
therefore be regarded only as approximate.
24The Persians, it is clear, were operating in both
Osrhoene and Mesopotamia, assaulting the Roman fortresses of the Tur Abdin (which had
never before fallen to them), such as Cephas, and Amida further north, and Resaina, Edessa and
Callinicum in the south. Given the large Roman expenditure on fortifications over the sixth
century, especially under Justinian, it should not occasion surprise that the conquest took so
long even in the absence of a Roman army in the field.
25The crumbling of the Roman defences
brought about a large emigration westwards, across the Euphrates ( Chr. 724, 146/17,
AG 922), 26
as it became ever clearer that Khusro’s war was going to be very different from that waged by his
grandfather. Roman resistance will have been further weakened by the capture and execution of
Narses in 604 or 605. SeePLREIII, Narses 13 and Stratos 1968: 60.
In either 606/7 (Chr. 1234, 86 [221/122],
AG 919, 27Mich. Syr. X.25 [391a/122 n.279], AG
918, cf. Jac. Ede. Chr.324/38) 28or 608/9 (Chr. 724, 146/17, AG 920, cf. Agap. PO8.449)
Mardin and Amida fell, followed by Resaina in summer 607 or 609. Cephas was captured six
months before Mardin, and upon learning of this, the Roman garrison at Mardin withdrew,
leaving the fortress’s defence to local monks (Mich. Syr. X.25, 390–1a/122 n.279). Thus both the
inability of the Romans to relieve the embattled cities and the tenacity of their inhabitants are in
evidence; but as the campaigns continued, their zeal waned. Khusro, no doubt in an effort to win
the support of the inhabitants of his newly conquered territories, restored the anti-Chalcedonian
bishops to their sees, removing those who supported the council. Some bishops who had been
expelled by Domitian in 598/9 and had sought refuge in Egypt now crossed into Persian territory
to resume their positions. Allegiance to their creed now took precedence over loyalty to the
empire, and the victories of the Persians were seen as a judgement on the beliefs of the
Chalcedonians (Chr. 1234, 89 [224/125–6], Geo. Pis. Contra Severum,PG92.1625, 47–55,
Mich. Syr. X.25 [389–91c/126 n.283], Barhebraeus, Chronicon ecclesiasticum, 263–7).
were no Roman counter-attacks, perhaps because of the rebellion of Heraclius, which began in
608; but even before then, Roman forces may have preferred to remain west of the Euphrates.
This war was thus already quite different from those of the sixth century: not only were there no
Roman counter-strikes, but the Persians clearly aimed at annexing the cities and territories they

overran. As theChronicle of 1234 mentions (86, 221/122), tribute was exacted; anti-Chalcedonian
bishops were installed; and, unlike previously, the Persians systematically captured every fortress east
of the Euphrates before proceeding further.
In 609/10 (Chr. 724, 146/17, AG 921, cf. Chr. Pasch. 698/149, a.609), Edessa, Carrhae,
Callinicum and Circesium were taken: the Persians were now clearly concentrating on
mopping up the remaining cities of Osrhoene. Some at least seem to have surrendered to the
Persians rather than face a protracted assault. The capture of Edessa – a city thought to be
impregnable, protected by Jesus’ promise to King Abgar – must have had a particularly crushing
effect on Roman morale.
31Then, on 7 August 610 (Chr. 724, 146/17, AG 921), Shahrvaraz
captured Zenobia, the first city west of the Euphrates to fall. See Stratos 1968: 63, Flusin 1992:
II, 74.
We may note in passing that a reference in the Syriac common source (taken up by
Theophanes, A.M. 6100 [296], Mich. Syr. X.25 [391–2a/125 n.282], Chr. 1234, 88 [224/
125], to which should be added Niceph. Call. XVIII.43 [416b]) to a Persian invasion of Asia
Minor in 610, penetrating as far as Chalcedon, is inaccurate.
Armenia, 601–10
The Persian annexation of Roman Armenia was as slow and systematic as that of Mesopotamia.
Here the Persians had to regain territory ceded in 591, and it was not until the end of Phocas’
reign that they were in a position to penetrate beyond Roman Armenia. By far the most impor-
tant source available for events in Armenia is Sebeos, whose chronology is unfortunately often
obscure. A certain amount of approximation is therefore inevitable in the dates here offered. See
Thomson and Howard-Johnston 1999: lxi–lxxvii, esp. lxxii.
In the winter of 603, while the siege of Dara was being pursued, a Persian commander named
Dµuan Veh was appointed to the command of Persian forces in Armenia. Setting forth from his
base at Dvin in the spring of 604, he fought an unsuccesful battle against the Romans at E³ evard
(near Erevan) (Sebeos 107–8/59).
35In the following year, Khusro appointed Datoyean to
replace Dµuan Veh, and under his leadership the Persians defeated the Romans at the village of
Getik to the west of the plain of Shirak. The Armenians, who had loyally supported the Roman
forces, were massacred after the battle in their fort at Erginay. The victorious Persians then
withdrew to Atropatene. (Sebeos 108–9/59–60).
36Datoyean was in turn succeeded by Senitam
Khusro (in 605/6), who outflanked the forces of Theodosius Khorkhoruni, the Roman
commander, at Ang³ (Anglon).
37The Romans tried to negotiate terms for a withdrawal, but
were surprised by a Persian attack and forced to flee. Theodosius was captured, and the Persians
continued to advance westwards. Following a further Persian victory in Basean, to the west of
Theodosiopolis, various forts fell into Persian hands. (Sebeos 109–10/60–2).
38The next Persian
commander in Armenia was Ashtat Yeztayar, who was accompanied by Maurice’s son (or
pretender) Theodosius. In 606/7 Ashtat defeated the Romans in Basean and pursued them as
far as Satala. He then returned eastwards, invested Theodosiopolis, and persuaded the city to
surrender by presenting Theodosius to the defenders. Other Roman fortresses fell soon after -
wards, including those of Citharizon, Satala and Nicopolis. (Sebeos 110–11/63).
Narratio de rebus Armeniae, 109–13 (p.41): After his (Maurice’s) death, the Persian
king Khusro took over the land of Armenia; (110) at that time 40the heretic Abraham
was established as patriarch. 41(111) And in the same year he obliged the bishops and

priests andhegoumenoi to anathematise the Council in Chalcedon and to depart from
the country. They pronounced the anathema and the controversy ended. (112) After
three years there was warfare in Phasiane (Basean) and the Persians slaughtered the
Romans, and the city Kitris (Citharizon) was besieged, as was Theodosiopolis, in the
fifth year of Phocas and the twentieth year of Khusro.
42(113) And they took many
other cities, in one of which was the catholicosJohn, who was in Armenia, but subject
to the Romans.
The Persians had now penetrated deep into Roman Armenia. Probably in the following year,
607/8, Shahin, the new Persian commander, defeated the Romans near Theodosiopolis and
expelled them from Armenia (Sebeos 111/64).
44That each year battles continued to be fought
seems to imply that the Romans were more willing (or better able) to defend Armenia than
45Two years later, in 609/10, perhaps after continuing Roman resistance, the
inhabitants of Theodosiopolis, including the catholicosJohn, were deported to Hamadan
(Sebeos 112/64). See Garsoïan 1999: 356–64.
46While the Armenian Church was able to regain
its unity and even prosper under Persian rule, the way now lay open for the invaders to push
deeper into Asia Minor. See Garsoïan 1999: 368–84.
The continuing Persian advance (610–12)
It was not until the autumn of 611 that the Persian advance was pursued further. Their progress
beyond the Euphrates was greatly assisted by the unrest which had broken out in Syria and Pal-
estine in the wake of Heraclius’ revolt in Africa and the invasion of Egypt by his ally Nicetas in
608. The outbreak of rebellion sparked off conflicts between circus factions, leading to the
intervention of imperial troops, and in 609 or 610 to the murder of the patriarch of Antioch,
Anastasius II. According to many sources, the Jews were heavily involved in the fighting,
although whether as faction members or as opponents of Christians is unclear.
responded by appointing Bonosus as comes Orientisin 608 or 609 to repress the violence, and he
exacted savage retribution on the Green faction in Antioch. But in 609 he was forced to move
into Egypt to oppose Nicetas’ forces advancing from the west. Although initially victorious, he
was defeated by Nicetas outside Alexandria in late 609.
49Heraclius overthrew Phocas and was
crowned emperor in October 610, but this did not mark the end of his struggle for power.
Comentiolus, one of Phocas’ brothers, and the commander of a Roman army in the East,
refused to accept the new ruler and undertook a rebellion at his winter quarters in Ancyra late in
610. Heraclius, whose régime was still insecure, quickly opened negotiations, first through a
monk named Herodian, and then through the former magister militumPhilippicus. Probably
early in the new year the revolt came to a swift end with the assassination of its leader by Justin,
the commander of the Armenian forces under Comentiolus.
While the details of Heraclius’ revolt and seizure of power are not strictly relevant to our theme,
it is clear that they had a considerable impact on the ability of the Romans to resist further
Persian attacks. Several points may be noted. First, and most obviously, Roman forces which
might otherwise have been available for the defence of Syria and Palestine were taken by
Bonosus to Egypt, thus leaving the field open to the Persians. Other troops which might have
defended Armenia withdrew to Ancyra in 610. This would help to explain the willingness of
cities such as Antioch to capitulate to the invaders. Second, the bloodshed between 608 and 610

was considerable. Not only were there fewer soldiers left in the Levant after Bonosus’ departure,
but those civilians remaining were weakened after the strife and reprisals, and no doubt reluc-
tant to engage in further conflict.
51Divisions must have remained within the populace, and thus
some at least will have welcomed the opportunity afforded by an enemy attack to take revenge.
The Persian armies were therefore faced with a weakened and divided enemy, and against this
background their rapid and sweeping conquests become more comprehensible
The accession of a new emperor failed to persuade Khusro to end hostilities. According to
some sources, Heraclius asked for peace upon gaining power, but Khusro refused, killing the
Roman ambassadors (Sebeos 113, Chr. 1234, 91 [226/127], Mich. Syr. XI.1 [403a/128
n.287], Agap., PO8.450, Chr. Seert 82,PO 13.527).
53The Persian advance westwards
continued in earnest in 611; the preceding year had perhaps been spent clearing the southwest
bank of the Euphrates (the Roman province of Euphratesia), where Roman strongholds such as
Zenobia and Sergiopolis may still have been holding out.
54In Armenia, meanwhile, the depor -
tation of the population of Theodosiopolis was being carried out. The Persian onslaught in 611 was two-pronged. In the north, Shahin proceeded to Caesarea
in Cappadocia, which he captured without difficulty: the Christians, according to Sebeos, evac -
uated the city, while the Jews welcomed the invaders. Here, however, the Romans made an
attempt at a counter-strike. Probably in autumn 611 Priscus, whose attempt to take command
in the East in 589 had proved so disastrous, was sent to surround the Persians in Caesarea. The
blockade proved effective, but in summer 612 the Persians managed to escape back to Armenia
(Sebeos 113/66, Nic. 2.9–22, Theoph. A.M. 6103 (610/11), 299, Chr. 1234, 91 [226/127]).
Heraclius, who had visited Priscus during the siege, therefore decided to take over command of
the eastern front in person, something no emperor had done since Valens in the 370s.
V. Theod.–154.9 (123–4): When now the disorder of the army had been
brought to an end and had reached a peaceful outcome, not much afterwards the Per-
sian people made an incursion as far as Caesarea, the metropolis of Cappadocia. We of
the monastery and all in our territory were in great fear, suspecting that they would
venture even as far as us. And when the servant of God was called upon to move us
from the monastery because of this great fear, he prayed to God and said to us, ‘Do not
fear, children, the onrush of the (Persian) people. For God has been called upon by me
and will not allow them to come against our territory; nor shall I see with my eyes a
foreign invasion here.’ After this, Priscus, the most glorious patrician and comesof the
excubitors, went off with an army against the Persians; (and) again a rumour arose
that they had surrounded (p.124) this people in Caesarea and that they were dying of
hunger and would either swiftly surrender or be destroyed. To these things the servant
of Christ said, ‘None of these (things) will come about, but they will completely
escape and depart. And if we do not turn back to God and repent so that he may be
reconciled with us as with the Ninevites, they will come again with a great force and
lay waste all the land as far as the sea. However, confident in my God, I say to you that
in my lifetime God will not abandon my old age nor will a barbarian incursion take
place in my time.’
56And this was the case by the grace of God. (154) There came to
him letters from the most pious Emperor Heraclius and the patriarch Sergius (to urge)
him to come to the capital city so that they might receive his blessing. He therefore
went out from the monastery and reached the capital. It happened that before his
arrival the emperor departed and went off to Caesarea to join forces with the

protopatrician Priscus and our army which was with him, against the Persians, with
the result that the holy man was thenceforth obliged by the patriarch Sergius to await
the return of the emperor.
In the south, the Persian advance under Shahrvaraz was swift. Its progress is difficult to follow
with precision, however. Perhaps in May 611 Apamea and Antioch fell, followed by Emesa,
which surrendered after receiving a guarantee of safety. Again, there was some Roman resis-
tance: in a battle at Emesa, following the city’s capture, a Roman army was defeated (Chr. 724,
AG 922, cf. Chr. 1234, 91 [226/127], Theoph. A.M. 6102 [609/10], 299, Agapius, PO
8.450). 58At both Caesarea and Emesa it is interesting to note that it was only after the fall of a
city that Roman forces intervened – perhaps because it was not until mid-611 that Heraclius
was able to organise a response. Clearly the Persians were driving southwards towards Palestine,
a region famous for its wealth.
The Roman counter-strike (613)
Although Caesarea had fallen to the Persians in 611, the important fortress of Melitene still
remained in Roman hands. It was against this city therefore that Shahin directed his army,
by way of Theodosiopolis, in 613; and, having captured it, he proceeded to join forces with
Shahrvaraz (Sebeos 113/66).
60At this point Heraclius embarked on a bold counter-attack,
having elevated his infant son Heraclius Constantine to the rank of Augustus. Philippicus,
succeeding to Priscus’ command as he had 24 years earlier, led the Roman army at Caesarea
eastwards into Roman, then Persian, Armenia, penetrating as far as the province of Ayrarat.
Khusro was therefore obliged to recall his armies, which hastened to deal with the Roman
incursion. Philippicus then withdrew to Roman territory via Theodosiopolis, while the
Persians, wearied by the forced marches undertaken to overtake the Romans, turned
instead to Mesopotamia (Sebeos 114/67–8).
61Philippicus’ campaign was probably merely a
diversion, allowing Heraclius, his brother Theodore, and Nicetas to combine forces in Syria.
There, in a closely fought battle, they were finally defeated by the invaders (Sebeos 114/67–8,
Agap. PO8.450).
62See Pernice 1905: 68–9, Baynes 1914: 36–7 (placing the episode in 611),
Thomson and Howard-Johnston 1999: 205–6.
V. Theod. Syk. 166 (153–4):During the days of Lent, when the Emperor Heraclius,
beloved of Christ, was going down from the royal city to Antioch of the East, to mar -
shal his forces against the Persians, he came up to the monastery to get his (Theo -
dore’s) blessing. The blessed man arose and met him at the entrance of the church of
the holy martyr George; and they embraced one another, and went into the church of
the glorious martyr and into the church of the Archangel, and the blessed man gave
him a blessing, commending him to God. And he gave gifts of blessing to him (p.154)
– fine flour, apples and choice wine – and urged him to taste (them). But the emperor,
because of (his) great haste, declined to taste them and to receive the gifts bestowed on
him, saying ‘Keep these things for me, father, and pray for us. And on my way back I
shall come and then take these things, and shall stay here as long as you bid; and I shall
enjoy at great leisure both your gifts and your all-holy prayers.’ But the blessed man
said to him, ‘Now listen to my exhortation, child, for on your return you will not find

me. I have a great road (to travel) and I intend to be away.’ But the emperor said to
him, ‘God will so arrange (things), holy father, that I will find your holiness here and
will feast with you
63at leisure.’ He then received his (Theodore’s) blessing and went
off to the city of Antioch, in which he began (his) war against the Persians, (together
with) Nicetas the patrician and comeswho was there also. The blessed man grieved
that he had left behind his gifts, saying, ‘If he had taken them, it would have been a
token of his victory, and he would have returned with joy. His leaving these things
behind is a sign of our defeat, and if he had not come up and received the blessing of
the saints, then great sorrow would have overtaken him and all of us.’ Having sat
down for a short time in the chapel of the Mother of God, bowing (his head) and
becoming thoughtful, he soon raised (his head) and said to me, ‘Know, George, (my)
child, that by the command of God this emperor will be long-lived, for he will reign
for at least thirty years.’ And indeed it occurred in accordance with
his word.
The Romans withdrew northwards after their defeat, closely followed by the Persians. In
another engagement fought near the Cilician Gates, the Romans, although initially victorious,
were defeated. The Persians then took over the province of Cilicia (Sebeos 115/68). See Stratos
1968: 106–7, Kaegi 1973: 329, Flusin 1992: II, 79.
Once the Romans had been pushed back from Syria, the way southwards, towards Palestine and
Egypt, lay open to the invaders. Later in 613 the city of Damascus was captured by Shahrvaraz.
Like Emesa, it surrendered to the Persians, agreeing to pay tribute; according to Theophanes,
however, many prisoners were taken (Chr. 724, 146/17,
AG 924, Chr. 1234, 91 [226/128],
Mich. Syr. XI.1 [401a/128 n.287], Theoph. A.M. 6105, 300). See Stratos 1968: 107–8, Flusin
1992: II, 79 and n.54.
The fall of Jerusalem (614)
The following year, Shahrvaraz’s drive southwards continued. From Damascus he proceeded
southwards to the vicinity of Bostra and Adraa.
Tabari, I, 1007/327 (Nöldeke 300): 64Qaysar (Caesar) sent a man called Qatma with
an army of Romans 65and Khusro sent one with Shahrvaraz, and they fought a battle
by Adhri‘at (Adraa) and Busra (Bostra), and it is the nearest part of the al-Sham to you
66The Persians joined with the Romans, and the Persians overcame them. The
unbelievers of Quraysh disliked it; then God revealed ‘Aliph lam mim’ … 67(tr.
Qur’an, 30.2–5: Aliph Lam Mim. (3) The Romans have been defeated (4) in the
nearest part of the land, but after their defeat they will be victorious (5) after many
years – government belongs to Allah before and afterwards – and on that day the faith -
ful will greatly rejoice. ( … ).
Passing through Phoenice and Galilee, Shahrvaraz first directed his march to Caesarea, on the
Mediterranean coast; from there he proceeded to Jerusalem via Diospolis (Lydda) (Sebeos 115/
68, Ant. Strateg. 3 [503], V. Georg. Khozeb.VIII.33 [133]).
68Although the new patriarch of

Jersualem, Zachariah, was inclined to come to terms with the invaders, the circus partisans
insisted on resistance. Roman troops were summoned from nearby Jericho, but fled upon
seeing the number of Persians surrounding the city (Ant. Strateg. 5–7 [504–5]).
69The Persian
siege, which began in mid- or late April, lasted twenty days. When the city walls fell, a massacre
ensued (Sebeos 115–16/69, Ant. Strateg. 8 [506–7]).
70A mass grave of Christians recently
uncovered near the Pool of Mamilla has been associated with the capture of the city. See Reich
1996: 26–33, 60.
The contemporary evidence of Sophronius shows that resistance at Jerusalem was stiff neverthe -
less. Cf. Hist. Her. 429/228 with Mahé 1984: 223. See Clermont-Ganneau 1898: 36–7 and
Baynes 1914: 198.
Sophronius, Anacr.14.69–102 (105–7): On top of the ridges of the walls stood the
population, (like Mount) Olympus, (so that) all the children and wives felt (only)
trivial concerns. (p.106) O Christ, grant (us) to see Persia burning soon instead of the
holy places! Possessed of a steadfast mind, they kept the approaching Mede from the
strong walls by showers of missiles and rocks. Then indeed, with raging spirits, the
Persian, barbarian that he was, after thousands of clashes, employed siege engines.
Having set fires everywhere beneath the wall, (as well as) an army of siege engines,
destroyed the strong wall and came to be inside the city. Equipped with bloody sword,
he cut down the people – the city of sacred and holy old men, children, and women.
(p.107) O Christ, may you curb by the hands of Christians the ill-fated children of
impious Persia! Accomplishing everything with cruelty, he despoiled the holy city and
with blazing fire burnt the holy places of Christ. After shouting an insult to God, who
had once died there, he despoiled the sacred spoil and with the spoil marched (off). O
Christ, may you curb by the hands of Christians the ill-fated children of the dreadful
parent, Persia!
Most sources report that the Jews collaborated with the Persians, both before and after the siege,
and took the opportunity to settle scores with their Christian oppressors. Many inhabitants,
including the patriarch, were deported to Persia (Ant. Strateg. 9–18 [507–11]). Outside Jerusa -
lem, the Jews may have undertaken some campaigns in Palestine themselves, laying siege to
Ptolemais and Tyre according to Eutychius (ed. Breydy, 121–2/101–2).
72Churches in the
vicinity show signs of destruction which have been associated with this episode. See Schick
1995: 27–9, Russell 2001: 47. The Persians for their part withdrew to Damascus by way of Jeri -
cho after the capture of the city (Ant. Strateg. 15 [510], V. Georg. Khozeb.VI.29 [127], VII.31
[130]). See Schick 1995: 21.
Both Palestine and Syria also suffered at the hands of Arab raiders at this time. The collapse of
the frontiers will have meant that normal arrangements between the Romans and neighbouring
tribesmen will have broken down. With no subsidies on offer, both tribes hitherto allied to
Rome and their opponents may have sought to profit from the unsettled conditions. See Sartre
1982a: 193, Kaegi 1992: 52–6, Shahîd 1995: 640–1.
General accounts of the fall of Jerusalem, with widely varying casualty figures (from 36,500
to 90,000) are to be found in numerous sources; the various versions of Antiochus Strategius’
account also offer differing figures. Mention is generally made of the seizure and removal of the

True Cross from Jerusalem; Nicetas succeeded in salvaging the sacred sponge and the spear
which had pierced Christ’s side (Ant. Strateg. 23 [515–16], Theoph. A.M. 6106 [613/14],
300–1,Chr. 724, 146/17,
AG 925,Chr. 1234, 93 [226–7/128], Mich. Syr. XI.1 [403–4a/400],
Chr. Pasch. 704–5, Tabari, I, 1002/318 [Nöldeke 291]). 75
TheLife of George of Khozeba offers a glimpse of the panic in the region at the time of the
siege of Jerusalem. The Persians probably withdrew northwards past Jericho and Bostra. See
Vailhé 1901: 646, Baynes 1914: 199, Di Segni 1991: 144.
V. Georg. Khozeb. VII.30 (127.19–129.13): When the Persian invasion had then
reached (p.128) as far as Damascus, there was no inconsiderable confusion in this
country. At one (moment) therefore our holy father – this George – while sitting on a
rock and warming himself by the ray of the sun (for he was withered through excessive
abstinence), (and) completely burning with the longing of spiritual love for the fulfil -
ment of the divine will, called upon merciful God, imploring him with tears to spare
his people. And a voice came to him: ‘Go to Jericho and you (will) see the works of
men.’ And he arose and finding some from the monastery going down to Jericho,
went down with them. When they arrived at the gardens near the city, suddenly he
heard in the air a great sound of crowds (of people) throwing one another into confu -
sion, striking and crying out, as if in combat. Raising his eyes to the air, he beheld it
filled with some Indians attacking one another as though under arms, and striking
(one another) as in war.
77And the earth shook and trembled beneath him. The broth-
ers said to him: ‘Come here, father, (and) let us go to the city. Why have you been
standing there for so long, gazing at the sky?’ But he said to them with tears and lam-
entation, ‘Let us flee, brothers, and return (home). Do you not see and sense the earth
shaking?’ And when he had said this, lo, suddenly some armed men on horseback
came forth from the city, and other young men on foot, and youths wearing greaves,
and spears were in their hands, running hither and thither.
78(p.129) And the brothers
realised that this was the shaking of the earth of which the old man had spoken. They
returned to the monastery in great fear, for he reported to them the vision in the air.
When the old man came back to his cell, he grieved and lamented the coarseness of the
people rather than their foolishness or impiety. Then, going outside, seated on a rock
(facing) towards the ray of the sun (for this was his task on account of the slightness of
his body), he called upon and besought God, saying: ‘Lord God of mercies and Lord
of pity, who wants all to be saved and to come to the recognition of the truth, raise
your rod and strike your people, because they wander in ignorance.’ At once he beheld
a rod of fire in the air stretching from the holy city (Jerusalem) to Bostra.
79And the
holy man knew that the people would be heavily punished; and he grieved and
lamented over all these things.
The Persians soon afterwards gave permission for the rebuilding of the city, and the patriarch of
Alexandria, John the Almsgiver, even sent materials to assist in this (Leontius, V. Ioh. Eleem.
80Extensive reconstruction took place after 617, when the Persians barred further Jews from
migrating to the city. Zachariah’s replacement in office, Modestus, raised funds from Chris -
tians throughout the region, while according to the Khuzistan ChronicleKhusro’s minister

Yazdin contributed generously to the rebuilding (Sebeos 116–18/70–2,Chr. Khuz.27). See
Flusin 1992: II, 174–7, Schick 1995: 41–4 and Thomson and Howard-Johnston 1999: 208–9. 81
Many refugees from Palestine sought refuge in Egypt at this time, where they received a kind
reception from the patriarch. 82
Anon., V. Ioh. Eleem. 9:After Rasmiozan (Shahrvaraz), the governor – or rather the
chief general – of Khusro, the Persian king, had sacked the holy places of Jerusalem,
this grievous report came to the ears of the thrice-blessed patriarch. When he heard of
this outrageous and audacious deed, and learnt that all the holy places had been com -
mitted to flames, he sat down and composed a lament as though (he were) one of the
inhabitants of (those) places. He lamented the desolation of these (places) not for one
day, nor two nor twenty or twice this amount, but for a whole year, grieving and
groaning bitterly as he struggled to surpass in his laments Jeremiah, who had once
lamented the capture of this city. This lament he did not compose (to put it simply)
just as it occurred (to him), so that it might be consigned to oblivion, but rather he is
said to have committed it to writing. In addition to this he sent a certain man,
Ctesippus by name, who at that time ran the monasteries of Ennaton,
83a man most
beloved of God, to see the destruction of the holy places in Jerusalem. And through
him he sent a great quantity of gold, a mass of grain, wine, oil and beans, and he pro-
vided clothes for lay-people and monks, and a large amount of food for the sake of the
sick, as well as very many beasts of burden for the transport of these t
He did not concern himself only with those captured in the cities, but also cared
with great attention for those in monasteries, and above all those in convents, who had
suffered in this way. Of these (monastic buildings) a great number had been destroyed
by the Persians, and about a thousand of the nuns had been taken captive; he (John)
sent a great amount of gold on their behalf and, having taken a record of them, he
restored them all to monasteries. The governors of the Persians, when they learnt of
the exceeding charity and sympathy of the man who was in truth well named ‘the
almsgiver’ – for even the foe respected the virtue of the man – they greatly desired to
see him; they provided money for Dion, who happened to be governor at the time, so
that he might go off and see him. Furthermore, he despatched Theodore, the bishop of Amathus, to rescue those
who had been taken prisoner by the Madieni,
85as well as Anastasius, the superior (of
the monastery) of the mountain of Antony the Great, and Gregory, the bishop of
86Through them he effected the rescue of very many men and women
captives by ransoming them in return for the payment of a great quantity
of gold. 87
The Persians at Chalcedon (614–15)
The capture of Jerusalem imbued the Persians with still greater confidence. In the summer or
autumn of 614 Shahin set off from Ctesiphon, bound for the heart of the Roman empire;
among his troops was a certain cavalryman, Magundat, who during the campaign deserted the
army and made his way to Hierapolis. By the end of 614 or early 615 Shahin had reached

Chalcedon on the Bosporus, and proceeded to invest the city; elements of his army may well
have been despatched to devastate other parts of Asia Minor, since archaeological evidence
establishes that Ephesus was destroyed in 614 (Acta Anast., ch.8,Chr. Pasch.706, Theoph.
A.M. 6107 [614/15], 301, Nic. 6, Sebeos 123/79, Tabari, I, 1002/319 [Nöldeke 292]).
Roman response was two-fold. First, the tactics already employed in 613 were repeated:
Philippicus undertook another incursion in the East, and thereby forced Shahin to abandon his
siege and follow him (V. Anast. ch.8).
89Second, taking advantage of the proximity of the
Persians, Heraclius sought to open negotiations to end the war. The emperor visited Shahin
outside Chalcedon, and the Persian general received him courteously (Nic. 6, Sebeos 122–3/
78–9, Thom. Arts. 89–91/156–8). Three Roman ambassadors were therefore sent by the
Senate to Khusro, and Shahin withdrew from the city (Chr. Pasch. 706, Nic. 7, cf. Sebeos
122–3/78–9). The text of the letter borne by the ambassadors is preserved in the Chronicon
Paschale. See Stratos 1968: 115–17, Thomson and Howard-Johnston 1999: 211–12.
Chr. Pasch. 707.1–709.24: God, who created all things and maintains (them) with his
own power, has bestowed a gift worthy of his own goodness on the race of men – provi -
dential care for the kingdom, by means of which we have been considered worthy to live
without disturbances, or, falling upon certain difficulties, we find a solution. Giving
thought to this divine thing – we mean the royal care – and before anything else, we
beseech your very great Clemency to consider us worthy of forgiveness, since we have
dared to undertake the present appeal to your might contrary to the previous system
which operated. For we know the custom which prevailed in previous times, which
ordained that when some strife arose between the two states, the rulers of each of them
would resolve the points at issue through reports (presented) to one another. But
Phocas, once he became a traitor to the Roman state, broke this arrangement. For,
having gradually corrupted the Roman army in Thrace, he suddenly attacked our royal
city and slew Maurice, who piously reigned over us, and his wife, as well as his children
and relatives and many of those in charge. And he was not content with the great evils he
had committed, but also did not pay fitting honour to your very great Clemency, with
the result that then you, aroused by our errors, brought the affairs of the Roman state to
such dire straits.
91(p.708) Our current piously reigning emperor and his ever-memo -
rable father, realising the things done by that corruptor, resolved to free the Roman state
from the excessive violence of that man. And this they did, discovering (then) how it
(the state) had been humbled by your might. And after the death of the tyrant, when our
emperor wished to take his own relatives (with him) and return to his own father in
Africa, urging us to select the man we wished as emperor, he was only just persuaded by
our entreaties to accept the kingdom. Because of the confusion which prevailed in the
two kingdoms, as well as the civil strife,
92he did not have the opportunity, as should
have been done, to pay the honour owed to the very great might of your Serenity
through an embassy.
93We therefore resolved to disregard the custom which we
mentioned above, and, being mere men, to employ entreaty towards so great a supreme
king, despatching some of our number who should be considered worthy of your foot -
steps. But because of what has been happening in the meantime, we have not dared to
do this until now. When, however, Shahin the most glorious Babmanzadag,
commander of the Persian army, was in the region of Chalcedon and met our most

pious emperor and us, and was asked by all to hold talks about peace, he said that he
himself did not have such authority, but that he required (authority) about this matter
from (p.709) your Clemency. And now he has sent an answer to us through the
95promising under oath that your very great might will receive those sent
by us as is fitting, and will release them unharmed to return to us, and that he was
ordered by your Beneficence to do this. Encouraged again by this sequence of events,
and above all by God and by your greatness, we sent your slaves Olympius, the most
glorious ex-consul, a patrician and praetorian prefect, and the most glorious Leontius,
a patrician and city prefect, and Anastasius, the most God-beloved priest and syncellus,
who, we beseech you, may be received as befits your very great might;
96and (we entreat
that you) restore to us quickly and safely peace, which is pleasing to God and which
befits your peace-loving might. We beg your Clemency to hold our most pious
Emperor Heraclius as (your) genuine child,
97for he eagerly wishes to do the service of
your Serenity in all things. In doing this, you gain a double glory for yourself, both for
your valour in wars and for the gift of peace. And henceforth we shall enjoy peace
through your eternally remembered gifts, taking the opportunity to offer prayers to God
on behalf of your long good health, and holding your good deed in remembrance
through the ages of the Roman state.
Neither Roman initiative bore fruit. The ambassadors were executed by Khusro, while Shahin
returned to Chalcedon, having presumably dealt with Philippicus’ incursion, and captured the
city (either in late 615 or 616) (Theoph. A.M. 6108–9, 301, Mich. Syr. XI.1, 404a/128 n.289
[year 7 of Heraclius], Nic. 7.20–2).
Consolidation of the Persian position (615–18)
Little is known of specific campaigns in this period; we are reliant for the most part on the evi-
dence of numismatics and archaeology in reconstructing the history of these years. Sardis was
destroyed in 616,
99and in the same year the Roman mint at Cyzicus ceased operation. But a
Roman army continued to hold out in Seleucia in Cilicia Tracheia, as is clear from the presence
of a mint there between 616 and 617. In the following year (and perhaps later) coins were
minted instead at Isaura Vetus in Isauria, away from the coast;
100the cause of this retreat may
have been the construction of a Persian fleet, since it seems that Constantia (Salamis) came
under attack around 617. Thanks to the intervention of the patriarch of Alexandria, John the
Almsgiver, who was in Cyprus at the time, the situation was resolved pea
Anon., V. Ioh. Eleem. 13:Hearing of the utter destruction of the Roman state at the
hands of the Persians, he (the patriarch John) wished to journey to the emperor and to
be an ambassador concerning a peace. And indeed he put together a valedictory
speech and addressed everyone, (but) it was not accepted by the people that he should
leave the city. When therefore the Persian armies had utterly devastated Syria,
Phoenice and Arabia and still other cities, the evil men threatened to take Alexandria
itself. At that point indeed he became aware through God of a deadly plot being
hatched against him, and he sailed away to his fatherland, Cyprus. A certain general,

Aspagourius by name, after being sent against Constantia in Cyprus and not being
received by those in the city, 102 had armed himself for war against them, and they in
turn had armed themselves against him. They intended to proceed at once to close
quarters for a great slaughter of one another, had not the disciple of peace, the all-
admirable John, anticipated (events), reconciled (them) and turned both sides to
peace, bringing them together.
If it was the Persians who were attacking Constantia, they do not appear to have gained control
of the whole island. 103Instead, they concentrated their efforts on the invasion of Egypt, the rich -
est province of the eastern empire. Shahrvaraz probably began his assault in 618, passing by
Pelusium on the well-trodden invasion route into the country.
Heraclius’ measures (615–19)
The situation in Constantinople itself was fraught; according to Nicephorus (8), Heraclius even
considered abandoning the capital and moving to Carthage instead. 105With the devastation of
Asia Minor by the Persians and the assaults on the Roman Balkan provinces, and even Greece,
by the Avars and Slavs, the economic resources of the empire were shrinking alarmingly. The
city of Thessalonica survived a siege by the Avars in c.618, but only with difficulty.
106 See
Haldon 1990: 43–4, Foss 1994: 49, Whittow 1996: 76–7. Desperate measures were called for,
even as the Persian attack on Egypt removed the one remaining significant source of grain and
wealth. Heraclius therefore introduced a new silver coin, the hexagram, in 615. This allowed
official salaries to be halved when paid in the new coin, which was also inscribed on the reverse
with the appropriate entreaty Deus adiuta Romanis(‘God help the Romans’) (Chr. Pasch. 706,
a.615). See Grierson 1968: 17–18, 101, 270–4 with pl.10, Yannopoulos 1978: 2–8, Hendy
1985: 494.
107At the same time, the weight of the copper folliswas reduced from approximately
11 grams to 8 (and still further later in the war). 108In 618 a charge of 3 follesper loaf of bread was
introduced at Constantinople; hitherto it had been free. Later in the same year the distributions
were suspended altogether (Chron. Pasch. 711 [a.618]). See Whitby and Whitby 1989: 164
n.449, Stratos 1968: 100–1.
The Persian conquest of Egypt (618–19)
Probably because Egypt was now cut off from the remainder of the Roman empire, the Roman
sources offer no account of the fall of Alexandria. 110The most complete account is to be found
in the Khuzistan Chronicle (see Chapter 15 below), according to which the city was betrayed by
a certain Peter; that in Severus’ History of the Patriarchs of Alexandria (PO1.484–5) is of little
111Leontius’ Life of John the Almsgiver (52, Dawes and Baynes 44b) further informs us
that Nicetas and the patriarch were able to flee the city before its fall, probably in June 619 (Chr.
724, 146/17,
AG 930). 112The Persians gradually extended their control of Egypt southwards; by
mid-621 the whole province was in their hands. See Altheim-Stiehl 1991, 1992a (noting that
the Persians were already in Oxyrhynchus in January 620), 1998 (a convenient summary). The
serious economic consequences of the province’s fall have already been noted. As elsewhere,
following the destruction which accompanied the invasion, a period of peace and even recon -
struction followed.
113We may note also that copper dodecanummiaminted in Alexandria under
the Persians feature a bust of Khusro between a sun and a crescent (common Sasanian motifs on

coinage) on the obverse, and a cross on the reverse: here too the king was clearly prepared to
conciliate local opinion. See Grierson 1968: 233–4, 336–8 with pl.18 and Flusin 1992: II, 127.
The Persian conquest of Asia Minor (619–22)
The noose around Constantinople now tightened further; nor was there any obvious source of
hope for those remaining in the capital. The Balkans continued to suffer devastation at the
hands of Avars and Slavs.
114In Asia Minor the last functioning mint – that at Nicomedia –
ceased operating in 619. See Hendy 1985: 416. Perhaps in 620, but more probably in 622, the
important central Anatolian city of Ancyra fell into Persian hands. A Persian fleet was also
active, and the island of Rhodes fell to the invaders in 622/3 (Chr. 724 147/18,
AG 934,
Theoph. A.M. 6111 [618/19], 302, Chr. 1234, 96, 230/133, Mich. Syr. XI.3, 408a/133 n.300,
Agap. PO8.458).
115Evidence for Sasanian activity along the Aegean coast is also provided by a
hoard of coins discovered on the island of Samos, probably buried in 623. See Oeconomides
and Drossoyianni 1989: 163–75.
It should be noted, however, that the Persians did not have a secure hold of all of Asia Minor:
Roman troops were clearly able to assemble and operate in many provinces there later in the
620s without hindrance. Archaeology similarly confirms that some cities at least, such as
Aphrodisias, were spared destruction.
116In general the Persians seem to have preferred to lay
waste Asia Minor, while keeping Palestine, Syria, Mesopotamia and Egypt more or less intact
under their own administration. See Morony 1987: 91–2, Flusin 1992: II, 95, Kaegi 1992: 45.

The source material for the final phase of the last Roman–Persian war is more extensive than for
the first phases. Theophanes offers a continuous account of several campaigns, and the
Chronicon Paschale provides similarly detailed information. The poems of George of Pisidia
commemorate certain episodes during the war, but tend to focus more on the emperor’s per -
sonal involvement in masterminding the Roman recovery. It has been plausibly suggested that
behind all three sources, as well as (Pseudo-)Sebeos’ notices, lie campaign reports by Heraclius
and his staff, sent to the capital from the front in order to boost morale and to reassure anxious
citizens. See Howard-Johnston 1994: esp. 72–8 and Thomson and Howard-Johnston 1999:
lxviii, M. Whitby 1994 and 1998. Speck 1988 offers a more sceptical perspective.
1For the
other sources used, see the notes on pp.182–3.
The campaign of 622 2
In the spring of 622 Heraclius was at last ready for a counter-offensive against the Persians. A
costly peace had been agreed with the Avars in 620, and Heraclius had therefore transferred the
Roman forces in the Balkans across to Asia Minor in the following year (Theoph. A.M. 6112,
3In 622 the emperor, with the backing of the patriarch Sergius, took over the can -
delabra and vessels of the church and melted them down to make gold and silver coins (Theoph.
A.M. 6113, 302.34–303.3, Nic. 11.21–3). See Haldon 1984: 171–2 and Whitby and Whitby
1989: 158–9 n.441 (accepting the date of 622 for this measure).
Parast. synt. chron. 42.7–10:And after the same Phocas was burnt, the ox 5was
melted down by Heraclius for the treasury of the guards (skoulkatameion) and in order
to cross over to Pontus for the sake of raising troops; for there was a guardpost
(skoulkaton) in Pontus.
On 5 April Heraclius left Constantinople, entrusting the city to the patriarch Sergius and the
general Bonus as regents for his young son Heraclius Constantine. On the following day he
reached Pylae, on the gulf of Nicomedia, by sea. He then assembled his forces somewhere in
Asia Minor, probably in Bithynia (Geo. Pis. Exp. Pers.I.154–II.11, Theoph. A.M. 6113,
302.32–4, 303.6–12). See Oikonomides 1976: 2, Howard-Johnston 1999: 3 and n.11,

Thomson and Howard-Johnston 1999: 213. 7The size of this combined army is unknown, but
it clearly must have been the largest force fielded by the Romans in the recent past: Theophanes
rather vaguely states that Heraclius ‘collected his forces and added a new army to them’
(303.11–12), while George of Pisidia specifically mentions the diverse nature of the forces
assembled by the emperor (Exp. Pers. II.165).
8But before he could lead them into battle, it was
necessary to revitalise their broken morale and to increase their effectiveness. Military exercises,
described by George of Pisidia (followed by Theophanes), were held; the image of Christ, not
made by human hands,
9was displayed to the troops; and Heraclius roused his troops for battle
by emphasising the religious nature of the struggle against a barbarian who had plundered
the holy sites of Christianity (Geo. Pis. Exp. Pers.II.12–202, Theoph. A.M. 6113,
Probably in July Heraclius proceeded to Armenia, where he inflicted a defeat on an army
led by a Persian-allied Arab chief (Geo. Pis. Exp. Pers.II.203–34, Theoph. A.M. 6113,
11The emperor was prevented from pushing further east by the arrival of the
Persian general Shahrvaraz, who had wintered in Pontus and quickly seized control of the
passes leading east from Armenia (Geo. Pis. Exp. Pers.II.256–60).
12Heraclius then outma -
noeuvred his opponent, prompting him to try to deflect the emperor by invading Roman
territory, moving south through the Cilician Gates (Geo. Pis. Exp. Pers.II.261–344,
Theoph. A.M. 6113, 304.20–5).
13Shahrvaraz soon reversed his plan, fearing an invasion of
Persia through Armenia, and took to shadowing Heraclius’ army. An attempt at a night
attack by the Persians was abandoned, but soon afterwards, probably in early August, the two
sides finally met in an open battle from which the Romans emerged victorious (Geo. Pis. Exp.
Pers. II.345-III.304, Theoph. A.M. 6113, 304.25–306.7).
Almost immediately after this victory Heraclius was forced to leave his army to winter in
Armenia, while he returned westwards to deal with the renewed Avar threat (Theoph. A.M.
6113, 306.7–8).
15It is doubtful whether Heraclius had done much to reverse the Persian tide in
the East by his victory, but the effect on morale of the first Roman success of the war must have
been considerable. See Oikonomides 1976: 7–9, Howard-Johnston 1999: 4.
Distractions in the west (623)
Continuing pressure from the Avars obliged the emperor to arrange a meeting in June with
their ruler, the khagan, at Heraclea in Thrace. Heraclius got only as far as Selymbria before
learning that the Avars planned to ambush him. He was forced to flee, while the invaders
penetrated as far as the walls of Constantinople. Nevertheless peace terms were agreed, by
which the Romans agreed to pay 200,000 solidia year to the Avars and to hand over high-
ranking hostages (Chr. Pasch. 712.12–13.14, Nic. 10, 13.1–9, Theoph. A.M. 6110,
The Persians meanwhile continued to whittle away at Roman possessions in Asia Minor.
Shahrvaraz captured Ancyra, while a Persian fleet seized Rhodes and deported the island’s
inhabitants (Theoph. A.M. 6111, 302.22–3, Chr. 724, 147/18,
AG 934,Chr. 1234, 96, 230/
133, Mich. Syr. XI.3 [408a/133 n.300], Agap., PO8.458). 17However at the same time,
perhaps as a result of Heraclius’ successes in the previous year, Khusro started to take a harder
line in his newly conquered territories, as well as in his whole kingdom generally. In particular,
he seized the wealth of the churches of Mesopotamia and Syria, and increased taxes to provide

funds for his continuing campaigns (Theoph. A.M. 6112, 302.25–7, A.M. 6116, 314.23–6,
Chr. 1234, 96, 230/133, Mich. Syr. XI.3 [408a/408], Agap.PO8.451, 458). 18
Heraclius’ first counter-offensive in the north-east (624)
On 25 March 624 the emperor left the capital with his new wife Martina and his two children
by his previous marriage. They celebrated Easter on 15 April in Nicomedia, from where he
and Martina set off for Caesarea in Cappadocia (Chr. Pasch. 713.19–714.8, Theoph. A.M.
6114, 306.19–21, Sebeos 124/81).
19 From Caesarea Heraclius proceeded north, first to
Theodosiopolis, then eastwards to Dvin in Persarmenia, which he sacked, and then southeast to
Nakhchawan (Geo. Pis. Her. II.160–72, Sebeos 124/81, Thom. Art. 92–3/159).
20At the same
time, Shahrvaraz continued to campaign in Roman territory, from where he was recalled
because of the emperor’s incursion (Theoph. A.M. 6113, 306.21–3,
Theoph. A.M. 6114, 307.19–308.25: So the emperor, taking up his army straight -
away made for inner Persia, burning the towns and villages. And there happened at
this stage an awesome miracle. For at the time of the summer solstice the air became
damp, refreshing the Roman army with the result that fair hopes revived them. When
Heraclius heard that Khusro was in the town of Ganzak
22with 40,000 fighting men,
he rushed against him. He sent forward some of his subject Saracens as an advance
party and they encountered the watch of Khusro, some of whom they killed, whilst
others they captured and brought to the emperor together with their commander.
When he had learnt of this, Khusro, abandoning the town and his army, took to
flight. Heraclius gave pursuit, and some he overtook and killed, whilst the rest escaped
and scattered. And when the emperor reached the town of Ganzak (308) refreshed his army in its suburbs. The Persians who had taken refuge with him said
that Khusro had destroyed with fire all the crops in those parts on his way to the town
of Thebarmais>
23in the east, wherein were the temple of Fire and the treasure of
Croesus, king of the Lydians, and the deceit of the coals. 24Having seized these items,
he (Khusro) had marched to Dastagerd. 25Setting out from Ganzak, the emperor
reached Thebarmais; and he entered (it) and burnt down the temple of Fire and burnt
down the entire city; and he followed hard behind Khusro into the defiles of the land
of the Medes.
26 And in this difficult terrain Khusro went from place to place.
Heraclius, as he was pursuing him, plundered many towns and lands. When winter
had set in, he deliberated as to where he should winter together with his army. Some
said that they should do so in Albania, others that they should push ahead against
Khusro himself. The emperor ordered that the army should purify itself for three days,
and then, opening the holy Gospel, found a passage that directed him to winter in
27So immediately he turned back and hastened to Albania. As he had with
him numerous Persian captives, he was the object of several attacks by the Persian
armies on the intervening journey, but with God’s help he gained a victory against all
of them. In spite of the severe winter cold that overtook him on the way, he arrived in
Albania with 50,000 captives whom, in his compassionate heart, he pitied and liber -
ated. He granted them proper care and repose so that all of them prayed with tears (in

their eyes) that he should become the saviour even of Persia and slay Khusro, the
destroyer of the world. 28(tr. Mango and Scott, revised)
Movs¢s Daskhurants‘i II.10 (130.3–132.5): Now since, in accordance with his
desire, he (Khusro) had successfully imposed his will over all peoples and kingdoms,
and had become so powerful and behaved so arrogantly and believed that he had
derived his formidable and wonderful kingdom through his own deeds of valour, he
did not comprehend that the Most High is lord of an earthly kingdom and He gives it
to whom He wishes. So he began gradually to grow weak and to lose strength before the king of the
Greeks and he was not able to raise his head in accordance with his former power. For
the emperor suddenly notified all his commanders and officers of God’s favour in
front of him. Straightaway he ordered (them) to assemble in one place with all the
strength which was under their control. Everyone responded to the assembly that had
been arranged. No man had time for his friend, but without delaying they hastily beat
their ploughshares into swords and their scythes into spears. The weak and the mild
were encouraged saying, ‘We are strong and warriors.’ Having arisen with his whole
army, he himself became commander and leader in front of his forces. Leaving the
court of his palace in the hands of his son, he fastened a crown on him and established
him upon the throne of his kingdom in his place. He did not strike against the Persian forces, which had surrounded and blockaded
his countries and cities, and were holding them in submission, and he did not pass
near (p.131) to them and he did not provoke them to battle. Rather, having left them
there in his land, he came and went across the sea and having fashioned a route
through the land of the Egerians, he went through Armenia and passed across the river
Araxes. And he planned to take the great king Khusro unprepared.
And when it was reported to Khusro, he marvelled in his mind and said, ‘Is this not
he who was driven into an abyss from fear of me? What now is this?’ And he took
flight from his face, from the fortresses on the borders of Media; he went from there to
the country of Asorestan. And immediately he sent swift-arriving couriers and wrote
to his great commander Shahrvaraz very great oaths and threats. ‘My great dishonour
and wrath will be expiated if in this way you are able to hasten to arrive and not to
spare a single man or beast out of those who have dared (to come) into my presence.’
And the commander undertook the command, having read and heard the terrible
news. Straightaway he reviewed all the Persian forces. Having left in the control of gar -
risons the cities of the Romans and the Palestinians which had become subject to (his)
control, carefully he charged them to hold (them) until his return from the matters
which had arisen. And he himself set in motion the forces with chosen armed men,
and with swift horses he made haste to accomplish the command of the kin
g. Now the great emperor Heraclius, when he saw that the Persian king had fled from
his face, refrained from pressing after him. And he raided (p.132) through the regions
of Atropatene as far as the place called Gayshavan, a fortified place chosen by the Per -
sian kings in the months of heat for its healthy coolness, which is on the borders of the
country of Media. He plundered, ruined and enslaved the whole land and turned

from there; he intended to winter in the districts of the countries of Albania, Iberia
and Armenia. (tr. Greenwood)
Stalemate in the Transcaucasus (625)
Over the winter Heraclius sought to win support from the Transcaucasian kingdoms; he also
entered into negotiations with the Turks to the north of the Caucasus. See Howard-Johnston
1999: 17.
Movs¢s Daskhurants‘i II.10 (132.5–21): 30Consequently he (Heraclius) composed a
letter to the princes and leaders of (the) countries, (stating) that they themselves
should go out willingly to meet him, and that they should receive him and serve him
together with his forces in the days of winter. But if not, they would be reckoned
before him like heathens, their strongholds would be taken and the regions of those
countries would be enslaved by his forces. And when all the heads and princes of this
country of Albania heard this, they abandoned the great city of Partaw to the same
person (Heraclius) at the command of Khusro, (and) departed and secured themselves
in various places.
31Many of the Christians and heathens, skilled inhabitants of the
city, who by reason of their incapacity and infirmity were not able to escape and flee
from their presence, remained there in the city. A certain priest, named Zachariah, a holy man, who was a monk of the church of
Partaw, a mild and quiet man, placed himself over them and through oaths and vari-
ous means, he caused many Christian persons to be saved; through his prayers, he
acted as their surety. (He did this) also for the sake of Jews and heathens. Conse-
quently afterwards his action was praised and having been attested by everyone, he was
appointed to the office of chief bishop of the see of Albania. (tr. Gre
Heraclius spent most of the campaigning season of 625 in the Transcaucasus, eluding three
Persian armies sent to overtake him, and seeking to defeat them individually. In addition to
Theophanes and Movs¢s (below), Sebeos (125–6/81–4) provides important details concerning
the campaigns. See Manandjan 1950: 138–44, Stratos 1968: 159–63, Howard-Johnston 1999:
17–18, Thomson and Howard-Johnston 1999: 215–16 and Zuckerman 2002
Movs¢ s Daskhurants‘i II.10 (132.21–133.11): Now when the force of Romans
arrived, a countless, immense multitude camped in the province of Uti, beside the
torrent which was inside the boundaries of the village of (p.133) Ka³ ankatuk.
trampled and ruined the fine properties of vineyards and village possessions through
which they passed. And travelling from there, they camped beside the torrent Trtu
near to the village of Diwtakan.
Then the Persian force, which they called ‘New Force’, came and caught up with
them. And its commander was Shahraplakan. 34And from the faithful nobles of the
king there was an inspector and governor among them, one man whom they used to
call Granikan Sa³ ar; he came and descended upon him (Heraclius). The other
commander of the Persians came from Rome and turned Heraclius backwards and

drove him out through the country of Siwnik‘. 35For although very great losses had
occurred to the Persian forces, nevertheless they displaced, drove away and threw him
(Heraclius) into his own country and reoccupied the cities which they had taken from
him by force. (tr. Greenwood)
Theoph. A.M. 6115 (308.27–312.8): In this year Khusro, king of the Persians,
appointed as his commander Shahraplakan, an energetic man puffed up with great
vanity; and having entrusted him with an army (consisting of) the so-called
Khosro¢getai and (p.309) Perozitai, sent (him) against Heraclius in Albania. Having
advanced to the peaks of Albania, they did not dare face the emperor in battle, but
seized the passes that led to Persia, thinking to trap him. At the beginning of spring
(625) Heraclius set out from Albania and made (his) way towards Persia through level
plains that provided an abundance of food, even if, by this lengthy route, he was
covering a great distance. Shahraplakan, on the other hand, pushed ahead by the
narrow and shorter way so as to anticipate him in the territory of Persi
Heraclius exhorted his army, saying: ‘Let us be aware, O brethren, that the Persian
army, as it wanders through difficult country, is weakening its horses and debilitating
them. As for us, let us hasten with all speed against Khusro so that, falling upon him
unexpectedly, we may throw him into confusion.’ The troops, however, did not con -
sent to do this, especially the Lazic, Abasgian and Iberian allies. For this reason they
fell into misfortune. For Shahrvaraz, too, had arrived with his army, whom Khusro
had armed in full strength and sent against Heraclius by way of Armenia. (As for)
Shahraplakan, he was following Heraclius from behind and did not engage him,
expecting, as he did, to join Shahrvaraz and in this way give battle. When the Romans
had been apprised of the approach of Shahrvaraz, they turned to cowardice and fell at
the emperor’s feet, repenting with tears of their misguided disobedience; for they
knew how great an evil it is for a servant not to yield to his master’s wishes. And they
said: ‘(Stretch out) your hand, O lord, before we miserable ones perish. We obey you
in whatever you command.’ Then the emperor hastened to engage Shahraplakan
before the latter had been joined by the army of Shahrvaraz and, having made many
sorties against him both by night and by day, reduced him to a state of timidity.
Leaving both of them to his rear, he pushed on with all speed against Khusro. Now
two of the Romans deserted to the Persians and persuaded them that the Romans were
fleeing out of cowardice. Another rumour had also reached them, namely that Shahin,
the Persian commander, was coming with another army to (their) aid. (p.310) When Shahraplakan and Shahrvaraz learnt this, they strove to engage
Heraclius in battle before Shahin arrived and transferred to himself the glory of vic -
tory. Trusting also the deserters, they moved against Heraclius and, when they drew
near to him, encamped, wishing to engage him in the morning. But Heraclius set out
in the evening and marched all night; and when he had gone a long distance from
them, he found a grassy plain and encamped in it. The barbarians, thinking that he
was fleeing out of cowardice, pushed on in a disorderly manner so as to overtake him.
But he met them and gave battle. Having occupied a certain wooded hill and gathered
his army there, he routed the barbarians with God’s help and slew a multitude of them
after pursuing them through the ravines. 203

in his back.> 37As these struggles were going on, Shahin also arrived with his army, and
the emperor routed him and slew many of his men, whilst the rest he scattered as they
were fleeing; and he captured their camp equipment. Shahrvaraz then joined forces
with Shahin and gathered together the barbarians who had survived. And, once again,
they made plans to move against Heraclius. (As for) the emperor, he pushed on to the
land of the Huns
38and their difficult terrain, through rough and inaccessible places,
(while) the barbarians followed him from behind. Now the Lazi, together with the
Abasgi, took fright; they severed themselves from their alliance with the Romans and
returned to their own country. Shahin was pleased at this and, together with
Shahrvaraz, eagerly marched on against Heraclius. The emperor, having gathered his
army, encouraged his men and spurred them on with advice, saying: ‘Brothers, do not
let the multitude disturb you. For when God wills it, one man will
rout a thousand. So let us sacrifice ourselves to God for the salvation of our brothers.
May we win (p.311) the crown of martyrdom so that the time to come will praise us
and God will grant (us) our reward.’ Having encouraged the army with these and
many other words, he arranged the battle order with joyful countenance. The two
sides stood a little distance from each other from morning until evening, but did not
engage with each other. When evening had fallen, the emperor kept to his march; and
again the barbarians pressed on behind him. Wishing to overtake him, they changed
route, but fell into marshy ground, went astray and came into great danger. So the
emperor passed through the regions of Persarmenia. That country being under Per-
sian control, many men joined Shahrvaraz and increased his army. And when it was
winter, the multitude was dispersed in their own lands so as to take rest houses>.
39Heraclius, learning of this, planned to steal a battle by night. And when the
winter, then, had set in, and Shahrvaraz was not suspecting anything, he selected the
strongest horses and the bravest soldiers of the army and divided them into two; the
first part he ordered to move ahead against Shahrvaraz, whilst he himself followed
behind with the rest. So they hastened through the night and reached the village
Salbanon at the ninth hour of the night.
40The Persians who were there became aware
of the attack: they rose up and rushed against them (the Romans). The Romans slew
all of them, except one, who informed Shahrvaraz. Shahrvaraz, rising up and mount -
ing his horse, naked and unshod as he was, obtained his salvation by flight. Heraclius
overtook and destroyed by fire his wives and the flower of the Persians, i.e. the com -
manders, satraps and picked soldiers, after they climbed onto the roofs of their houses
and were preparing to fight; some he slew, some he burnt, whilst others were bound in
fetters, so that nearly no one escaped except for Shahrvaraz. (p.312) They took the
arms of Shahrvaraz, (namely) his golden shield, his dagger, spear, gold belt set with
precious stones, and boots. When Heraclius had taken these things, he moved against
the men scattered in the villages. These men, on learning of the flight of Shahrvaraz,
(also) fled without restraint. He pursued them, killed or captured many of them,
whilst the remainder returned to Persia in disgrace.
41(As for) the emperor, he joyfully
collected his army and wintered in those parts. (tr. Mango and Scott, r
Sebeos 125–6/81–3 offers more geographical precision than Theophanes. Shahrvaraz moved

from Nisibis, through Media, to P‘aytakaran in Albania. He and Shahin, with 30,000 troops,
nearly surrounded Heraclius near Gardman, east of Lake Sevan. Heraclius eluded both of them,
first moving south to Nakhchawan and then northwest into Bagrevand. With an élite force of
20,000 men, the emperor inflicted a crushing defeat first on Shahrvaraz’s vanguard (of whom
only one survived) and then on the main body of the Persian forces at Archesh (in the battle
described by Theophanes, immediately above). Shahrvaraz escaped, while Heraclius continued
west for the rest of the winter.
42See Thomson and Howard-Johnston 1999: 215–17 (noting
probable use of the same source by Theoph. and Sebeos) and Howard-Johns
ton 1999: 18.
The turning of the tide (626)
While Heraclius continued to outmanoeuvre Khusro’s armies, the Avars were advancing
towards the imperial capital. Shahrvaraz too reached Chalcedon and entered into communica -
tion with the Avars. But the city, already strongly fortified at the emperor’s orders, stoutly
resisted the attacks of the Avars and Slavs in late July and early August. See Bari¬i¨ 1954, Stratos
1968: ch.14, Howard-Johnston 1995b.
The Persians, lacking a fleet, were unable to do any more than observe.
43It is possible,
however, that they deliberately failed to intervene: according to Theophanes (A.M. 6118,
323.22–324.16), Mich. Syr. XI.3 (408–9a/408–9), Chr. 1234, 98, 232–3/136–7, and
Agapius (PO 8.461–2),
44Khusro wrote to the second-in-command at Chalcedon, Kardarigan,
ordering him to kill Shahrvaraz and lead the army back to Persia. The letter was intercepted
by the Romans, and Heraclius Constantine invited Shahrvaraz to Constantinople to show it
to him. The general therefore entered into an alliance with the Romans, altered the letter to
state that Khusro wanted 400 commanders killed, and by making it public secured the
backing of his army against the king. Whether because of such an arrangement, which may
well have been fabricated when Shahrvaraz entered into friendly relations with Heraclius in
629, or because he was threatened by the Roman army which had defeated Shahin (see
Theoph. below), Shahrvaraz’s forces then withdrew from Constantinople. See Mango 1985:
106–9 (accepting the alliance) and Speck 1988: 144–52, 293–8 and Howard-Johnston 1999:
19–22 and n.68 (rejecting it).
Theoph. A.M. 6116 (312.19–314.23): In this year, on the 1st of March, the emperor
Heraclius collected his army and took counsel as to which road he should follow: for
two roads lay before him, both narrow and difficult, one leading to Taranton, the
other to the land of Syria.
46And whereas the one to Taranton was superior, it lacked
every kind of food supply, whereas the one to Syria that went over the Taurus
provided a plentiful abundance of food. Everyone gave preference to the latter, even
though it was more precipitous and covered with much snow. Having traversed it
with great toil, in seven days they reached the river Tigris, which they crossed and
arrived at Martyropolis and Amida.
47Both the army and the captives rested (there).
From there the emperor was able to send letters to Byzantium (p.313) and to set out
everything from his point of view, and to cause great joy in the City. Shahrvaraz,
having collected his scattered army, went after him. The emperor picked a band of
soldiers and sent them forth to guard the passes leading to him; and sallying forth to
the eastward passages,
48he came face to face with Shahrvaraz. Having crossed the

Nymphius river, he reached the Euphrates, where there was a pontoon bridge made of
rope and boats. Shahrvaraz, having untied the ropes from one bank, shifted the whole
bridge to the other.
49When the emperor came and was unable to cross by the bridge,
he went by (it) and found a ford which he safely traversed – an unexpected feat in the
month of March – and so reached Samosata. Once again he went over the Taurus and
arrived at Germaniceia; and, going by Adana, he came to the river Sarus.
Shahrvaraz stretched the bridge back to its former place and, crossing the Euphrates
without hindrance, followed him from behind. The emperor crossed the bridge of the
Sarus and, finding an opportunity to rest his army and horses, encamped there and
gave them rest. Shahrvaraz (then) reached the opposite bank. Finding the bridge and
its forward bastions occupied by the Romans, he encamped. Now many of the
Romans made disorderly sorties across the bridge and attacked the Persians, among
whom they caused much slaughter. The emperor forbade them to sally forth indis -
criminately in case a way might open for the enemy to enter the bridge and cross it at
the same time they did, but the army did not obey the emperor. Shahrvaraz set up
ambuscades and, feigning flight, drew many of the Romans to cross over in pursuit of
him against the emperor’s wish. He then turned round and routed them, (p.314) and
killed as many as he overtook outside the bridge – (and so) they paid the price for their
disobedience. When the emperor saw that the barbarians had broken ranks in pursuit
and that many of the Romans who were standing upon the bastions were being slain,
he moved against them. A giant of a man confronted the emperor in the middle of the
bridge and attacked him, but the emperor struck him and threw him into the river.
When this man had fallen, the barbarians turned to flight and, because of the narrow-
ness of the bridge, hurled themselves into the river like frogs, whilst others were being
killed by the sword. The bulk of the barbarians, after being scattered to the edges of
the river, shot arrows and stood firm, preventing the Romans from crossing. The
emperor did cross to the other side and bravely opposed the barbarians with a few men
of his guard, fighting in a superhuman manner so that even Shahrvaraz was astonished
and said one Cosmas (a runaway Roman and an apostate) who was standing
close to him: ‘Do you see, Cosmas, how boldly the Caesar stands in battle, how he
fights alone against such a multitude and shrugs off blows like an anvil?’ For he was
recognized by his particular boots, and received many blows, although none serious nature in this battle. After they had fought in this battle all day,> when evening
came, they drew apart. Shahrvaraz became frightened and retreated through the
night. As for the emperor, he collected his army and hastened to the city of Sebastea.
After crossing the river Halys, he spent the whole winter in that land.
51(tr. Mango and
Scott, revised)
Since Heraclius will have reached Sebastea in April or May 626, Theophanes’ final statement
above must be rejected. His next entry in fact continues to recount events from the same year.
Theoph. A.M. 6117 (315.2–26): In this year Khusro, king of the Persians, made a
new army by conscripting strangers, citizens and slaves, raising the levy from every
nation. He placed this levy under the general Shahin and, taking another 50,000 men

chosen from the phalanx of Shahrvaraz, attached them to him. He called them the
Golden Spearmen and sent them against the emperor. (As for) Shahrvaraz, he des-
patched him with his remaining army against Constantinople, so that after establish -
ing an alliance between the western Huns (whom they call Avars) and the Bulgars,
Slavs and Gepids, they would advance on the City and lay siege to it. When the
emperor learnt of this, he divided his army into three contingents: the first he sent to
protect the City (Constantinople); the second he entrusted to his own brother Theo -
dore, whom he ordered to fight Shahin; the third part he took himself and marched to
52During his stay there he invited the eastern Turks, whom they call Khazars,
into an alliance. 53Now Shahin with his newly recruited army overtook the emperor’s
brother and prepared for battle. With God’s help (through the entreaties of the all-
praised Mother of God), when battle was joined a storm of hail fell unexpectedly on
the barbarians and struck down many of them, whereas the Roman array enjoyed
calm weather. (So) the Romans routed the Persians and slew a great multitude (of
54When Khusro learnt of this, he was angered at Shahin. And Shahin, because
of his great despondency fell ill and died. By order of Khusro his body was preserved in
salt and conveyed to him; and he subjected the corpse to many outrages. (tr. Mango
and Scott, revised)
The following extract from an account of the life and miracles of St Theodore the Recruit has
been connected with the victory of Heraclius’ brother. See Howard-Johnston 1995b: 134.
Zuckerman 1988: 206–10 associates it rather with Heraclius’ campai
gn of 622.
V. S. Theodori, ch.11 (53): Miracle three. For while the Persians remained before the
city (Euchaita), 55they were suddenly hit by a Roman expeditionary force. Boiling
with anger, they slaughtered many of the captives with the sword and set fire to the
city and the shrine of the holy man. But they were not quite able to escape (him),
(and) the warrior martyr overtook them. They had not yet marched far (from the city)
when another column of Roman soldiers attacked them on the mountain called
Omphalimus. They (the Romans) destroyed many, (while) a mass of stones sent from
the sky, like hailstones, killed the others by divine judgement when they reached the
river called Lycus.
56Thus none of those who had done such things returned to his
lands. ( … ) (tr. Zuckerman, revised)
In the eastern Transcaucasus, Heraclius’ contacts with the Turks began to bear fruit in the same
year: an invasion of Albania took place in mid-626, and the yabghu khagan demanded from
Khusro that he withdraw from Roman territory. See Baynes 1914: 665–6, Howard-Johnston
1999: 21.
Movs¢s Daskhurants‘i II.12 (140.17–143.20): Then, after that, in the thirty-sixth
year of Khusro (625/6), Caesar Augustus decided to contemplate ways how perhaps
he might be able to be cured from the terrible sorrow and dishonour. And during the
whole (p.141) period of his kingship he united to himself the whole Roman army
which he summoned to his assistance, in order to breach the great mount Caucasus,

which shuts in and dominates the regions of the north-east and in order to open the
gates of Ch‘or 57to draw out various peoples of barbarians and through these to drive
out the king of the Persians, the proud Khusro. Then he prepared and briefed one of
his nobles (whose) name (was) Andr¢,
58a highly talented and intelligent man, and
sent him with many promises of innumerable and countless treasures; ‘On condition
that they shall assist me out of great envy (of Khusro), I shall undertake to satisfy the
thirst of the savage, gold-loving people of long hair.’ Now when the heir to the king of the north, who was second to his kingship and
named Jebu Khak‘an (yabghu khagan),
59heard this, and saw the promises of very
valuable presents in return for plundering in a raid all the countries which were
under the control of the Persian king, then with great eagerness he gave a reply
saying, ‘I shall take revenge against the enemy. I shall leave and arrive in person as an
assistant to him (Heraclius) with my valiant forces, and I shall please his mind with
military engagements, with my sword and bow, just as his person desires.’ Then he
sent with the same noble, in order to satisfy the purpose of that agreement, select
mounted men, powerful in strength and skilled archers, in number about one thou -
sand. They rushed suddenly against the gates of Ch‘or and did not pay attention to
the militia and garrison of the king of Persia, who were arranged at the great gate.
Rather, swooping like (p.142) eagles beside the great river Kur, sparing no one who
came against them, they took a route through the country of Iberia and Eger; they
cut across the great sea as far as the royal palace.
60And being in the presence of the
great emperor Heraclius, they affirmed to each other oaths in accordance with the
custom of each one. Having received from him an order concerning their issuing
forth, they returned from there on the same route to their country, nothing being
suspected by anyone. Now at the entering of the thirty-seventh year of the same Khusro (626/7),
king of the north sent the promised fighting force, appointing his brother’s son as
commander, whose princely title they give (as) Shat‘. When he came, he galloped
through all the borders of this country of Albania and through a part of Atropatene.
He put to the edge of the sword many individuals, Christian and heathen alike. Who
will be able to discover and set down in writing the number of those who were made
prisoner by them? When they had encamped beside the river Araxes, he despatched messages to the
great king Khusro, revealing their uniting with the emperor and their coming to his
assistance. And the transcripts of the messages to one another are as follows. ‘If you
will not turn your face from the king of the Romans and yield to him all the countries
and cities which you have seized through your violence, and (if) you will (not) des -
patch all the prisoners taken from his land, which you have at present under your con -
trol, together with the wooden cross which all (p.143) Christian peoples worship and
glorify, and (if) you will (not) summon outside his borders all your forces, the king of
the north speaks in this way, the lord of all the earth, your king and (king) of all kings,
“I shall set my face against you, you governor of Asorestan, and in place of one evil
which you employed against him, I shall pay you back double. I shall move against all
your borders with my sword in the same way that you moved with your sword against

his borders. I shall not release you, and I shall not hesitate to act against you according
to that statement which I have relayed to you.”’Then when the great Khusro heard all this, he was provoked like a terrible flood, or
like a lion against hunters, or like a bear bereft of young; so was he. Although he saw
them united and risen against him, despite this, in accordance with his cunning, he
did not display his fear or his hiding from his face, but with arrogance and great
rebuke, he replied, ‘Depart, say to your king and our brother khagan that your house
has been venerated and honoured long since by my ancestors and by me as if our
beloved brother. For this reason we have been grafted onto one another in alliance
through sons and daughters.
62So it was not right or worthy for you to cast off your
limbs and be led astray by the words of my servant, the valiant one 63of the Romans.’
The messenger returned to his country. (tr. Greenwood)
The final Roman counter-strike (627–8)
Heraclius probably returned to Constantinople for the winter of 626–627, allowing his troops
some rest before returning to the offensive. Then in spring 627 he set off for the Transcaucasus
once more, arriving first in Lazica. In the meantime, his diplomatic overtures to the Turks
continued to bring results, and a large-scale invasion of Albania was undertaken by the Turkish
yabghu khagan. The two sides then converged at Tiflis in Iberia. See Howard-Johnston 1999:
Theoph. A.M. 6117 (315.26–316.16) (cf. Chr. 1234, 98, 233/137, Mich. Syr. XI.3
[409a/409], Agap. PO8.463–4): 64Now the Khazars, after breaking through the
(p.316) Caspian Gates, invaded Persia, i.e. the land of Adraigan, 65under their com-
mander Ziebel 66who was second in rank after the khagan. And in whatever lands they
traversed, they made the Persians captive and consigned the towns and villages to the
flames. The emperor, too, set out from Lazica and met them. When Ziebel saw him,
he rushed forward, embraced his neck and did obeisance to him, while the Persians
were looking on from the town of Tiflis. And the entire army of the Turks fell prone
to the ground and, stretched out on their faces, reverenced the emperor with an
honour that is unfamiliar among (foreign) nations. Likewise, their commanders
climbed on rocks and fell (flat) in the same manner. Ziebel also brought his adolescent
son to the emperor, and he was delighted by the emperor’s conversation and struck by
his appearance and wisdom. After picking 40,000 noble men, Ziebel gave them in alli -
ance to the emperor, while he himself returned to his own land. Taking these men
along, the emperor advanced on Khusro. (tr. Mango and Scott, revised)
Nic. 12.16–43 also describes the meeting of Heraclius and the yabghu khagan in some detail.
According to his account the emperor betrothed his daughter Eudocia to the khagan in return
for his assistance; and with the troops he received from the Turks he invaded Persia. See Mango
1990: 180–1 and Zuckerman 1995: 120–3 (on the significance of this dynastic marriage, which
came to nothing when the khagan was assassinated in late 629).

HVG224–6/233–5 offers a few details from the Iberian perspective. It describes how the
Iberian ruler Stephen remained loyal to the Persians and fortified himself in Tiflis. He opposed
the Romans courageously but he was killed and the city captured. In his place Heraclius
installed Adarnase as king of Iberia, a kingdom reduced in size in the west after Heraclius’ victo -
ries in 627–628. See Toumanoff 1963: 389–91.
Movs¢s Daskhurants‘i II.11 (135.5–140.14): Now in the thirty-eighth year (of
Khusro), which was a year of crisis and misfortune (and) of the murder of Khusro, the
same man came himself, whom we mentioned above, who was Jebu Khak‘an,
bringing with him his son as well. And no one was able to quantify the number of his
force. When the terrible, sorrowful news reached this land of Albania, it was decided
to defend this country of ours in the fortress of the great capital city Partaw. That
came about from the order of one man, whose name was Gayshak‘, who had been sent
by Khusro as head and prince of this country. He surrounded the majority of the
neighbouring provinces and wanted to consolidate himself by means of an alliance
with the magnates of this country and the inhabitants of the city (in order) to oppose
them (the raiders). But he watched to see what events would occur to the garrison of
the great city of Ch‘or, and to the garrison of the wonderful walls (for) which the kings
of Persia had exhausted their country through vast expenditure, collecting architects
and devising various materials for the construction of the astonishing project (by)
which they barred and enclosed (the pass) between mount Caucasus and the great sea
of the east.
Now at the approach of the universal wrath which was facing us all, however, the
waves of the sea extended and struck it (the city of Ch‘or) and demolished it from (its)
foundations. Because (their) terror increased on seeing the ugly, insolent, (p.136)
broad-faced (men), without eyelashes, in feminine dress with locks of hair, as they
charged, trepidation possessed them (the defenders), especially on seeing in front of
them the well-drawn, expertly aimed (bows) which were showering (arrows) upon
them like very heavy hail. When these ravenous shameless wolves charged against them, they (the defenders)
were massacred indiscriminately in the streets and passages of the city. Their eye did
not spare the handsome, the beautiful, the young men or women, nor the insignifi -
cant or useless. They did not pass over the crippled or the old, nor did they show pity
nor did their heartstrings writhe with compassion for the children, those who
embraced murdered mothers and suckled blood from their breasts rather than milk.
But like having lit a fire in straw, they went in through one gate and went out through
the other. They left work for the beasts of the field and birds of the air in it. Then grad -
ually the waves were stirred against us. Now when our head and prince heard about all this, who was guarding and holding
the city of Partaw, he wanted to speak to the multitude who had fortified themselves,
who had gathered in the great fortress, because (their) terror had increased, as to what
it would be right to do. He opened his mouth but from great fear at that time, he was
unable to repeat his plan, because his courage had disappeared and shaking seized him
from foot to head, and his knees were knocking together. And when the multitude

perceived that the man had completely lost courage, they raised a clamour and said,
‘Why have you shut us in completely at this time, to give us with our wives and chil-
dren into the clutches of bloodthirsty animals? Also how shall we be able to (p.137)
leave and flee before them, such a large common crowd of this city, because behold the
exterminating enemy has come and is three miles from us?’ Then each man said to his
neighbour, ‘Why are we (so) passive (as) to make this city a tomb for ourselves? Let us
abandon our goods and wealth, let us leave and depart; perhaps we shall be able to save
ourselves.’ And having raced in common to the four gates of the city, they hastened to
escape to the mountainous region of the province of Arts‘akh.
When the enemy became aware of what had happened, they strained after the fugi -
tives and caught up with one group at the foot of the mountain which is opposite the
village of the large region of Ka ³ankatuk‘,
71which is in the same province of Uti, from
which I also am. And at the darkening of the day, they were not able to spend time
injuring many, but a few fell into their clutches, some of whom they massacred; having
placed others forcibly behind the baggage, the carts and the beasts of burden, they
returned from there to their camp. And by the protection of God, they did not continue
to press after the multitude of fugitives. During that night, everyone crossed, just as in
an earlier time, the Hebrews (crossed) through the Red Sea, and they slipped away into
the inaccessible province of Arts‘akh. In the same way, that prince, by name Gayshak‘,
saved himself along with his whole house and he slipped away to the regions of the
Persians, but he was not able to establish himself in the same office of prince. After all this, the mountain streams rose, and a rapid river gushed against the country
of Iberia, and they surrounded and blockaded the fine, commercial, renowned great city
of Tiflis. And it was brought to the attention of the great emperor (p.138) Heraclius. He
assembled all the forces of his power, (and) straightaway he came to his assisting ally. On
drawing near, they received gifts and royal presents; they rejoiced on seeing one another.
Then from that time there was visible the miseries of the unfortunate persons, those
who were shut in the fortress. Calamity upon calamity apprehended them. But because
the time (of the city’s capture) had not yet arrived, which stood in the future, it
happened in this way, that when Khusro learned about the meeting of both great kings
at that city, before the siege (began) he sent rapidly a force to their assistance for the
defence of the city, the eager and valiant warrior, his commander Shahraplakan, and
elite cavalry of his bodyguards and courtguards with him, about one thousand men (in
number). When the inhabitants of the city saw the reinforcements despatched,
consisting of strong experienced warriors, they grew much stronger in themselves and
began to deride the two kings. For although they could see the countless multitude of
the forces of the north and west, drawn up around the city in the likeness of mountains
– the land trembled and shuddered from the multitude – and with the same (army)
engines and four-wheeled (machines) and various other devices prepared by the hands
of Roman engineers, by which they threw stones accurately in order to demolish the
circuit wall by means of very large stones; and (although they could see) the very large
bloated skins full of stones (and) full of sand, by which they (their besiegers) had
compelled the great river Kur, which surrounds one side of the city, to rush and burst
against the circuit wall, they were not at all faint-hearted at any part of the proceedings;

(p.139) rather they encouraged one another and they rebuilt and reconstructed the
demolished parts of the wall.Now when the forces of the two kings were dispirited and used up and exhausted,
and a considerable number of infantry had fallen in battle, the kings consulted one
another and said, ‘Why is this loss to our forces (taking place)? Is it not the case that
when we have bound the strong man, we shall pillage his house as we may wish?’ Then
the zealous great emperor Heraclius organised and arranged whatever it should be
right to do. He said to that man who had come to his assistance, ‘Turn, go peacefully
with your forces this year to your place. For we see your cool disposition, and behold
you will not be able to endure at the coming of summer time the burning heat of the
country of Asorestan, where the royal residence of the Persians is, beside the great river
which is the Tigris. And at the coming of another year, at the diminishing of the
months of heat, may you hasten to advance to here so that we may realise our plans. I
shall not cease to contend with the king of the Persians and having ventured forth, I
shall harass this country and his subjects, and I shall contrive in such a way through
cunning that he will be killed by his own people.’ And it happened that when the inhabitants of the city learned of their dishearten -
ment and weakening, they became even more arrogant in themselves and they began
to mock the (eventual) cause of their own destruction. They brought one large
pumpkin and they painted upon it a picture of the king of the Honk‘ (Huns), one
cubit long and one cubit wide. They stretched one vine branch in the place of his eye
lashes, which no one was able (p.140) to make out. And in the place of his beard, an
impudent baldness and in the place of his nose, nostrils one span wide, and with a
number of moustache hairs so that anyone could recognize him. Having carried that
(out), they placed it upon the circuit wall opposite them and they shouted to the
forces and said, ‘Here’s the emperor your king, where he is standing. Turn, worship
this one. It is Jebu Khak‘an.’ And having taken a spear in hand, they pierced the
pumpkin in front of them – the one which had been made similar to his figure. In the
same way, they ridiculed the other king, made a mockery and jeered and called him
impure and a sodomite. When the kings saw and heard this, they nourished resent -
ment and became swollen (with rage); they hoarded and collected rancour in their
own persons (in their determination) that not a single soul out of all of them should be
saved – (of) those who were under their rule
72– until revenge should have been
exacted for those insults with which they had been ridiculed by them. Having turned
their faces, they left with insults. (tr. Greenwood)
Movs¢s describes the actual fall of the city at II.14 (151–3/94–5). He apparently dates it to the
following year, but it seems more probable that it took place later in 628. See Howard-Johnston
1999: 24 and n.77.
While the yabghu khagan prosecuted the siege of Tiflis, Heraclius moved southwards with an
army comprising Roman forces and a substantial Turkish escort; he also secured the support of
many Armenians.
73In the late summer and early autumn, Roman control was restored to the

region, while the Persians sought to muster more armies to meet the new threat. See Howard-
Johnston 1999: 24, Speck 1988: 135–8.
Theoph. A.M. 6118 (317.11–26):In this year Heraclius, by invading Persia together
with the Turks starting in the month of September – an unexpected move, since it was
winter – threw Khusro into a state of distraction when he learned this. But the Turks,
seeing the winter and the constant attacks of the Persians, could not bear to toil
together with the emperor and started, little by little, to slip away, and (eventually) all
of them had left and returned home.
74Now the emperor addressed his own army,
saying: ‘Know, brothers, that no one wishes to ally with us, except God and His
Mother who bore Him without seed, and (and) so that He may show His might,
who trust in His mercy> He sends down His aid.’ (As for) Khusro, he collected all his armies and appointed Rahzadh commander
over them, a most warlike and brave man, whom he sent against Heraclius. The
emperor (meanwhile) was burning the towns and villages of Persia and destroying by
the sword the Persians he captured. On the 9th of October of the 15th indiction he
reached the land of Chamaetha,
75where he rested his army for one week. (As for)
Rahzadh, coming to Ganzak, in the emperor’s rear, he followed him, while the
Romans, in front, were destroying the crops. Trailing behind, like a hungry dog, he
was sustained with difficulty on the emperor’s crumbs. (tr. Mango and Scott, revised)
Sebeos 126/83 describes Heraclius’ route in more detail. Initially he moved westwards, into
Gogovit, the region south of Dvin, but then turned southeast, passing to the west of Lake
Urmia towards Ganzak. See Manandjan 1950: 148–53, Stratos 1968: 205–6, 208–9 and
Thomson and Howard-Johnston 1999: 218–19. From Chnaitha Heraclius moved west across
the Zagros through the Keli Shin pass, heading down towards the river Tigris; one detachment
at least may have employed a more northerly route, passing through Marga (Hist. Rabban bar
‘Idta, 1280–1319, p.67–8/253–5). See Howard-Johnston 1999: 25, Fiey
1966: 7–9.
Theoph. A.M. 6118 (317.32–323.22, 324.16–325.10): On the 1st of December the
emperor reached (p.318) the Great Zab river, which he crossed, and encamped near
the town of Niniveh. Following him, Rahzadh, too, reached the ford and, going three
miles downstream, found another ford and crossed. The emperor sent out the com -
mander Baanes
76with a few picked soldiers; the latter encountered a company of Per -
sians and, after killing their captain, brought (back) his head and his sword, which was
all of gold. He killed many more and made 26 captive, among whom was the sword-
bearer of Rahzadh. This man announced to the emperor that Rahzadh was intending
to attack him, (since) thus Khusro, who had sent him 3000 armed men, had com -
manded; but these men had not yet arrived. Learning these things, the emperor first
sent away his camp equipment and himself followed,
77seeking a place in which to give
battle before the 3000 had joined (the enemy). And when he had found a plain suit -
able for fighting, he addressed his army and drew it up in battle order. Upon arriving
there, Rahzadh also drew up his army in three wedge formations and advanced on the
emperor. Battle was given on Saturday, the 12th of December. The emperor sprang

forward in front of everyone and met a commander of the Persians in battle, and, by
God’s might and the help of the Mother of God, threw him down; and those who had
sprung forward with him were routed.
78Then the emperor met another (Persian) in
combat and cast him down also. Yet a third assailed him and struck him on the lip
with a spear, wounding him; but the emperor slew him too.
79And when the trumpets
had sounded, the two sides attacked each other and, as a violent battle was being
waged, the emperor’s tawny horse, called Dorkon, was struck in the thigh by the
infantry, receiving a spear (blow) to its thigh. It also received several blows of the
sword on the face, but, wearing as it did a coat of armour made of fibres, it was not
hurt, nor were the blows effective. (p.319) Rahzadh fell in battle, as did the three divi -
sional commanders of the Persians, nearly all of their officers and the greater part of
their army. As for the Romans, fifty were killed and a considerable number wounded,
but they did not die, save for another ten. The battle was waged from morning until
the 11th hour. The Romans captured 28 standards of the Persians, not counting those
that had been broken, and, having despoiled the dead, took their corselets, helmets
and all their arms. And the two sides remained at a distance of two bowshots from one
another, for there was no rout.
80The Roman forces watered their horses at night and
fed (them). But the Persian horsemen stood until the seventh hour of the night over
the bodies of their dead; and at the eighth hour of the night they moved (off) and
returned to their camp; and taking it up, they went away and encamped in fear at the
foot of a rugged mountain. The Romans took many all-gold swords and belts interwo-
ven with gold and set with pearls, and the shield of Rahzadh, which was all of gold and
had 120 laminae, and his all-gold breastplate; and they brought in his caftan together
with his head, and his bracelets and his all-gold saddle. And Barsamouses, the prince
of the Iberians who are subject to Persia, was taken alive.
81 battle having taken place between Persians> and Romans in that the battle did not
cease for the whole day. The Romans won, but this occurred only by God’
s help.
After encouraging his army, the emperor pushed on against Khusro with a view to
frightening him and so that, having sent (him) forth, he might summon Shahrvaraz
from Byzantium.
83On the 21st of December (p.320) the emperor learned that the
army of Rahzadh – as much of it as had escaped from the battle – had been joined by
the 3000 men despatched by Khusro and had reached Niniveh as it followed behind
him. After crossing the Great Zab, the emperor 84
with 1000 men to ride forward and seize the bridges of the Lesser Zab> before Khusro
had become aware (of it). After riding 48 miles in the night, 85George seized the four
bridges of the Lesser Zab and, finding Persians in the forts, took them alive. On the
23rd of December the emperor reached the bridges, crossed them and encamped in
the mansions of Yazdin; he rested both his army and his horses and celebrated the feast
of Christ’s Nativity in that place.
When Khusro became aware that the Romans had seized the bridges of the Lesser
Zab, he sent a message to the army (that had been) under Rahzadh that they should
try very hard to overtake the emperor so as to return to him. Making haste, they
crossed the Lesser Zab in another place and anticipated the emperor, and (now)
marched in front (of him). (As for) the emperor, he came upon

Dezeridan, which he destroyed and handed over to the flames, (while) the Persians
crossed the bridge of the river Tornas and encamped there. 87The emperor came
upon> another palace of Khusro called Rousa 88 and this, too, he destroyed. He
suspected that the enemy were going to fight him at the bridge of the river Tornas; but
when they saw him, they abandoned the bridge and fled. (So) the emperor crossed
without hindrance and reached another palace called Beklal;
89here he built a hippo -
drome and destroyed it. 90Several of the Armenians who accompanied the Persians
came to the emperor and said: ‘Khusro with his elephants and his own
army (p.321) is encamped five miles from the palace called Dastagerd, in a place called
91and he has given instructions that his forces should assemble there and
fight you. There is a river there that is difficult to cross, and a narrow bridge, and many
cramped spaces between buildings, and fetid streams.’ The emperor, after taking
counsel with his officers and his army, remained in the palace of Beklal. He found
therein in one enclosure 300 corn-fed ostriches, and in another about 500 corn-fed
gazelles, and in another 100 corn-fed wild asses, and all of these he gave to his army.
And they celebrated the 1st of January there. They also found sheep, pigs and oxen
without number, and the whole army rested, enjoying (these things) and praising
God. They caught the herdsmen of these cattle and accurately learned from them
that Khusro had learned on the 23rd of December that the emperor had crossed
the bridge of the Tornas, and forthwith set out from the palace of Dastagerd
92loading all the money he had in the
palace on the elephants, camels and mules that were in his service, and he wrote to
the army of Rahzadh that they should enter that same palace and the houses of the
noblemen and take away anything they found therein. Therefore the emperor sent
one half of his army to Dastagerd>, while he himself went by a different road to
another palace called Bebdarch,
93and having destroyed this too and consigned it
to flames, they thanked God for having wrought such wonders by the entreaties of
the Mother of God. For who had expected that Khusro would flee before the
Roman emperor from his palace at Dastagerd and go off to Ctesiphon, when, for
24 years, he would not suffer to behold Ctesiphon, but had (p.322) his royal resi -
dence at Dastagerd? In his palace of Dastagerd the Roman forces found 300
Roman standards which the Persians had captured at different times. They also
found the goods that had been left behind, namely a great quantity of aloes and big
pieces of aloes wood, each weighing 70 or 80 lbs, much silk and pepper, linen
garments beyond counting, sugar, ginger and many other goods.
94Others found
silver bullion, silken cloaks, woollen rugs and woven carpets – a great quantity of them
and very beautiful, but on account of their weight they burnt them all. They also
burnt the tents of Khusro and the porticoes he set up whenever he encamped in a
plain, and many of his statues. They also found in these palaces an infinite number of
ostriches, gazelles, wild asses, peacocks and pheasant, and in the hunting park huge
live lions and tigers. Many of the captives from Edessa,
95Alexandria and other cities
sought refuge with the emperor, (as did) a large multitude from other cities. The
emperor celebrated at Dastagerd the feast of the Epiphany,
96cheering and refreshing
the army and horses while destroying the palaces of Khusro. These priceless,

wonderful and astonishing structures he demolished to the ground so that Khusro
might learn how great a pain the Romans had suffered when their cities were laid
waste and burnt (by him).
Many of the palace stewards were also seized and, on being interrogated as to when
Khusro had departed from Dastagerd, they said: ‘Nine days before your arrival he
heard of your presence and secretly made a hole in the city wall near the palace.
this way he went out unhindered (p.323) through the gardens, he and his wife and
children, so there should not be a tumult in the city.’ And neither his army was aware
of it nor his noblemen until he had gone five miles from there, at which point he
announced that they should follow him in the direction of Ctesiphon. And this man
who was incapable of travelling five miles in one day, travelled 25 in his flight. His
wives and children, who previously had not laid eyes on one another, now fled in dis -
order, jostling one another. When night had fallen, Khusro entered the house of an
insignificant farmer to rest, but could scarcely be admitted through the door. When,
later, Heraclius saw that (door), he was amazed. In three days Khusro reached
Ctesiphon. Twenty-four years earlier, when he besieged Dara in the days of Phocas,
the Roman emperor, he had been given an oracle by the magicians and astrologers,
namely that he would perish at whatever time he went to Ctesiphon; and although he
would not suffer to go one mile in that direction from Dastagerd, he now returned to
Ctesiphon as he fled. Even there he did not dare stop, but crossed the pontoon bridge
over the river Tigris to the town on the other side, which is called Seleucia by us and
99by the Persians. He deposited all his money there and remained there
with his wife Shirin and three other women who were his daughters. His remaining
wives and his many children he sent to a stronghold forty miles further
east. (A (misplaced) section on the siege of Constantinople in 626 and a deal between
Heraclius and Shahrvaraz is omitted here.) Now Heraclius wrote to Khusro: ‘I am pursuing you as I hasten towards peace. For
I do not willingly raze Persia, but after being forced by you. Let us, therefore, throw
down our arms even now and embrace peace. Let us extinguish the fire before it con -
sumes everything.’ When Khusro did not accept these proposals, the hatred of the
Persian people grew against him.
100 He conscripted all the retainers of his noblemen
and all his servants and those of his wives and, having armed them, sent them to join
the army of Rahzadh and take a stand at the river Narbas, 12 miles from Ctesiphon.
He commanded them that when the emperor had crossed the river, they should cut
the bridge and the pontoon bridge. (As for) the emperor, he moved from Dastagerd
on the 7th of January and, after marching three (p.325) days, encamped 12 miles from
the river Narbas, where the Persian camp lay and where they had 200 elephants. The
emperor despatched George, turmarch of the Armeniacs, to go as far as the river and
ascertain whether the Narbas had a ford. And when he had found that they had cut the
bridges and that the Narbas had no ford, he returned to the emperor. Setting forth,
the emperor came to Siazouros and, for the whole of the month of February, he went
about burning the villages and the towns.
102 In the month of March he came to a vil -
lage called Barzan, 103 where he spent seven days; and he despatched the commander
Mezezius 104 on a foray. (tr. Mango and Scott, revised)

Sebeos 126–7/83–5 (cf. Thom. Arts. 93–5/160–2) reports in similar fashion the battle of
Niniveh and Khusro’s flight across the Tigris and downfall; see Thomson and Howard-
Johnston 1999: 219–20. Nic. 14 recounts Heraclius’ duel with Rahzadh slightly differently
from Theophanes, cf. Mango 1990: 182 and Speck 1988: 317–21. On other sources dealing
with Khusro’s fall (in much less detail) see Flusin 1992: II, 265–6 and n.6 (and add Tabari
1004–5/322–3). Heraclius’ withdrawal northwards was probably prompted by several factors,
among them his concern that Shahrvaraz might arrive on the scene (cf. Sebeos 127/84), his
inability to reach Ctesiphon, the opportunity to damage Persian lands in the Diyala valley, and
perhaps an awareness that events at the Persian court might overtake Khusro. See Baynes 1914:
674, Stratos 1968: 218–20, Howard-Johnston 1999: 5–6.
Movs¢s Daskhurants‘i II.12–13 (143.21–148.24):Then the emperor took his forces
and marched into the regions of the Persians. Being obdurate and determined, he
hastened to reach the court of the Persian king. Now when (p.144) the Persian king
saw that it (the campaign against Heraclius) had not been carried out in that way, but
that (the emperor had) been daring, had marched, and was coming against him, he
took flight before him and slipped away to his own capital, to the great Ctesiphon. He
transported his wives and concubines and children to the far side of the river Tigris.
He assembled his permanent force, which was small in size (and) close to him,
prepared it and sent it against the emperor. He attracted an individual through
distinctions and wealth, who was acknowledged at his court because of his valour – he
transferred another name to him and called him Rochveh (Rahzadh) – (and) he
promoted him by way of inducement and he struck against the waves. He appointed
him as commander of the force about to be sent against it (the Roman army). Out of
fear of his order, he reluctantly agreed, but he himself anticipated the vanquishing of
the hastily assembled and weak force of Khusro before the emperor. He wrote once
and twice and four times to his king Khusro out of concern and said, ‘If not of my own
death, I shall at least warn you in advance about the destruction of your forces. For if
you do not supply to me straightaway a force to assist me, although I am not afraid of
dying, then you will learn of (my death).’ Then the king ordered this reply to be composed: ‘Do not be frightened in the face
of them, but fight and overcome.’ But on the last occasion, being extremely disdain -
ful, he wrote to him saying, ‘If you will not be able to overcome, why should you not
be able to die?’ When he saw that harsh order, having raised his hands with his forces
to the sun and moon, he shouted with a loud voice and said, ‘My gods, judge between
me and my merciless king.’ (p.145) And having plunged into battle, he and his forces fell before the Greek forces and
became like dust which a storm carries away.
Now when the Persian nobles witnessed that great rout which happened to the
Persian forces, they began to complain to one another, ‘For how long will the streams
of blood flow in various places on account of battle, (the blood) of Aryan persons of
this country? For how long will we be in fear and trepidation of this blood-letting
king? For how long will our possessions and property be amassed to the royal treasury,
our silver and gold to his treasuries? For how long will the passes of roads remain

constrained and blocked, damaging the profits of trade of various regions? For how
long will our souls remain quaking and emptied from our persons by his terrible
command? Did he not in fact consume and swallow like the sea our most excellent
companions, the leaders of countries? Did not in fact many of our brothers perish on
various occasions, in companies, in divisions, in various torments by his command,
some even being drowned? Did he not snatch a man from his wife, and a father from
his son, and scatter (them) to distant peoples in the capacity of a servant or female
domestic, and have (them) assembled against ferocious enemies?’They muttered this and much similar about him, but no one dared to speak any -
thing openly until the time for the end of his life was reached. Then someone from his
trusted friends, a noble, arose, who was protector of the first-born of Khusro who was
called Kavadh.
107 Having formed a plan with him, he had (them) destroy the reputa -
tion of his father Khusro and the multitude of his brothers. And he took the office of
the king of Persia. Through the striving and directing (p.146) of his protector he
seduced and disturbed the very decorous ranks of the members of the court of Khusro.
And he diverted the hearts of everyone unawares in favour of his foster-child Kavadh,
in order to make him king in place of his father. II.13. But listen to me for a moment and I shall describe briefly the cunning of the
man, how or by what way he was able to arrest in a cage of death the terrible hunter,
the lion of the east, at whose roar alone distant peoples shook and trembled and those
close to his presence shrank, like melted wax. And that man
108 in that way in a
moment stole the hearts of everyone and rose up against him as if (he was) an orphan.
In performing all this, he did not summon another king or prince to his assistance, nor
did he organize distant peoples and tribes to bring reinforcements to his foster-child
against him. He merely sent secretly to the emperor Heraclius for him to stay put there
where he was with his forces for a few days.
He ordered official letters to be written on behalf of Kavadh to the magnates and
the leading members of the various contingents of the great court of the Persian king-
dom. ‘The authority over this kingdom has been removed from my father and given
to me; be prepared with a few cavalry.’ He established them at the head of the bridge
of the river Tigris near to the city of Veh-Ardashir opposite the gate of (p.147)
Ctesiphon where his father Khusro was, being guarded by his personal forces. A herald
proclaimed loudly from the right and the left of Kavadh, saying ‘Whichever man who
loves life and wants to see his days in mildness, let him hasten to come before this new
régime of Kavadh, for this man is king of kings.’ And they opened the gates of the for -
tress of Oblivion
110 and he summoned outside audaciously all the prisoners of the king
– those who had been held in the shadows of death for a long time, in a huge countless
multitude. And he said to them, ‘Come out, you unfortunate people, touched by
Khusro, for the gates of life have been opened for you by this newly-dedicated king, by
his son Kavadh.’ And each one having broken his chains, they rushed outside
together. The noise of their praise being shouted went up to heaven; they praised
Kavadh saying, ‘O king, may you live for ever.’ Having driven (off) the most select
horses of the stables of Khusro, they mounted (them) with their chains in their hands,
and galloped this way and that, abusing Khusro.

Then many of the various contingents of house-guards and bodyguards and others
similar to them 112 of the court of Khusro seized the symbols of their banners and
rushed to Kavadh his son. And (as for) those who remained in the court, the messen -
gers of Kavadh urged them to behave prudently and to arrest Khusro, because other -
wise they would perish. Now when Khusro heard the clamour of that shouting, he
said to those who were standing nearby, ‘What are the sounds of those cries?’ And they
were silent because they were ashamed to explain (them) to him. And the mixed sound
of trumpets pressed more and more. Again he said to those who were nearby, ‘What
are the disorderly (p.148) sounds of cries?’ And they said, ‘Because your son Kavadh
intends to become king in place of you, and everyone is running to him. Behold, he
has drawn up opposite this city beside the bank of the river and has released all the
prisoners of the great prison, those arrested by your command. And with joy everyone
is saluting him, calling (him) king.’ When he heard all this disastrous news, especially the release of the prisoners, he
was exhausted, he sighed, he lamented and was uncertain of himself. He was (not able
to) think when sitting nor contemplate when standing, because he saw the immense
force that had arrived against him. He went out on foot through the gate of his garden;
he went and hid among the trees of the garden and he stood concealing himself; his
soul was emptied by sudden fear. His eyes watched the merciless sword raised above
him. Then the whole contingent came against him and arrived around the palace.
Searchers went in after (him) and found him at that place where he had been hesi-
tating. Having conducted him outside the palace, they threw him into a hall which
they call K‘ataki Hndukn, the house of the Indians, in the palace of the one called
113 When he entered, he enquired about the place and the lord of the
house; having beaten his breast, he said groaning, ‘Woe is me, how unfortunate I am!
How I have been deceived by sorcerers, those who divined that I would be arrested in
India! And now behold it has been allotted to me in accordance with the description
of their vision, albeit metaphorically.’ Having kept him for that day, on the following day they removed his head with a
sword. The evil one was exterminated. And in the same way King Kavadh, having
mutilated his brothers’ feet and hands, intended to spare their lives in a crippled state,
but then after the complaint of many, these too were removed with the sword.
114 (tr.
As described above (and below), in late February Heraclius withdrew northwards to
Atropatene, leaving events in Persia to take their course. The devastation wrought by the
Roman army in the heart of the kingdom and the manifest failure of their king accelerated the
demise of Khusro.
115 He was replaced in a coup during the night of 23–24 February and
executed five days later; he was succeeded by his son Kavadh II Shiroe, who immediately
entered into negotiations with Heraclius. It was not until the arrival of Kavadh’s embassy on 24
March, as emerges from the Chronicon Paschale, that Heraclius learned at Ganzak of the fall of
Khusro. The account of the Chronicon Paschalebelow is the text of a despatch sent by Heraclius,
which was read out in St Sophia in Constantinople on Sunday 15 May 628; it is therefore a
contemporary source of the highest value. On the events described see Christensen 1944:

493–6, Stratos 1968: 226–8, Howard-Johnston 1999: 5–6, 26, Thomson and Howard-
Johnston 1999: 221.
Chr. Pasch.727.15–734.17: May you, the whole world, shout to God (in victory),
may you serve the Lord in merriment, enter His presence in exultation and know that
the Lord himself is God. He made us and not we ourselves. We are His people and the
sheep of His pasture. Enter His halls with hymns and give thanks to Him. Praise His
name, since Christ (is) Lord and His pity (endures) for ever, and His truth (endures)
through the generations. Let the heavens rejoice (p.728) and the earth exult and the
sea make merry and all things in them.
116 And let all of us Christians, praising and glo -
rifying (Him), give thanks to God alone, rejoicing greatly in His holy name. For the
arrogant Khusro, who fought with God, has fallen. He has fallen and has been over -
thrown, down to the underworld; and his memory has been obliterated from the
earth. He who was puffed up and spoke unjustly in his arrogance and contempt
against our Lord Jesus Christ the true God and his undefiled mother, our blessed
Lady, the Mother of God and ever-virgin Mary, he, the impious one, has perished
with a crash. His labours rebounded on his own head, and onto his head descended his
injustice. For on the 24th of February past of the first indiction (628), there was an
uprising against him by Shiroe, his first-born son, as we indicated to you by another
despatch; and all the Persian nobles and armies, and the whole force assembled from
various places by the accursed Khusro, joined the side of Shiroe with Gusdanaspes
too, the former commander of the Persian army. That God-hated Khusro prepared to
turn to flight, but was prevented and taken prisoner in the new fortress which had
been built by him for the guarding of the monies collected by him. (p.729) And on the 25th of the same February Shiroe was crowned and proclaimed
king of the Persians, and on the 28th of the same month, after he had placed the iron-
bound God-hated Khusro in great pain for four days, he killed this same arrogant,
proud blasphemer, who fought against God, by a most cruel death so that he might
know that Jesus, born of Mary, crucified by the Jews, as he himself had written in blas-
phemy, is almighty God;
117 and he treated him (Khusro) according to how we had
written to him (Kavadh Shiroe). And thus did that man who fought against God
perish in this life. He departed by the road of Judas Iscariot who had heard from our
almighty God that ‘it would have been better for that man if he had not been born’;
he departed to the unquenchable fire prepared for Satan and those worthy
of him. Through our other despatch to you from our camp near Ganzak which contained
the account (of events) from the 17th of October up to the 15th of March, we indi -
cated how God and Our Lady the Mother of God acted with us and our Christ-loving
expeditionary forces (in a way) beyond human comprehension; and how the God-
hated and accursed Khusro fled from Dastagerd to Ctesiphon, and how his palaces
were destroyed along with many districts of the (p.730) Persian state; and in what
manner Shiroe was able to make his move against him. After we had composed that despatch and sent it on the 15th of the present month
of March, we were concerned to know the subsequent events surrounding Khusro and
Shiroe, and we sent (letters) to various places, as far as Siarsuron
119, and as far as the

Lesser Zab, Kalkhas, (the mansions) of Yazdin, by both roads (leading from) our most
successful expeditionary forces, as well as through the Saracens who are in the control
of our Christ-loving state.
120 (This we did) in order, as has been said, to know precisely
what had taken place there. On the 24th of the same month of March scouts brought
to us at our camp near Ganzak one Persian and one Armenian, who handed over to us
a memorandum (addressed) to us from a certain a secretisof the Persians with the
name Khosdaes and the rank (of) Rasnan.
121 It contained (the news) that Shiroe,
having been proclaimed king of the Persians, had sent him to us, along with other
nobles too, and this memorandum of Shiroe to us, and that, having reached Arman,
he (Khosdaes) had resolved to despatch to us the two aforementioned men so as to
have some men sent (to him) who would keep him and those with him safe. It was
realised by us that he had observed many corpses of Persians along the road, slain by
our most successful (p.731) army – about 3000 men from the Narbas (onwards);
frightened by this, he feared to reach us without an escort. And on the 25th of the
same month of March we sent to them the most glorious magister militumElias called
Barsoka and the most magnificent Theodotus the drungarius
124 with younger soldiers
and twenty pack-horses with saddles, who would meet and bring them safely to us.
With them we had resolved to send also Gusdanaspes Rhazei, the commander of a
thousand men who had come to us when trouble broke out between Shiroe and
Khusro. On the 30th of March we received a reply at our camp near Ganzak from
Elias and Theodotus and Gusdanaspes that they had found much snow in the Zagros
mountains and that they had brought Persians and horses out from their forts and had
thus dug through the snow; (they added) that it had come to their attention that the
envoys sent by King Shiroe were nearby, but that they had were unable to cross the
Zagros mountains. From this we – and all the men of our Christ-loving expeditionary
forces – realised all the more that the favour and goodness of God had guided us and
guides us and preserves (us yet). For if it had fallen out that we had tarried (p.732) in
the Zagros mountains, and if the storm had taken place in this way, our most success-
ful expeditionary forces would have come to great harm, since adequate provisions are
not to be found in these places. For from the time when we moved from Siarsuron,
that is from the 24th of February, until the 30th of March, it did not cease snowing.
But coming to the districts of Ganzak through God(’s help) we found many provi -
sions for men and horses; and we stayed in the city of Ganzak, which is suitable and
has about 3000 houses, and in the surrounding villages, so that we were able to reside
for many days in one place. We ordered our Christ-loving expeditionary forces to
place their horses in the houses of the city during the winter and to keep one each in
our camp; for our fortified camp is near the city. The superintendent of the city of
Ganzak and all its landowners, when they learned that we had crossed the Zagros
(mountains), withdrew and departed to mountain regions to more secure
forts. When we received the two men sent by the Persian king Shiroe, we despatched one
of them – that is the Persian – with other men too, to the superintendent of Ganzak,
who was forty miles away and in a secure fort; and (p.733) we addressed a memoran -
dum to him, to the effect that he should prepare sixty horses for the envoys, so that
they might depart immediately and without hindrance to the Persian king Shiroe.

The superintendent, receiving those sent by us and this memorandum of ours, praised
us and the Persian king Shiroe for many hours – he himself and all to be found there.
For they had been informed by the Persian who had been sent as an envoy that
Khusro, who fought against God, had perished and that Shiroe had become king of
the Persians. And the superintendent replied to us that he was making ready the horses
of the envoys in response to our commands to him, and that when he received our
second order containing (the news) that the envoys had arrived, he himself in person
would come, bringing the requisite horses; (and he said that) he would undertake
every task and service. While we stayed in our camp near Ganzak until the 3rd of
April, Phaiak thea secretisand Rasnan
125 arrived around the second hour, on a
Sunday. And that same hour we received him and he gave us the memorandum of
Kavadh Shiroe, the most clement Persian king, which contained his proclamation and
(announced) that he wished to have peace with us and with every man. We resolved
therefore to attach after our present despatch a copy of the memorandum of Kavadh
Shiroe the most clement (p.734), as well as to attach beneath our reply to him. We
stayed in this same camp of ours near Ganzak until the 7th of April, that is for 27 days,
and on the 8th of the same month we despatched the same Phaiak the a secretisand
Rasnan, having attended to him and all those with him as far as possible, for he had
with him distinguished personages; and we despatched with him the most magnifi-
cent tabularius Eustathius.
126 And thus we have faith in our Lord Jesus Christ the good
and almighty God and in our Lady the Mother of God, because they have arranged all
our affairs by their goodness. And with (the help of) God we moved our camp on the 8th of the same month,
intending to make our way to Armenia. But may you fare well, unceasingly and zeal-
ously praying on our behalf, so that God may grant us, as we desire, to see you (again). (p.735) Copy of the memorandum from Kavadh Shiroe, most clement king of the
Persians, to our most pious emperor Heraclius, protected by God. ‘From Kavadh
Sadasadasakh to Heraclius, the most clement Roman emperor, our brother, we offer
the greatest thanks.
127 To the most clement emperor of the Romans and our brother.
We, through the support of God, have with good fortune been vested with the great
diadem and have attained the throne of our fathers and forebears. Since we have been
so beneficently deemed worthy by God to attain such a throne and dominion, we have
decided that if there is anything for the good and service of mankind which can be
done by us, we have, as is fitting, beneficently ordered that it be done. When God
honoured us with this great throne and dominion, we had the policy of releasing each
and every person held in prison. And for the future, if there is anything for the good
and service of mankind and this state which could be ordered by us, we have ordered
(it), (p.736) and it has been done. And we have a policy of such a kind that we shall
live in peace and love with you, the emperor of the Romans and our brother, and the
Roman state, and the other peoples and other petty kings around our state. Because of
the delight of your brotherhood, the emperor of the Romans, at our accession to this
same throne
128 … directly … your brotherhood and we sent orders to your brother -
hood that Phaiak the a secretisand Rasnan, our (advisor), be present. But your broth -
erhood knows the disposition, love and friendship which we hold towards your

brotherhood, not only towards it but also towards your state. Because of your love,
may you order that the men from this state who have been captured (p.273) by the
army of your brotherhood be released. In addition, may you order these men by every
means to come to our ancestral state. And concerning the firm and eternal peace to be
(established) … ’(p.737) ‘In the name of our Lord Jesus Christ and God. Emperor Caesar Flavius
Heraclius, faithful to Christ, emperor of the Romans.
129 We offer the greatest thanks
and report that we received the memorandum sent to us by your filiality 130 through
Phaiak Khosdaes the a secretisand Rasnan. And we recognise that with the support of
God you have been vested with the royal diadem for the good fortune (of all), and are
seated on the throne of your father and forebears. We have rejoiced all the more and
devote ourselves to God so that he may deem you worthy to occupy the throne of your
fathers and forebears for many years in good health, in eternal good fortune and in
great peace. And that which you disclosed to us through your memorandum,
(namely) that when you sat on so great a throne and took power, you had the policy of
releasing those held in prison for various causes; and furthermore that you had
ordered that those things be done which were for the good and benefit of men (p.274)
and had such an intention …
Nic. 15 briefly reports the imprisonment and death of Khusro, as well as the exchange of letters
between Heraclius and Kavadh Shiroe. According to Nicephorus, Heraclius called Kavadh his
child (15.15) and asked for the return of the Holy Cross; he also enquired about the fate of the
ambassadors sent in 615. Kavadh replied that he would return the cross when he found it and
that the ambassadors had all died, all but one executed by Khusro. See Mango 1990: 182–3 and
Speck 1988: 322–4.
Theoph. A.M. 6118 (325.10–327.16): 131 Gusdanaspes, 132 who was captain of a
thousand men in the army of Shahrvaraz, 133 went over to him (Heraclius) together
with five others, three of whom were captains and two officers of other rank; and he
brought them to the emperor. This man Gusdanaspes announced some vital news to
the emperor, saying that ‘When Khusro fled from Dastagerd and went to Ctesiphon
and Seleucia, he fell ill with dysentery and wanted to crown his son Mardanshah who
was born to Shirin. And he crossed the river again and brought with him Mardanshah
along with Shirin and her other son Shahryar. As for his first-born son
134 and
his brothers and wives, he left them on the other side of the river. When Shiroe
became aware that Khusro was intending to crown Mardanshah, he was troubled and
sent his foster-brother to Gusdanaspes with this message, ‘Come across the river that I
may meet you.’ But Gusdanaspes was afraid to cross on account of Khusro and
declared to him, ‘Write to me through your foster-brother, if you wish.’ So Shiroe
wrote him the following: ‘You know how the Persian state has been destroyed by this
evil man Khusro, and now he intends to crown Mardanshah and has scorned me, the
first-born. If you tell the army (p.326) that they should accept me, I shall increase
their pay and make peace with the Roman emperor , and we
shall live in plenty. (So) strive with your men that I should become king. I will then

promote and acclaim all of you, and yourself in particular.’ I informed him through
his foster-brother that I would speak to the army and labour to the best of my ability.
And I spoke to 22 captains and won them over to my views, as well as other officers
and many soldiers. I announced this to Shiroe, who instructed me that on the 23rd of
135 I should take some young regulars and meet him at the pontoon bridge of
the Tigris river, take him to the army and move against Khusro. (He stated also) that
Shiroe had with him the two sons of Shahrvaraz, the son of Yazdin, and many other
sons of noblemen, and the son of Aram – all picked men. If they are able to kill
Khusro, well and good; but if they fail, all of them, with Shiroe, will go over to the
emperor. He sent me to you, O lord, because he feels ashamed before the Roman
Empire; for, once upon a time, it saved Khusro and, on his account, the land of the
Romans has suffered many ills. Because of his arrogance, he says, the emperor will not
trust me.’ Now the emperor sent this man back to Shiroe and indicated that he should open
the prisons and let out the Romans confined therein, and give them arms, and so
move off against Khusro. Shiroe obeyed the emperor and, after releasing the prisoners,
attacked his parricide father Khusro. The latter tried to escape, but failed and was cap -
tured. They bound him securely with iron fetters, with his hands tied behind his back,
and placed iron weights on his feet and his neck, and so led him into the House of
Darkness, which he himself had fortified and rebuilt for the deposit (p.327) of monies
therein; and they starved him by giving him a paltry amount of bread and water. For
Shiroe said, ‘Let him eat the gold he collected in vain, on account of which he starved
many men and made the world desolate.’ He sent to him the satraps that they might
insult him and spit upon him, and he brought Mardanshah, whom he had wished to
crown, and slew him in his presence, and they killed all his remaining children in front
of him, and he sent all his enemies that they might insult him, strike him and spit
upon him. After doing this for five days, Shiroe commanded that he should be killed
with bow and arrows, and thus slowly and in suffering he gave up his wicked soul.
Then Shiroe wrote to the emperor to give him the good tidings of the slaying of the
foul Khusro; and after making with him a permanent peace, he handed back to him all
the imprisoned Christians and the captives held in every part of Persia together with
the patriarch Zachariah and the precious and life-giving Cross that had been taken
from Jerusalem by Shahrvaraz, when the latter captured Jerusalem.
136 (tr. Mango and
Scott, revised)
As the Chronicon Paschale (p.734) relates, while Heraclius returned westwards, an envoy was
sent to Ctesiphon to arrange terms for concluding the war. Although Kavadh Shiroe released
Roman prisoners (Theoph. 327.12–13 above), he lacked the power to give orders to Shahrvaraz
to evacuate his conquests in the west.
137 (Sebeos 128/86). Heraclius was therefore obliged to
despatch forces to reconquer the occupied territories. The praefectus sacri cubiculiNarses was
entrusted with a large force, which, according to Ant. Strateg. 24 (516), drove the Persian into
138 More information is provided by Chr. 1234(below), cf. Theoph. A.M. 6119,
327.19–24, Mich. Syr. XI.3 (409a–10/409–10), Agap. PO8.465–6. 139 See Flusin 1992: II,
285–7, placing these events in 628, before the death of Kavadh Shiroe

Chr. 1234, 100–2 (234.26–236.12):(…)He(Shiroe) made with him (Heraclius) a
treaty and agreement, so that Heraclius would receive all the lands which belonged at
that time to the Romans, and the Persians would stay inside their former boundaries,
and those of them who were in the lands of the west would go over to the country of
the Persians. (p.235) Shiroe became king of the Persians in the year 19 of Heraclius
and 7 of Mohammed. After this, Heraclius marched from the east to arrive in Syria,
and he sent before him Theodoric (i.e. Theodore) his brother, to cast out Persians
who were living in the cities, according to the agreement which existed between him
and Shahrvaraz from before
140 and confirmed between him and Shiroe.
(101) Theodoric, the brother of the king, went over into the cities of Mesopotamia
and told the Persians who were in them to leave and go to their land. In addition, they
had already been informed in letters from Shahrvaraz and Shiroe of the treaty between
the Romans and the Persians. Then the emperor advanced in the footsteps of his
brother and, when he had arrived, he set up governors and Roman garrisons in the
cities. When Theodoric reached Edessa, he notified the Persians there of those things
which were (happening) and of the coming of the emperor; but the Persians scorned
him and did not listen to his words, and replied to him, ‘We do not know Shiroe and
we will not surrender the city to the Romans’. There were Jews who dwelt in Edessa
standing with the Persians on the wall, and because of their hatred for Christians, and
to ingratiate themselves with the Persians, they insulted the Romans and mocked
Theodoric and made him listen to bitter taunts. After this, Theodoric incited a violent
attack against the city and beset it with a hurling of rocks from a siege engine, and
when the Persians (who were) in the city were crushed together, they accepted an
amnesty to return to their land. A certain Jew, whose name was Joseph, when he
feared the destruction of his people, threw himself down from the wall and went off to
Heraclius, and found him at Constantia. He came before him and sought to persuade
him to forgive his people (p.236) the offence of insulting Theodoric, his brother, and
to send (instructions) to him so that he would not take vengeance on them. Theodoric
had entered Edessa and had taken control of it, and made the Persians go to their land.
He sent (men) to gather all the Jews who had insulted him, and then he began to kill
them and to plunder their houses. Suddenly Joseph the Jew arrived, bearing a letter
from the emperor, which ordered his brother not to harm them. (102) Afterwards Theodoric left Edessa and crossed the Euphrates and arrived at
Mabbug (Hierapolis), where he expelled the Persians who were in Syria and Phoenice.
In those days, Emperor Heraclius reached Edessa and began to dwell in the palace at
the head of the source (of the river). ( … ) (tr. M. Greatrex)
At Edessa Heraclius hoped to heal the divisions between supporters and opponents of the
Council of Chalcedon by taking communion from the bishop of Edessa, Isaiah, but he was
brusquely rebuffed. Henceforth the emperor was to prove more forceful in his dealings with the
Monophysites (Chr. 1234, 102, 236/140, Mich. Syr. XI.3 [408–9c/411–12], Agap. PO
141 The situation was further complicated by Kavadh’s death in October 628, upon
which he was succeeded by his young son Ardashir (Chr. Khuz. 29, Sebeos 129/87–8). 142

The peace treaty of 629
Despite the inglorious death of Khusro almost a year earlier, Roman control of the Near East
had not been fully restored by January 629 by any means. Mesopotamia, Osrhoene, and pre-
sumably Syria and Phoenice had been regained by Theodore, but Shahrvaraz remained in
143 Nothing is known of Persian troops stationed north of Syria. Heraclius was forced
therefore to deal with Shahrvaraz, clearly now a more powerful figure than the new king in
Ctesiphon (Sebeos 129/88, Thom. Arts. 96/162). On 17 July 629 the emperor met Shahrvaraz
at Arabissus to arrange terms. Since already in June Shahrvaraz had started to evacuate Egypt,
the two leaders had evidently been in contact for some time before their meeting (Chr. 724,
146/17–18). A western Turk invasion around the same time failed to dissolve the good rela -
tions now established. See Flusin 1992: II, 290–1, Zuckerman 1995: 118 and n.30, Howard-
Johnston 1999: 27–8.
Chr. 724, 147.18–24, cf. 139.13–18 ( AG 940 = 629): In that year, in July, Heraclius,
emperor of the Romans, and Shahrvaraz the patriciusof the Persians, met each other
at a pass in the north, which is named Arabissus Tripotamus. They built a church
there and called it Eirene (Peace) by name, and they spoke about peace there with each
other, and it was determined that the Euphrates was the boundary between them.
Thus they made peace with each other. (tr. M. Greatrex)
Nic. 17.1–19: Shahrvaraz, when he had heard that Khusro and Shiroe, Kaboes and
Hormizd had died, returned from the land of the Romans and wrote an apology to
Heraclius, (claiming) that what he had done to the Romans he had not performed
willingly but according to the order of the one who sent him, and he sought permis-
sion to come to him (Heraclius) and to present (himself) as a slave. After being reas-
sured by a sworn assurance from the emperor, he came to him and promised to give
him money from Persia with which he might repair whatever he had destroyed in the
land of the Romans. Meanwhile the son of Hormizd fell victim to a plot and was slain,
and Shahrvaraz sought from the emperor the Persian crown.
145 He granted him (this),
and they agreed between themselves that all the Roman territory under the Persians
should be restored to the Romans. When peace had been concluded, Shahrvaraz
immediately returned to the Romans both Egypt and all the eastern lands after with -
drawing the Persians (that were) there; and he sent the life-giving Cross to the
emperor. Heraclius conferred the dignity of patrician upon Nicetas, the son of
Shahrvaraz, and gave the latter’s daughter Nike in marriage to his own son
Theodosius, born of Martina.
146 ( … ) (tr. Mango, revised)
Sebeos 129–30/88, cf. Thom. Arts. 96/162 reports that Shahrvaraz agreed to evacuate the east -
ern provinces after Heraclius offered him the Persian crown and military support if he should
need it.
147 When they met (i.e. at Arabissus, presumably) Shahrvaraz pledged to return the
Cross as soon as he found it and to fix the frontier where Heraclius wished it. See Thomson and
Howard-Johnston 1999: 224.
The significance of the terms reported in Chr. 724, if correct, should not be underestimated: the
Roman frontier would have been shifted further west than it had been since the first century

AD . The whole of the Roman provinces of Mesopotamia and Osrhoene, including the city of
Edessa, so recently retaken by Theodore, would be handed over to the Persians, and Antioch
would lie within striking range of Persian armies. We think it more likely that its reference to
the Euphrates is an error. Cf. Flusin 1992: II, 290 and n.145; contra, Howard-Johnston 1999:
148 There are indeed scraps of evidence to indicate a continuing Roman presence on the
Tigris, as far south as Tagrit. See Fiey 1987: 98–9, Kaegi 1992: 153–
7. It remained for Shahrvaraz to wrest the Persian throne from the young Ardashir. From the
Khuzistan Chronicle (p.29) theChronicle of Seert (below), Sebeos 129/88, Hist. Her.432/231,
and Chr. 1234 , 103, 238/142, it appears that Heraclius provided military assistance to his ally.
The chronology of Shahrvaraz’s march to Ctesiphon is uncertain; the only well established date
is that he seized the throne on 27 April 630. Given that he had to assemble his army and march
from the Roman East to the Persian capital, he is unlikely to have reached it until the end of
629. See Flusin 1992: II, 306–9, Zuckerman 1995: 119 n.30, Thierry 1997: 172–6 (on the
general David), Thomson and Howard-Johnston 1999: 223–4.
Chr. Seert 93,PO 13.556: Then troubles (at the Persian court after the assassination
of Kavadh Shiroe) began to brew. And so they wrote to Shahrvaraz, who resided in the
country of the Romans in the service of the Emperor Heraclius. And they asked him
to return to them because he was the only survivor from the progeny of Sasan. But he
declined out of fear for himself, and because he did not relish violating his compact
with Heraclius that he would stay with him. When the latter knew of this, and of his
staunch loyalty to him, he gave him a large force
150 and sent with him a commander,
known as David. And Shahrvaraz headed towards al-Mada’in with the intention of
fighting with Ardashir, son of Shiroe. When he came near al-Mada’in, they closed the
gates before him and prevented him from entering it. But he continued to employ
deceitful devices until he had the gates opened, and so he entered (the city), killed
Ardashir and seized the kingdom from him. Because of this, the army’s commander-
in-chief was angered and a group of soldiers went over to him and refused to recognise
Shahrvaraz. So Shahrvaraz fought with him victoriously, and captured him, and thus
the kingdom incontestably became his. And Shahrvaraz honoured those of the
Romans who were with him and returned the wood of the Cross which Khusro had
taken from Jerusalem, and placed it in the box, which he despatched with David, who
had been sent with him from the country of the Romans.
151 Forty days later he killed
Shamta, son of Yazdin, and crucified him. 152 Then one of the relatives of Khusro
assailed him unawares, and secretly assassinated him. (tr. Shahîd)
Tabari, I, 1062/401, offers more details on how Shahrvaraz, with 6000 men, captured
Ctesiphon by securing the loyalty of the son of Adhur Gushnasp, Ardashir’s leading general,
and another commander. See Flusin 1992: II, 306–7.
The return of the Cross and the fall of Shahrvaraz (630)
The date of the return of the Cross to Jerusalem has been the cause of much controversy. As has
been noted, at some point in late 629 or early 630, Shahrvaraz captured Ctesiphon, and the
young Ardashir was overthrown. But not until 27 April did Shahvaraz make himself king. It is

possible therefore that he ruled as a regent for Ardashir before killing him, or that he ruled the
kingdom for some months before formally becoming king. The date of Shahrvaraz’s seizure of
power bears on the return of the True Cross, for it is generally now agreed that it was carried
into Jerusalem by the emperor himself on 21 March 630 and then deposited in the church of
the Holy Sepulchre. Since theKhuzistan Chronicle(p.29), Nicephorus (above) and the Chroni-
cle of Seert (above) all credit Shahrvaraz with the restoration of the Cross, it must be supposed
that he was in a de factoposition of power before his formal accession – whether as a regent for
Ardashir or simply as uncrowned king. See Flusin 1992: II, 293–309 (arguing for a regency),
Zuckerman 1995: 119 n.30 (noting that no ancient source refers to a regency) and Thomson
and Howard-Johnston 1999: 224 (accepting the regency).
The return of the Holy Cross to Jerusalem, escorted by Heraclius from Hierapolis, made a huge
impression on contemporaries. It was even asserted that the Cross had been preserved intact
throughout its time in Persia. Accounts are found in Sebeos 131/90, Geo. Pis. In Restitutionem
S. Crucis, Ant. Strat. 24 (516), Sophronius, Anacreontica18,Translat. Anast. 1, Theoph. A.M.
6120 (328.13–15), Nic. 18, Chr. 1234, 103, 238/142. See Frolow 1953: 94–105, Thierry
1981: 205 and 206 fig.1, Speck 1988: 336–77 (dating the restoration to September 628),
Flusin 1992: II, 309–19.
On 9 June 630 Shahrvaraz was assassinated and his place taken by Boran, a daughter of
Khusro II. To bolster support for her régime, the new ruler sent several prominent churchmen
as envoys to the Romans, including the catholicosIsho-Yab. They met the emperor at Beroea
and conceded all the territories which the Persians had gained – not only Mesopotamia and
Osrhoene, if these had indeed been ceded by the Romans (see p.227 above), but even those
which Maurice had annexed in 591 (Chr. Khuz. 30,Chr. Seert 93,PO 13.557–60, Translat.
Anast. 3, Sebeos 131/90).
154 The future had seldom looked better for Roman rule in the East.
See Sako 1986: 121–4, Flusin 1992: II, 320–1, Howard-Johnston 1999: 29 and n.92,
Thomson and Howard-Johnston 1999: 227.

(First part)
Some events from (Ec)cl¢sastik¢(i.e. from ecclesiastical accounts) and Cosmostik¢(i.e.
accounts relevant to secular matters) from the death of Hormizd, the son of Khusro,
to the end of the Persian kingdom.
Hormizd reigned for twelve years, and imposed the yoke (of servitude) on the nobles
and on all the world. One of the chiefs of his troops, Bahram by name, (of) Rhagae, who
had been sent by him to the regions of the Turks, rebelled against him; and having
assembled many troops, he prepared for war with the king. But when the nobles who
were at the royal court,
3who hated Hormizd, heard of the rebellion of Bahram, they
conspired with each other, took Hormizd from his throne, and put out his eyes; and in
his place they appointed his son Khusro. When the news reached Bahram, he was very
angry, not because he liked Hormizd, but because he (himself) had not performed the
deed. (Therefore), having prepared his troops and got ready for war with Khusro, he
proceeded to march against Khusro. Since Khusro saw that Bahram’s side was stronger
than his own, he fled from him and hurried along the south road,
4that is, through
Peroz-Shapur (Pirisabora), ‘Anat (Anatha), Hit and Circesium, and went to take refuge
with the Roman emperor Maurice. While he was making his journey in flight, the
catholicos Mar Isho-Yab did not set out with him;
5and (so) Maurice greatly blamed
Khusro, in that he was not joined by the patriarch of his kingdom, (p.16) and all the
more in that Mar Isho-Yab of Arzanene was a wise and clever man.
6Khusro developed a
great hatred of the catholicos, (firstly) because he had not set out with him, and secondly
because he had not met him on the road when he heard that he had set out, after
Maurice had given him troops. But the catholicosdid not meet the king because he
feared the wickedness of Bahram, in case he should ruin the Church and stir up a perse -
cution against the Christians.
7Maurice handed over many troops to Khusro, and they
set off to the east; when Bahram heard (this), he set off from Mahoza 8with his troops
and fled to Azerbaijan (Atropatene). 9Khusro, with Persian and Roman troops, encoun -
tered him (in battle); the Romans gained a victory and Bahram’s contingent was
defeated. But Khusro with great joy turned back (to Persia). For it was said that the
image of an old man appeared to Khusro, as he seized the bridle of a horse and was going
forth to war; and when he told his wife Shirin (of this) after his return from the war, she
said to him, ‘This is Sabrisho, bishop of Lashum’; and (Khusro) considered the matter
and was silent.

At that time the brothers Bindoes and Bestam, who were imprisoned by Hormizd,
left prison, and were of great help to Khusro, because they were from the family of his
11Then (Khusro) sent Bestam with a large army to the regions of the Turks,
but he (Bestam) left Bindoes at his (Khusro’s) court. When Bindoes rebuked Khusro
about matters of state, he (Khusro) planned to kill him; Bindoes fled, going forth in
the direction of his brother Bestam. But when he was passing through the region of
Azerbaijan, the marzbanthere heard (about him), made dinner for him, and seized
him; and he sent him to Khusro. Having learnt (of these things), his brother gathered
forces from the Turks and Dailamites and proceeded as far as Mahoza; but a certain
Turk plotted against him, killed him, and sent his head (p.17) to Khusro.
12As for
Bindoes, the king ordered that they (his men) cut off all the limbs of his right side; and
he sent him to Beth Lapat
13and crucified him there. They also placed the head of
Bestam on the neck of Shapur, the son of Bahram, who had opposed him (Khusro);
and they mounted (Shapur) (on) a camel and led him around the royal
court. But Isho-Yab, the leader of the Christians, because Khusro had a great hatred
indeed of him
14in that he had not set out with him for the territory of the Romans,
and furthermore on account of the slander of Timotheus the chief doctor of Nisibis, 15
strenuously protected himself from the king. After a little time, when the patriarch
went down to Hira of the Tayyaye to see Nu‘man, king of the Tayyaye, who had
become a Christian,
16and when he reached the place of Hira, he was struck down by
disease and died in the area whose name is Beth Qushi. But when Hind, the sister of
17heard (of this), she left Hira with the priests and the faithful, and with a
great procession they brought the body of the holy man into (the city); and Hind
placed it in the new monastery which she had built.
18The Church was for a time
without a leader, and then the synod was gathered by the command of the king to
elect the head. The king sent (a message) to them that they should bring forth
Sabrisho (the bishop) of Lashum and appoint (him) as (their) chief; they hurriedly
brought him forth and appointed him as their head.
19And he was greatly honoured all
his days, both by the king and by both his (the king’s) Christian wives, that is by
Shirin the Aramaean
20and Maria the Roman. 21
There was at Nisibis the metropolitan Gregory the Kashgarite. It is not possible for
the tongue to narrate the disputes and many quarrels which Satan caused between
these two good men (Gregory and Sabrisho).
22But (when) at Gregory was at
Nisibis for a short time, they excommunicated Gabriel the son of Rufinus because he
was well-versed in the course of the stars and the signs of the Zodiac;
23and they seized
(p.18) and brought over Gregory, the bishop of Kashgar. (Now) at Nisibis, which is
(also known as) Antioch of Mygdonia, which is named thus
24because of the gardens
and parks which are there, since it is of the Persians and the Romans,
stupid, troublesome and quarrelsome men gathered from every region, especially on
account of the famous school which was there. But there was (a certain) Henana of
Adiabene, an interpreter (of Scripture);
25(and) because (regarding) various matters he
had blamed in his teachings the ecumenical interpreter, 26the zealous Gregory did not
tolerate this. 27He wished also to correct clergy who corrupted their practices along
with the rest of the faithful, but they did not obey him. But they arrested a certain

deacon, who was called Bar Ta‘l¢(son of foxes), while sacrificing a white cock in a
grove which is outside the city; (Gregory) summoned and punished him. He also
disgraced and scattered everywhere certain monks who (had committed) disgraceful
acts (and) lived around Mount Sinjar; they were Messalians.
28From that time there
were many accusations against him from the Nisibenes and from those outside (the
city); (therefore) the king sent for him to come (to him), and ordered that he stay in
the Shahdost monastery. (Gregory) shook the dust from his feet at the gates of Nisibis,
and then departed.
29Mar Sabrisho wished to depose Gregory (from his office), but the
bishops did not give in to him; then the king ordered him to return to his land. 30He set
out (there) and built a monastery for himself in the region of Kashgar, in the place
which is called Bazza of the Euphrates;
31 there he founded a monastery, and he
brought many to the fear of God. For they say that after the abdication of Gregory it
was not accorded to Mar Sabrisho to perform miracles as previously. Then Nisibis seceded from Khusro; when the king heard (this), he sent the
nakhveragan, a royal leader, against them with (p.19) a great army and elephants, and
with him Mar Sabrisho. The inhabitants of the city shut the gates before him; then,
through the wheedling of the catholicos, and because thenakhveragangave an oath to
them that he would not repay them with evil, they opened the gates to him. But when
he entered, he went back on his agreement, and he took the notables who were with
them and tortured them; he plundered their houses, destroyed all they possessed, and
finally slew them by every manner of death. And (thus) the curse of Gregory against
them was fulfilled, and Mar Sabrisho himself understood this.
There was at that time (a certain) Gabriel Derustbadh 33from Singara, a chief
doctor, beloved by the king for the reason that he had drawn blood from the arm of
Shirin, and she had had a son and had called him Mardanshah, when previously she
had not had sons;
34Gabriel, although previously he had been a heretic, 35wished to
be enrolled in the side of the orthodox; 36but because he had put away the lawful wife
whom he had taken, who was a confessor of noble lineage, and had taken in
marriage two pagan wives, with whom he lived in pagan fashion, the catholicosurged
him to put away the pagans and to take (back) the lawful (wife), but he refused; then
he returned to the side of the heretics, and brought many evils upon the followers of
our side.
When Khusro was fleeing from Bahram and had reached Roman territory, he is
said to have asked Nu‘man, the king of the Tayyaye, to travel with him, but he
(Nu‘man) refused; he also requested a very noble horse from him, which he did not
give him.
38And Khusro also asked for the daughter of Nu‘man, who was most beau -
tiful, and Nu‘man did not agree; but he (Nu‘man) replied, ‘I would not give my
daughter as a bride to a man who enters marriage in the manner of beasts.’ Khusro put
all these things together and kept them in his heart. When (p.20) Khusro had a pause
from wars, he wished to exact vengeance on his enemies, and among them Nu‘man.
One day he invited Nu‘man to dinner, but in place of food he broke scraps made of
hay before him; and Nu‘man was very embittered, and sent (word) to his fellow
tribesmen, the Ma‘add.
39 They devastated many places and took captives from
Khusro; and they penetrated as far as the ‘Arab. 40When Khusro heard (of these

things), he was disturbed, and enticed him (Nu‘man) to come to him by all sorts of
means, but he refused. But one of Nu‘man’s interpreters, (who was) from the island of
41Ma‘ne by name, conspired with Khusro and said to Nu‘man, ‘The king really
loves you’, and swore by the gospel, ‘The king will not hurt you’. Mawiyah too, the
wife of Nu‘man, said to him, ‘It is more fitting for you to die with kingly status than to
live expelled and driven away from kingly status.’ But when he reached the royal
court, (Khusro) did not kill him, but ordered him to remain at the gate; afterwards,
however, it is said that he slew this illustrious confessor by deadly po
Then a certain man, Phocas by name, rose against Maurice, the emperor of the
Romans, and slew him together with his sons and wife; but one of his sons escaped,
Theodosius by name, and came to Khusro. He was received with great honour by
the king. He ordered the catholicosto admit him to the church, (where) the royal
crown was placed on the altar, and then placed on his head, according to the ritual of
the Romans.
43Then Khusro handed over an army to him (Theodosius), and he
went against the Romans. Phocas too sent many troops, who pitched camp in Beth
Washe, beyond the city of Dara;
44and they fought a battle with Theodosius and
crushed his troops. He reported to Khusro, ‘I do not have the strength to stand
against the Romans’. And Khusro set off in winter with many troops from Mahoza
and assailed Roman territory (p.21); and the catholicoswas with him.
45The forces of
Phocas went out against them, and after the detachments fell upon each other, many
on both sides were killed. They threw down a noose for Khusro,
46but one of his
(Khusro’s) bodyguards, Mushkan by name, cut it; on the next day, the battle (line)
was drawn up, and the Romans gave way before the Persians.
47(Then) the king
fought against Dara and constructed mounds; (soldiers) made tunnels beneath the
wall and set fire (there). And through the many things they did they broke through
the wall; and in this (city) they poured out blood like water. But the bishop of Dara
struck with an iron bar the vein which is known as the ‘catholicos’ of the body,
while he lay on his bed, the blood flowed out and he died. For he was fearful of the
king, who had sworn, ‘I will kill him by forty means of death’. From that time
Khusro had power over the Roman territory. Dara was crushed in the fourteenth
year of Khusro.
But when the king was fighting against Dara, a certain Radh 50came down to the
churches of Seyarzur 51and overturned them; and when the faithful, together with
their bishop Nathaniel, saw (this), they did not tolerate it; and they rose up against
Radh and expelled him. He came to Khusro at Nisibis, and spurred him on, saying to
him, ‘You are fighting on behalf of Christians, while I am persecuted by Christians.’
And without consideration, the king summoned Nathaniel, the bishop of Seyarzur,
and gave an order to imprison him for six years; and afterwards he crucified him. For
although Khusro in appearance showed favour to the Christians for the sake of
Maurice, he was, however, the enemy of our people.
But Mar Sabrisho was seized by a serious illness at Nisibis; and the king sent (a
message) to him and asked him to release Gabriel from his (sentence of) excom-
munication, but he refused. He made a will and sealed (it), and he ordered that they
carry him to his monastery. The Nisibenes, however, wanted to deposit (p.22) the

body of the holy man in their church, but the king refused them, since he had heard
the order of the catholicos; and his (Sabrisho’s) disciples placed his body on a camel,
and carried it to his monastery.
Then Gregory, (who was) from Porath, 54was madecatholicosat the instigation of
Shirin, who was a (fellow) native, although all the clergy, along with the king, sought
out Gregory of Kashgar, who had been expelled from Nisibis. But he undertook unac -
ceptable practices in his pontificate, lived on for a few years, and (then) died.
account of the scheming and hatred of Gabriel towards the Church, the Church
remained for a while without a leader and was not allowed a voice on account of the
accusations of Gregory; and they appointed in the Church Mar Aba the archdeacon,
(who was) from Ctesiphon, to lead it, an upright and wise man. The Church
remained for a long time without a head.
56But Gabriel, (who was) from Singara,
much threatened the orthodox, and expelled our people from the monastery of Mar
Pethyon and the monastery of Shirin and from other (monasteries), and these he colo -
nised with followers of the party of the heretics.
At that time Ionadab the Adiabenean was distinguished in the Church through the
freedom of speech which he had before God and (through) his love for the king 58and
he obtained a letter from the king by which he had power over all the mountain where
(the monks) lived in the monastery of (St) Matthew, the deceivers of Mosul.
when his desire had been fulfilled by the king – to expel and persecute them in every
way – the cunning plots of Gabriel did not allow (him to finish the operation). Bar
Hadbeshabba of Holwan
60 published his writing(s); and for practices of virtue
Shubalemaran of Karka de Beth Selokh was famous, and Aphrahat of (the region of)
the Zab (rivers) and Gabriel of Nehargul, a great man and performer of
(p.23) Then Gabriel roused the king against us, so that we should (have to) come to
debate with the followers of his party. And since there was no catholicosin the Church,
there voluntarily came down in person to the debate Ionadab the metropolitan of
Adiabene, Shubalemaran of Karka de Beth Selokh, George of Mount Izala, and also
the bishop of Nehargul and Sergius of Kashgar, (who was) from Tell Pahare.
held the debate in the royal court; Gabriel and the followers of his side came off worse,
while our orthodox men triumphed. The king reproved Gabriel, (urging him) to
desist from this impertinence, but he refused. But Gabriel uttered bitter insults against
the orthodox and denounced Gregory of Izala in the presence of the king, because he
had abandoned the law of the Magi and become a Christian and reviled (the gods)
Hormizd and Saturn. The king sent (a message) and put him in prison for one year;
then he crucified him in Veh-Ardashir (Seleucia), in the middle of the hay market.
The faithful then seized his body and deposited it in the monastery of St Sergius in
Notable at the royal court at this time was Yazdin, from Karka of the Garamaei; 64
this man was a defender of the Church like Constantine and Theodosius, and built
churches and monasteries everywhere, as an equivalent of a heavenly Jerusalem. He
was as beloved by Khusro as Joseph in the eyes of Pharaoh and more besides, and in
consequence of this he was famous in both kingdoms – that is of the Persians and of

the Romans. It is said that this Yazdin each day from one morning to the next used to
send one thousand staters to the king. 65
By practices of virtue there flourished at this time Mar Babai of (Mount) Izala. He
became the priest for the monastery after Rabban Mar Abraham of Kashgar. 66Many
industrious brethren came forth from the monastery, such as Mar Jacob, who
founded the monastery of Beth ‘Abe,
67Mar (p.24) Elias, who built a monastery on the
bank of the Tigris at Hesna ‘Ebraya, 68and Mar Babai bar-Nesib[n]aye. 69That blessed
man (Babai the Great) abandoned all he owned, and went up to the monastery of Mar
Abraham and became an anchorite; finally he departed from there, and set out to
build a monastery in the vicinity of the monastery (of Mar Abraham) itself. Large
numbers of brethren settled there; and although he was (one) of the notables in the
(secular) world, he chose to occupy himself in the holy labours of penitence, his prac -
tices surpassing the written word.
70When Yazdin learnt of this, he came to see him.
When he saw him (practising) every (form) of asceticism and dead (in) body, Yazdin
stood at his feet; (but) the holy man sent him away. After a while Yazdin brought to
him a golden cross which had many settings in it like rubies, and many emeralds, and
in its middle was placed a fragment of the actual wood of the cross on which our
saviour was crucified, together with other items to adorn his monastery.
71But Satan,
the lover of controversy, caused much bickering and great disputes between those two
fortified towers of the fear of God;
72and it did not settle down or end until the end of
their lives. The followers of Mar Babai the Great did not allow anyone to enter their
monastery until they should pronounce an anathema against the excellent Mar Babai
the Nisibene, whom they named Babai the Small (lesser). We have said these things
briefly, for their practices arose and blazed forth more than the sun, and bear witness
that they have many doctrines concerning their party and the purity which they
73Mar Babai the Great published many doctrines, debates and interpretations.
The holy Mar Babai the Nisibene published books on the practices of anchorites,
which greatly fascinate their listeners, as well as sermons on penance, which were
made in metrical (style).
(p.25) Then Khusro gathered his troops and entered Roman territory. He
appointed two commanders, (whom) he sent to the west. They crushed Mardin,
Amida, Martyropolis (Mayfarqet) and Edessa, made bridges over the Euphrates, (and)
crossed opposite Mabbug (Hierapolis). One of the commanders, whose name was
Shahrvaraz, hurried to Jerusalem and struggled to make (the inhabitants) open the
gates to him, but they refused. And he prepared an assault against (the city) and,
having built mounds against it and broken through the wall, he entered it. He seized
the bishop and the nobles of the city and punished them on account of the wood of
the cross and the items in the treasury. Because the divine power had broken the
Romans in the presence of the Persians, and because they had shed the pure blood of
the Emperor Maurice and his sons, God left no place hidden, but rather they
revealed (each one) to them (the Persians). They revealed to them the wood of the
cross, which had been placed and hidden in a herb garden. 76They made many chests
and sent (the wood) to Khusro with many vessels and items of great worth. When it
(the furniture) reached Yazdin, he held a great festival, and with the permission of the

king, he took a piece of (the wood for himself), and sent (the wood) to the king. The
king placed it as a token of honour with the sacred vessels in the new treasury which he
had built at Ctesiphon.
Then the Persian troops besieged Alexandria, which is enclosed by walls and
surrounded by the water of the Nile 78and has weighty doors. Alexander built it on the
advice of his teacher Aristotle. For a while they (the Persians) fought against it but
were unable to capture it; then a certain man, Peter by name,
79went over to them. He
had (earlier) come down to Alexandria from Qatar, so that he might be instructed in
philosophy; and he said to the Persian commander, ‘I will hand over the city to you’.
This Peter had one day found (p.26) at the bottom of some book in the public
archives of the city (a passage) written thus: ‘When affliction arises for Alexandria,
(the city) will be seized from the west gate which looks towards the sea.’ The Persians
prepared themselves, and seized (some) small fishing vessels, and embarked in them.
With the fishermen, they entered the city in the early morning while it was (still) dark,
protected by the fishing (vessels).
80They killed the guardians of the gates, opened (the
gates) to their allies, and announced the victory of Khusro on the wall. All the men
were seized by fear; and the wind took hold of many ships onto which had been loaded
the treasures of the Church and the nobles (of the city), so that they might be taken to
sea for safety, and brought them to the camp of the Persians. They sent the (treasures)
to Khusro, along with the keys of the city.
81But when an envoy bearing keys reached
Yazdin, he had golden keys made in the night in place of the (keys) and sent them to
the king, to entice the king further to favour him. But after Jerusalem had been
crushed, the detestable Jews threw fire into all the churches in it (the city). In this way
the conflagration destroyed the church of the Resurrection, which had been built by
Constantine and Helena and adorned with marble and priceless mosaics. In addition,
the sons of the crucified one (the Jews) went up to the Persian commander, and said to
him, ‘Behold! All the gold and silver and treasures of Jerusalem were placed beneath
the tomb of Jesus’. For they were contriving to destroy the place of (his) tomb. And
when he (the commander) had given them the opportunity, they dug down three
cubits around (the tomb), and found a box inscribed with these words: ‘This is the box
of Joseph, the counsellor who gave burial to the body of Jesus.’ But when the
commander heard of the Jews’ plot he threw them out sternly.
82When Yazdin heard,
he made it known to the king; and he issued an order concerning the Jews, by which
their wealth was confiscated and they were crucified. But as regards the body (p.27) of
Joseph, he (Joseph) had ordered before his death that his body be deposited at the side
of the grave of our Lord. Then Yazdin sought from the king (permission) to rebuild
the churches of Jerusalem and, having sent not inconsiderable moneys, he renovated
(them) with all their beauties; and everywhere he built churches as well as monasteries.
When the Persian commander also heard that many riches were to be found in the
church of St George of Lydda, 84he sent a large number of his soldiers, but they were
unable to enter, however, being held back by divine power. But at last he (himself)
went (forth) with great anger, and when he reached the door of the church, he urged
on his horse to enter by force; but the hooves of the horse stuck to the ground and he
could neither move on forwards nor go backwards. (Thus) God showed that although

he had allowed him to enter Jerusalem, (he had not done so because) his power was
weak, but in order to punish the Romans, who said that Khusro could not hold power
over Jerusalem. He (the Persian commander) vowed, ‘If I escape, I shall make an
object of silver in the likeness of the church of St George’. And so it turned out, and,
behold, this miraculous object has hung in the church until now.Afterwards, while Khusro was living in Dasqartha de Malka,
85the Caesar Heraclius
gathered many troops and came down against him; Khusro was disturbed by his pres -
ence and was in great fear. Heraclius went down to the northern regions; and he
ravaged, devastated and took captives (in) all the northern regions. When he drew
near to Dasqartha, Khusro fled from him and went to Mahoze. But it was said that
when he (Khusro) had determined to take flight from Dasqartha, he heard the sound
of a wooden bell; and he was disturbed and struck (p.28) his loins, and his stomach
was relieved. Shirin said to him: ‘Do not fear, O God’, to whom he (replied): ‘Now
am I a God, when I am persecuted by a priest?’ This he said because he had heard that
Heraclius had taken the office of priest;
86but he swore, ‘If I obtain victory, I will leave
no church and no wooden bell in my entire kingdom’. Yet the apprehension and
anxiety which had seized him from the sound of the bell (did so) because he supposed
that the Romans had wooden bells with them, and, behold, had come already to
Dasqartha. He (Heraclius) removed all the royal treasure (and) led off captives; and,
having devastated many places, he withdrew.
Then many of the (Persian) troops revolted against Khusro; but Shamta, the son of
Yazdin and Nehormizd, rose up and made Shiroe the son of Khusro king and gathered
many troops around him.
88But when Khusro heard (of these things), terrors seized
him and fatal pains struck him. Having abandoned his kingdom in the night, he
escaped; and he had two servant boys with him, who joined him from his household.
They fled and hid in the royal garden, and when he saw the troops following him, the
boys cried to each other.
89He (Khusro) reached out his hand into a hedge, so that he
might proceed further alongside it and escape, but he could not go on because of his
fear. And they seized him and led (him) off and imprisoned him in the house of a cer -
tain man whose name was Mihraspend;
90they gave bread to him to sustain (his) life.
Then Shamta and Nehormizd sought (permission) from king Shiroe the son of
Khusro to kill Khusro; and when he allowed them, they came and entered (the place)
where he was guarded. Shamta lifted the sword to strike him, and Khusro, weeping
before him said to him, ‘How did I wrong you, that you would kill me?’ And Shamta
did not strike him, but Nehormizd struck him on his shoulder with an axe, and
repeated (the blow) on the other (shoulder). His son Shiroe mourned for him, (p.29)
and they brought him into the tomb of the kings. This Shamta accomplished, because
when his father Yazdin had died, Khusro had plundered his house and inflicted many
tortures on Yazdin’s wife;
91but Nehormizd (acted) because he (Khusro) had slain his
father. 92But Khusro, the son of Hormizd, reigned for 38 years.
There was peace and tranquillity in the days of Shiroe, his son, for all the Chris -
tians. But the nobles of the king, together with Shamta, formed a plot and killed all
the children of Khusro; and Merdanshah, the son of Shirin, (was) among them.
Afterwards Shamta was accused in his (Shiroe’s) presence of wishing to take over the

kingdom; and he (the king) sent (men) to imprison him. Shamta then escaped, but
they followed (him) and came upon him in the city of Hirtha of the Tayyaye (Hira).
The king cut off his right hand and put him in prison.
Isho-Yab, from Gedhala, was made head of the Church. 95Although in early youth
he had married a woman, he had not been barred because of this and was made bishop
of the city of Balad. Finally he was elevated to the rank of catholicos, being endowed
with every virtue. Shiroe, when summer came, set off for Media, according to the custom of the king;
a pain in the stomach seized him, and he died on the journey.
96He had reigned eight
months. 97
Then they made reign in his stead Ardashir, his son – his son by the Roman Anzoy 98
– although he was still a small boy. 99But one of the leaders of the Persian army, who
was aligned with the Caesar Heraclius, Feruhan by name (Shahrvaraz), 100 when he
heard of the boy Ardashir’s rule, equipped Roman and Persian troops, entered
Mahoza, overcame the army of the Persians, and went up and killed Ardashir. He
brought Shamta, the son of Yazdin, out of prison and crucified him in front of the gate
of the church of Beth Narqus,
101 because he had insulted the daughter of this (p.30)
general (i.e. Shahrvaraz). 102 He dismissed the Romans who had accompanied him,
and they returned to Heraclius. 103 Together with them he sent to Heraclius the wood
of the Lord’s cross, which they had brought back from Jerusalem and had been depos-
ited in the Persian treasury, along with a countless number of presents.
104 Ardashir had
reigned one year and six months; this Feruhan (Shahrvaraz), the general who killed
Ardashir, reigned forty days.
105 As he was leaving Mahoza one day, a certain body-
guard struck him from behind with a spear, and he died and was reviled by all the
The Persians made Boran, the wife of Shiroe, 107 rule over them. When she came to
rule, she wisely sent Mar Isho-Yab the catholicosto Heraclius in order to make peace
with him. Cyriacus of Nisibis accompanied him (Isho-Yab), as well as Gabriel of
Karka of Beth Garmai and Marutha of the city of Qastra.
108 They were received in a
very friendly fashion by the Emperor Heraclius, who did for them everything they
109 Boran the wife of Shiroe, who ruled over the Persians, died by strangula -
tion in the end. 110 (tr. M. Greatrex)

The eastern frontier (363–630)
The epigraphic record of the eastern frontier is markedly uneven. Large numbers of inscriptions
from Syria and Phoenice Libanensis allow us to build up a fairly detailed picture of local
defences and reactions to enemy invasions in this region. But elsewhere, including Mesopo -
tamia, Osrhoene, Armenia and the Transcaucasus, very few inscriptions have been found. Inev -
itably, therefore, this chapter will concentrate on the southern part of the frontier, a region
where Arab raiders were often more of a threat than Persian armies.
1It should be emphasised
that we offer here only a selection of inscriptions, concentrating on those which are most often
referred to in the secondary literature, but have not always been translated. It is not our aim to
provide an exhaustive assessment of Roman defensive work along the frontier, which would
involve a systematic examination of literary and epigraphic evidence. See Mouterde and
Poidebard 1945: 237–40, Liebeschuetz 1977, Whitby 1986a, Gregory 1997.
Fourth–fifth centuries
It is difficult to discern any particular pattern to building work in the East in the late fourth and
fifth centuries, partly because most of our evidence is clustered in the sixth century. Under
Valens, however, there is record of building work at Amida, no doubt partly required by the
swelling of the city’s population with refugees from Nisibis.
CIL III.213, Amida (367/75): see Chapter 1.
In Syria, on the other hand, it has been suggested that some forts were abandoned by troops and
were taken over by hermits. At any rate the desert area around Chalcis became the haunt of
anchorites and Arab tribesmen.
2Other places, however, such as Zebed, at the eastern approach
to the Jebel Shbeyt, were strengthened in the second half of the fourth century. See Mouterde
and Poidebard 1945: 167. Further south, we have clear evidence for continuing investment in
frontier defences: the following inscription testifies to the building of a mansio, a hostel, to pro-
tect travellers in the desert. See Isaac 1992: 176–7, Gregory 1997: 9
IGLS 2704 = AAES355, Khan al-Abyad, south-west of Qaryateyn: 3A plain that is
dry indeed, and hateful enough to wayfarers, on account of (its) long wastes and (its)
chances of death close at hand, for those whose lot is hunger, than which there is no
graver ill, you have made, my lord, a camp, adorned with greatest splendour, O

Silvinus, guardian of the most stronglimesand of the cities by the trust of (our) mas -
ters, revered in all the world; and you have contrived that it abound in water celestial,
so that it may be taken under the yoke of Ceres and of Bacchus. Therefore, O guest,
with joy pursue your way, and for benefit received sing with praise the doings of this
great-hearted magistrate, who shines in peace and in war. I pray the gods above that
he, enjoying higher rank, may continue to found for (our) masters such camps,
Map 6Provenance of inscriptions discussed in Chapter 16

arduous though they be, and that he may rejoice in children who add honour to their
father’s deeds. 4(tr. Prentice, revised)
The fifth century, in general a peaceful one for the eastern frontier, provides occasional glimpses
of defensive building work.
IGLS2090, Burj el-Qai: + In the year 768, in the month Xandikos (April 457),
through the effort of Maximus and Bassumus (this lintel) was erected. 5
The following inscription from the fifth century testifies to the problems besetting communi -
ties of Phoenice and Syria (and even Palestine), among which were both Arab raiders and the
Roman troops entrusted with guarding the region. See (e.g.) Mayerson 1989, Isaac 1992:
235–49, Whittaker 1994: 135–41.
IGLS 2501bis, region of Emesa: 7… of asylum … But we wish that this same (?) place
be inviolate and unaffected by violence in every way possible, and in no way troubled
by anyone in any way, and we decree that for the future (its inhabitants) should resist
any damage, ill-treatment or annoyance either by the armed forces or (on the part) of
those of the Saracen race or of anyone else attempting to wrong them, and again that
they should have certain rights of ownership under whatever form it might be, com-
pletely immune from tax save for the capitation tax … Donation of the most pious
(emperor ?) … granted … ’
The reign of Anastasius (491–518)
Under Anastasius a more coherent policy of defence emerged, most notably in the construction
of the powerful stronghold at Dara on the frontier (see Chapter 5).
Granaries and depots were important to troops and inhabitants in the event of enemy invasion.
It is clear that they were constructed not only by imperial officials but also by churchmen.
role of the church in public building projects is noteworthy: churchmen were used on several occa -
sions to supervise building work even of a military nature (as at Dara, but also at Scythopolis after
the Samaritan revolt of 529–530. See Di Segni 1995: 321–2, Trombley and Watt 2000: xlviii-l,
Feissel 2000: 88.
IGLS 2081, Arethusa: (Under the most) holy bishop Andreas this most beautiful
(public) granary (was erected) from its foundations on the 9th of the … month, year
810, indiction 6 (498).
In general, funding for defensive works came either from city or provincial funds. The former
were limited, and so recourse to the latter, which were administered by the province’s governor,
was frequent. Often the emperor would be thanked for channelling this money into a city’s
infrastructure. See Di Segni 1995: passim, esp. 330–1, Jones 1964: 759. Outside the cities, mili -
tary officials might engage in the construction of forts or barracks, as might also the church. But
here wealthy local individuals might also come to the assistance of their region, as in this case

from Taroutia Empor¤n (modern Kerratin, south of Chalcis). 10See Di Segni 1995: 326–8,
Gregory 1997: I, 93, Di Segni 1999: 150–2. On the context of the inscriptions below see Foss
1997: 234.
IGLS 1630 = PUAES III B 992, Taroutia Empor¤n, a lintel from a tower: Thus
wisely guarding his country, John, abounding in good counsels, supplying the gold
unsparingly, provides a tower, a place of safety for (his) friends, through the effort of
Paul (the) deacon, in (the) year 821 (
AD 509/10), in (the) name of God (the) Saviour.
(tr. Prentice)
IGLS 1631 = PUAES III B 993, Taroutia Empor¤n, a lintel found close to the
preceding inscription: He who in the emperor’s business has proved a faithful servant
in toils, John, has made this guardhouse secure, a defence of safety in this region,
being son of Azizos by name, brother of Paul … (tr. Prentice)
IGLS 1725, Umm et-Tine, south of Taroutia Empor¤n, a lintel: May the Lord
guard your entrance and your exit. (This building) was (built) under (the direction) of
the most glorious [C]apito[l]inus. Indiction 10, in the month Gorpieus, year 828
(September 516).
The reigns of Justin I, Justinian and Justin II (518–78)
A wealth of evidence for building work during the reigns of Justin I and Justinian survives. On
the one hand, there is the detailed account provided by Procopius’ Buildings (De Aedificiis)
II–III, recording the extensive building activity of Justinian on the eastern frontier. At the same
time, the epigraphic record from Syria for this period exceeds that of any other in antiquity:
northern Syria was more densely inhabited at this time than ever before.
12The inscriptions fre-
quently attribute the initiative for the defensive work to individuals who do not appear to be
imperial officials; much work in particular seems to have been undertaken by churchmen. The
two sources – inscriptions and Procopius – must both be taken into account in assessing impe-
rial involvement in frontier defence; although there are few references to imperial involvement
in the epigraphic record, it is clear from Procopius that much work was undertaken by Justinian
in the East.
In general, it is only inscriptions which allow Justinian’s fortification work to be dated:
Procopius’ Buildingsmerely ascribes a wide range of projects to the emperor without assigning
any date to the work involved.
14Three phases may be distinguished within the period 518–578.
(1) 518–32
In this phase there was a considerable amount of defensive work undertaken in Syria and along
the frontier with Persia, see Chapter 5. Fortifications were also restored further south, as at
Scythopolis: see Di Segni 1995: 318–19, translating the relevant inscriptions. An uneasy period
of peace with Persia prevailed up to the late 520s, but raids by Lakhmid Arabs presented a sig -
nificant threat to Syria. See Whitby 1986a: 727 (who regards 518–27 as a period of peace),
Lauffray 1983: 38–9 and 141 (suggesting that work was also carried out at Zenobia in this

IGLS481 = PUAES III B 1105, Kfellousin, a four-storey tower: + In the year 570
(?), the third of Xanthikos, indiction 15 (April ?522) through Sabatius, deacon, con -
tributing of his own toil.
IGLS 2155, basalt lintel from Garion (Gour), between Emesa and Raphanea :
station-post (metaton) of St Longinus and St Theodore and St George, year 836 (
524/5). 16
IGLS 1768, Hawa, north of Ruweyda, among important ruins, lintel of a tower:
The tower was erected (?) with God’s (help) for the safety and health of the brothers
and servants (?), year 840, 6th (?) indiction (528/9).
Seyrig 1950: 239, Palmyra: Under the supervision of Flavius Platanius Serenianus
the most eminent duxof the East … 18
Roussel 1939, Hierapolis: inscription commemorating the Eternal Peace. See p.97
(2) 532–40
During the period of the Eternal Peace, there was little building work in the East. Justinian con -
centrated his efforts in the West, and by the time of Khusro’s invasion in 540, Roman defences
were in a poor state. See Chapters 6–7.
IGLS 1789, Ruweyda, lintel of a tower: + Chi-mu-gamma. Lord, help your servant
Thomas (the) deacon. + Year 851, indiction 3, Jesus Christ. (
AD 539/40). 19
(3) 540–78
Following Khusro’s devastating invasions in the early 540s, more effort was invested in fortify-
ing the eastern provinces, particularly towards the end of Justinian’s reign (the early 560s). 20See
Whitby 1986a: 728–9, Lauffray 1983: 35–7, Sauvaget 1939: 122, Tate 1996: 334. However,
the construction of city defences and forts could not compensate for a general reduction in
troop numbers in the region. The Roman presence on or near the Strata Diocletianahad been
decreasing well before this period, and Justinian intensified this trend by relying on the Jafnid
kings to defend much of the southern section of the frontier. In noting the inscriptions below
therefore, we must keep in mind that ‘there is practically no inscriptional evidence for the pres -
ence of the Roman army’.
SEG 41.1510 = Mango and Mango 1991: 466, perhaps from Tell Beshme, possibly
from a work of fortification: Under our lord of the world, Flavius Justinian, eternal
Augustus and the most glorious duxThomas, through the effort the most glorious
comes domesticorum Cyrus, (this work) was completed in the month of Audynaeus,
indiction 5.
IGLS 1889 = PUAES III B 871, K¤me Olban¤n (Halban), north-east of Epiphania,
lintel of a tower facing north: + In the year 854, the month of Dystrus, indiction 6
(March 543). + (This is) the work of John and Symeon, stonemasons from K¤me

Humann and Puchstein 1890: 405 no.5, Constantia: Thomas, the most holy
bishop, putting his hope in God, began the labour (of building) this granary in the
month of Artemisium, indiction 6, of (the) year 854 (May 543).
IGLS145–7, Cyrrhus: see p.111.
IGLS 1598, Idjaz, about 3 km south of Taroutia Empor ¤n, lintel on a wall that termi -
nates in a tower: I shall not fear the myriads of people besetting me on every side,
because the Lord is my protector. Lord my God I hope in you. Save me from those who
persecute me and deliver me. You, Lord, will protect and keep me. Protect me, Lord,
like the maiden of (your) eye. In the shadow of your wings shelter me. Some (enemies)
are in chariots and others on horses, but for our part we glory in the name of our Lord
God. Play the man and let your heart be strong, all of you who hope in the Lord. (Chi-
Rho). I am a house of peace, keeping the children of prosperous people safe with walls of
stone. + In the tenth indiction of the year 858 (546/7).
25(tr. Trombley)
IGLS 1809, Ma‘an, east of Apamea, lintel of the west gate of the fortress in three
fragments: Justinian, our most pious and gloriously triumphant emperor, who saves
all the cities by his unsparing munificence, restored this fort also in the year 859 (
547/8). Many years to the most distinguished count John, who takes delight in
building. Many years to Theodore the renowned a secretis. 26
IGLS348 = AAES 305, Chalcis, on the east side of the north tower: With the aid of
God the whole eastern face was built from its foundations out of the pious generosity
of our most serene … master Flavius Justinian, Eternal Augustus and Emperor and
under the supervision of Longinus, the most glorious and most renowned former gov-
ernor, ex-consul and most glorious magister militumand of the most glorious former
consul Anastasius and of the most magnificent and illustrious engineer Isidore in the
14th indiction, year 862 (= 550/1).
IGLS 1811, Qasr el-Mouharram (Qasr el-Berouj), north-east of Emesa, tower B,
lintel of the gate: Christ Jesus, become for us a protecting God, a house of refuge and a
mighty tower in the face of the enemy. Raise up this house, plant in it the glory of your
name for eternity (through) the prayer of Mary the Mother of God and all your saints.
Amen. This lintel was placed in the year 862, in the month of Artemisius, indiction 14
(May 551).
IGLS 1682, Androna, lintel across the main gate: + This is the gate of the Lord. Just
men shall enter within it.++Itiscustomary for others to serve the many by (the)
contribution of funds. May you, Thomas, best and most wonderful (of men), illumi -
nate the daily life of your fatherland. You appear as a saviour through the acts in which
you are prudent, while God our Saviour takes care to share in your plans. + + We laid
the foundations of the fortress with (the help of) God, through the munificence of
Thomas and the effort of Jacob his nephew in May, on the twentieth day, in the sixth
indiction of the (year) 869 (
AD 558). + + The lintel was set in place in November, on
the first day, with the help of God in the eighth year of the indiction of the year 871
(November 560). + +
29(tr. Trombley, revised)
The peace concluded in 562 did not mark an end to defensive work in the East; the Romans had
perhaps learnt their lesson after the Eternal Peace of 532. 30

IGLS1841–4, Qasr ibn Wardan: extensive ruins have been discovered here, in a style
more similar to buildings in Constantinople than surrounding Syrian constructions.
IGLS 1842 offers a date of November 564; recent work indicates continuing
construction under Justin II, apparently under the supervision of a strat¢lat¢sGeorge.
The site contains a church, a palace and barracks for troops. See M. Mango, ODB
III.1764 (with references), C. Mango 1976: 151 and now de Maffei 1995: 109–12.
IGLS 1726, Tell Hazne, about 18 km north-west of Ruweyda, lintel of a gate of the
tower: + I, Thomas, periodeutesby the grace of God, having prayed and besought God
for a means of atonement for my sins, erected this tower for the glory of his name,
praising (him) for its foundation(…).Indiction 13 of the year 874 (563). + (The)
Lord of the armies (is) with us; our protector is the God of Jacob. Holy God, holy
(and) mighty, holy (and) immortal, (who) was crucified for us, have mercy on us, who
was crucified for us, have mercy on us. (In a cross in the middle: Light. Life. Alpha.
IGLS 1743, Abu Habbe, north-east of Apamea, a fortress of which only the north
wall remains: … (The God of the armies is with us); our protector is the God of Jacob.
+ In the name of the father and of the son and of the Holy Spirit this good work (was
accomplished) under Macedonius through John, his notarius, in the month of
Daisius, indiction 14, year 877 (June 566).
The role of the Ghassanids on the frontier
The Ghassanid allies of the Romans assisted in defending the frontier not only with their forces,
but also by undertaking building work themselves. See Shahîd 1995: 489–518, cf. 258–61,
Foss 1997: 250–2, Sartre 1982a: 183–8, Key Fowden 1999: 149–73 (on the so-called
praetorium of Mundhir the Jafnid at Sergiopolis).
Waddington 1870: 2562c, al-Burj (just south of Dumayr), on a stone on top of the
entrance to the fort: Flavius Alamundarus (Mundhir), (the) renowned patrician and
phylarch, giving thanks to the Lord God and saint Julian for the sake of his salvation
and that of his most glorious children, built (the tower).
The late sixth and early seventh centuries
Defences continued to be worked on in the period after Justin II (from 578 to the 610s),
although the number of inscriptions decreases. Whether it should be inferred from this that less
work was being carried out is uncertain.
IGLS 288 = AAES318, Anasartha, lintel: By the gifts of (his) majesty (the) city,
despising the inroad of the barbarians, set up at its gates its benefactors, (the) Saviour
Christ, (her) gloriously victorious sovereigns, (the) renowned (commander), the pre -
fects of the praetorium, also (?) its most holy bishop, (and the?) most glorious

engineer, in the month Gorpieos (September), in the 906th (?) year, indiction 13. +
Jesus Christ, Emannuel. + God over all. 36(tr. Prentice)
Imperial involvement continued nevertheless, as the following inscription, and others from the
same site, clearly attest. 37
IGLS2125: Gagar al-Amiri, south-west of Arethusa, part of a huge lintel: (in accor-
dance with) the decision of Constantine (Tiberius Augustus?), (as) the (aforemen -
tioned) Tiberius, (the emperor) of pious reputation (and memory, and as too) our
Emperor (Maurice) beloved of Christ, having inherited the piety of his aforemen -
tioned father (decreed), and once Gregory the venerable (patriarch was involved), the
surrounding wall (peribolos) (was erected) from its foundations by the generous (?)
(gift of the most beloved of God) Sergius, son of Leontius, presbyter and para-
monarius … year 894 (582/3).
IGLS281 = AAES 319, Anasartha, city gate: + (Phocas) and Leontia, 39our most
pious sovereigns, O Lord protect! + A pious branch that sprang from noble stock,
Gregory, the renowned, and adorned with the fruits of his virtue, presented to God
this wall also, in sparing his own country (the expense). Indiction 8, in the 916th year
Building work went on in the early seventh century, and the inscriptions thus provide evidence
for Roman resistance to the Persian invasions.
IGLS 271, Mu‘allaq, north-east of Anasartha, a monastery church: + Chi-Mu-
Gamma. (Chi-Mu-Gamma). ++ This is the gate of the Lord. The just shall enter
within it. Upon this rock I shall build my church, and the gates of hell shall not prevail
against it. + This house of St Barapsabba was built through the efforts of the most distin-
guished … in the tenth indiction in the year 918 (606/7), … being the stonecutter.
(tr. Trombley)
Worthy of note too is IGLS2068 from Bserin, south of Epiphania, recording construction of a
barracks dedicated to St George in November 928 (616).

1 Cf. Robin 1996: 698 n.118 and Key Fowden 1999: 142.
2 Cf. Garsoïan 1998b: 1110, 1998c: 1158.
1 Cf. also Paschoud 1979: II, 218 n.91 (noting that several of the regions surrendered to
the Persians were not east of the Tigris). The precise status of Armenia in the treaty of
363 is unclear, and may well have been deliberately left ambiguous: see the perceptive
comments of Seager 1996: 283–4 and idem 1997: 267. For a general account of events
see (e.g.) Curran 1998: 78–9; on the chronology of the withdrawal see Dodgeon and
Lieu 1991: 237, with Matthews 1989: 185–6, Dillemann 1962: 308–12 on the
2 Turcan 1966 considers the differing attitudes of the sources. Whitby 1988: 205 n.14
believes that attitudes softened as time passed, but the passages of Agathias and Joshua
the Stylite below imply otherwise. See also Chrysos 1993, Teixidor 1995:
3 Reading memorabant with Fontaine 1977: I, 195, cf. II, 254–5 n.636.
4 The Tigris.
5 Surena was the name of one of the seven most noble families of Persia, but, like other
family names, was frequently misunderstood by Roman sources as a title. See
Christensen 1944: 105 and n.3. On the Persian initiative see Chrysos 1993: 168–9
(noting that Ammianus is supported by other sources).
6 A detailed analysis of this section of Ammianus is provided by Seager 1997: 266, who
notes some internal inconsistencies (over the strength or weakness of the Roman
7 The four days are 8–11 July. As Fontaine 1977: II, 256–7 n.645 and Matthews 1989:
186 note, Ammianus is guilty of serious exaggeration here: Corduene lay some 280 km
away and had been evacuated by the Romans four years earlier, cf. also Dillemann 1962:
8 I.e. the emperor more commonly known as Galerius.
9 Location uncertain, but Fontaine 1977: II, n.648, following Dillemann 1962: 213–14
identifies it with Rhabdion. Ball 1989 proposes an identification with Seh Qubba on the
Tigris in Iraq. As noted above (n.1), not all these regions are on the eastern side of the
Tigris; cf. also Chrysos 1993: 174–82, 187–8, noting that, contrary to what Ammianus
alleges, Shapur did not obtain all that he wanted. See Dillemann 1962: 219–20 on the
fifteen fortresses.

10Procopius was a relative of Julian, who in 365 unsuccessfully attempted to wrest power
from Valens. Cf. PLREI.742–3.
11 Reading nimiumwith Fontaine 1977: II, 259 n.650 for nimia.
12 A district of Atropatene: see Fontaine 1977: II, 29 n.58 and Dillemann 1
962: 300–1.
13 So Seyfarth. Fontaine 1977: I, 198, has the more likely Nevitta, on whom see PLRE
14 Dillemann 1962: 304–5 and fig.18 for the Roman route back. On the duration of the
peace see Chrysos 1993: 186–8, arguing that it was a crucial point gained by Jovian (and
that Nisibis and the other cities were therefore liable to be returned to Roman territory
after this time). We find this point unconvincing: the peace may have been intended to
last thirty years, but there is no reason (despite later Roman claims) to suppose that any
territory was ever due to be handed back.
15 Millar 1993: 102, 129, on these earlier attacks on Hatra. By the 230s, however, a Roman
garrison had been installed there.
16 Dillemann 1962: 310–11 on the Roman route here and Fontaine 1977: II, 264 n.663 on
the types of food mentioned by Ammianus.
17 30 km east of Singara, cf. Dillemann 1962: 311 and Fontaine 1977: II, 26
4–5 n.664.
18 The text of this may be found also in the Suda, I.401; a similar extract in Exc. de Virt.63 =
Joh. Ant. frg.181.
19 Libanius has just referred to earthquakes in Syria.
20 Very similar complaints may be found in Or.24.9 (tr. in Dodgeon and Lieu 1991: 261).
21 This reference to Roman defeats is directly at odds with Ammianus’ account, XXV.7.5.
22 The agreement to hand back Nisibis after 120 years is a clear fabrication, cf. Blockley
1992: 84 and 215 n.37. But the story persisted, finding its way into the Roman archives,
it appears: so Justin II claimed to Khusro I in the 570s, cf. Mich. Syr. X.1 (below, p.140).
See Chrysos 1976: 31–2, Luther 1997: 99 and below (d).
23 Or ‘generalship’: the Greek is ambiguous.
24 The names seem to have been somewhat garbled by Zos. Zalene is presumably a
misunderstanding of Arzanene, and Moxoene is omitted. Cf. Blockley 1984: 35 for a list.
The Greek term ethn¢has here been translated as ‘provinces’, although it literally means
‘peoples’. These Armenian principalities or satrapies were known as ethn¢(Latin gentes),
and were not technically provinces: cf. Paschoud 1979: II, 218 n.91 and Chrysos 1993:
174–6. They had not been under direct Roman control.
25 The tone of this excerpt, and the list of previous Roman surrenders contrasted with this
one (III.32), is more pessimistic than hostile. Cf. Paschoud 1975: 188–97 for a
comparison of Zos.’s treatment of the peace with other sources (esp. Ammianus,
Eutropius and Festus).
26 A reference to the three eastern provinces annexed by Trajan in his Parthian campaign,
but given up voluntarily by Hadrian upon his accession in 117. Millar 1993: 100–2 notes
how numerous fourth-century sources make this same comparison.
27 We take this as a reference to contemporary debate about the peace treaty and the events
leading up to it.
28 An allusion to Herodotus VIII.6, where the expression is used of Persian expectations of
their victory over the Greek navy shortly before the battles of Artemisia and Salamis in
29 An ironic criticism of Julian rather than Jovian.
30 Essentially the same account in Chr. Pasch.553–5; a more garbled version in Joh. Nik.
31 As Turcan 1966: 884, 886 notes, Theodoret is plainly in error in making such positive
claims for the state of the Roman army on its way back, as is clear from the eyewitness
account of Ammianus.

32Cf. Drijvers 1994: 201–14. On the generally more positive attitude of the Syriac sources
towards Jovian, see Teixidor 1995: 503–4.
33 I.e. the Romans had caused the more damage. Cf. Chrysos 1993: 171–2 on this
tradition, which Chrysos is inclined to accept.
34 Cf. Chrysos 1976: 31 and Nöldeke 1874: 280 on these details; also Josh. Styl. 7 (above)
and n.22 on this invention of a time-limit.
35 Cf. Ehling 1996 on Jovian’s emissions (for which see Kent 1981: 393 nos.109–12, 424
no.229); also Chrysos 1993: 170.
36 Tabari is here drawing on the Julian Romance, cf. Nöldeke 1874: 291–2 and Chrysos
1976: 31.
37 Cf. Teixidor 1995: 504–5 on this population transfer. He emphasises rightly that neither
the departure of many Romans from Nisibis nor the arrival of immigrants from the east
had a significant impact on the character of the city.
38 See Matthews 1989: 1–4 for an excellent modern evocation of the scene. Ammianus and
Ephrem were certainly eyewitnesses; Zosimus himself, equally certainly, was not, but he
is clearly relying on a first-rate source which shows notable parallels with Ammianus. Cf.
Paschoud 1979: II, 226 n.97, 228–32 n.99. See below on the work at Amida. Edessa was
another place of refuge for those forced to move, as is noted by Chr. 724 (in vol.1); see
Chrysos 1976: 27–8 and 27 n.1 on the destinations of the refugees.
39 Fontaine 1977: II, 269 n.678, following Dillemann 1962: 312, identifies the place with
40 If the reading ad Hibitamis accepted, it is probably another reference to Thebetha, cf.
Fontaine 1977: II, 271–2 n.686 and Dillemann 1962: 312.
41 The implication seems to be that the Persians tolerated the Christian community in
Nisibis. Presumably either some people remained in the city or others from the
countryside nearby were moved there, along with the Persians tranferred to the city (on
whom see Tabari, p.9 above). This would account for the continuing pro-Roman
sympathies of the city, cf. Blockley 1992: 28.
42 The body of Julian on its way to Cilicia. The city referred to is Nisibi
43 Chr. Pasch. 553–4 offers the same account.
44 Cf. Ps.-Dion. and Mal. (above) on the resettlement of the citizens of Nisibis at Amida;
Gabriel 1940: 178–82 notes an extension to Amida’s city walls around this time (see
below), cf. Oates 1968: 106, Gregory 1997: II, 65. Other Nisibenes, such as Ephrem,
emigrated to Edessa. See Chrysos 1976: 27 n.1.
45 Vanderspoel 1995: 168 for the date.
AD 370. In fact Valens arrived in the spring, and was back in Constantinople in winter
370/1; cf. Barnes 1998: 251. An earlier trip to Antioch (in 365) had had to be abandoned
on account of Procopius’ revolt, cf. Gutmann 1991: 166 and Greatrex 2
000a: 35–6.
47 Dillemann 1962: 105 connects this cession (otherwise unattested) with the return of
Rhabdion (identified with Ammianus’ Castra Maurorum, 304–5) to the Romans. No
other source knows of a peace treaty under Valens nor of the handing over of territory in
Mygdonia (i.e. the area around Nisibis).
48 Bird 1986 argues that the Breviariacomposed by both Festus and Eutropius and
presented to Valens in 369/70 were intended to incite the emperor to exact vengeance on
the Persians. Note (e.g.) Festus, Breviarium30.2 (the closing lines of the work, hoping
for an imminent victory over the Babylonians, i.e. the Persians).
49 As Paschoud 1979: II, 355–6 n.128 argues, Zos. must be referring to Valens’ journey in
spring 370. Zos. rightly refers to the emperor’s visit to Hierapolis that year – the
traditional mustering point for eastern campaigns – but fails to record his trip back to
Constantinople in winter 370–1. See Barnes 1998: 251–2 on Valens’ movements at this
time. The case mentioned at the end of 13.2 took place in winter 371/2.

50Vanderspoel 1995: 177 for the date (March 373; March 374 is also possible) of
Themistius’ speech.
51 Themistius Or.11.149b appears to refer to Valens being in Mesopotamia in late March
373. CTh XIV.13 places him at Hierapolis in August that year (cf. Seeck 1919: 245 and
PLRE I.212), although the date of the law is not entirely secure.
52 The version translated here is that printed by Gabriel 1940: 136. On archaeological
evidence for the expansion of Amida at this time, see n.44 above.
53 For a useful recent assessment of the role of the Arabs in this period see Isaac 1998a:
444–52. Graf 1989: 350 cautions against views of ‘increasing pressure’ on the frontier
here. A brief and judicious summary of the debate surrounding the term Saracen is
provided by Bowersock in LA680–1; cf. Matthews 1989: 349.
54 Mayerson 1980 is rightly sceptical about how much reliance can be placed on the
surviving accounts of Mavia’s uprising, since their primary interest is doctrinal matters.
Although the passage from Socr. below refers to Valens’ departure from Antioch (378),
this seems too late for the outbreak of the revolt, cf. Woods 1996: 268, putting the
ending of the revolt in 377, shortly before Valens’ departure from Antioch (dated to 378
by Barnes 1998: 253 and n.16). Note also that a contingent of Arabs was present near
Constantinople shortly after the battle of Adrianople, Ammianus XXXI.6.5–6 (and
perhaps before, cf. Shahîd 1984: 175–83). The account of Thdrt. HEIV.23.1 has been
omitted, since it is of little worth, as too is that of Theodore Lector and those who
followed him: cf. Shahîd 1984: 197–8 and Bowersock 1980: 481–2. The Tanukhids
seem to have been replaced by the Salihids as the chief allies (foederati) of the Romans
around 383. Shahîd 1984: 203–16 rather overplays the meagre information provided by
Pacatus, Pan. Lat. II (XII) 22.3 concerning a successful campaign against the Arabs at
this time.
55 Lucius was the Arian successor of Athanasius as patriarch of Alexandria, who oppressed
his orthodox opponents. Rufinus, writing in the first decade of the fifth century, is the
first source to write on Mavia; as Mayerson 1980: 124–6 notes, he was more interested in
the monk Moses and events in Alexandria than Mavia.
56 Socrates goes on to describe how the (orthodox) monk Moses was appointed as bishop
for Mavia and her tribesmen.
57 Soz. may be exaggerating in the extent of Mavia’s raids, but cf. Shahîd 1984: 146. On the
term klima here used for ‘region’, cf. ODBII.1133 (misunderstood by Shahîd, loc. cit.).
58 Woods 1998: 329 identifies the magister equitum et peditum per Orientem as Julius and
his subordinate with Maurus, dux Phoenices.
59 cf. Bowersock 1980: 490–3. See also Millar 1998: 85 on the importance of Sozomen’s
evidence here, noting that it is the first-ever reference to poetry in Arabic. Frye 1977: 11
oddly interprets the word ‘odes’ as a non-Greek term meaning ‘r
60 As Sartre notes, Sozomen’s reference to ‘the present reign’ is ambiguous, since it could
refer to Valens (in whose reign the episode of Mavia took place) or Theodosius II (in
whose reign he was writing). Isaac places the episode ‘in the last quarter of the fourth
century’. Shahîd 1984: 141 identifies Zocomus as the ‘first of the Salihids’, the
confederation allied to Rome during the fifth century (doubted by Sartr
e 1982a: 143).
61 Weissbach, REII.1 (1920), 618–19 and II.5 (1934), 1922–3 and Matthews 1989: 51–2
on Resaina. Since it had a bishop earlier in the fourth century, it was clearly not an
insignificant place even before Theodosius’ work, cf. Dillemann 1962: 107. Mal.’s claim
that Theodosius I extended the walls of Antioch is a confusion with Theodosius II,
however: see Downey 1961: 437 and 1941. Van Esbroeck 1997: 364–5 attributes the
foundation of both this Theodosiopolis and that in Armenia (see Chapter 4) to
Theodosius I, dating them to 383/4. For Litarba and Gindarus see Talbert 2000: 67 D4.
62 Valens perished in 378 at Adrianople, while Shapur II died in 379, to be succeeded by his
brother Ardashir II (379–83). Not all the embassies sent during this period are noted

here. That undertaken by Sporacius is to be found in Chapter 2; that by Stilicho
(Claudian,De cons. Stil. I.51–4) cannot be placed with certainty, cf. Blockley 1987: 230
and n.35.
63 The passage comes between a report concerning the death of King Athanaric (Jan. 381)
and one concerning the elevation of Arcadius to the rank of Augustus (J
an. 383).
64 This oration is dated to 1 January 383, cf. Vanderspoel 1995: 205.
65 i.e. previous emperors.
66 Cf. Seeck 1920: 453 on this passage.
67 Parallel notices in Chr. Pasch.563, Hyd. a.384, Cons. CPa.384, Socr. HEV.12.2. See
also Marc. com. a.385, a possible allusion to the division of Armenia, and certainly
evidence of further negotiations with Persia; cf. Croke 1995: 59.
68 Nixon in Nixon and Rodgers 1994: 475 n.73 rightly connects this passage with the
embassy of 384. Pacatus later (32.2) seems to allude to the division of Armenia; see
Chapter 2 below.
69 The brief notice comes just before Theodosius’ campaign against Maximus in 388.
Libanius, Orr.19.62, 20.47 confirms that the Persians were seeking peace at this time.
The division of Armenia (Chapter 2) cemented the peace. Following the accession of
Bahram IV in 388, a further Persian delegation confirmed the peace with Theodosius in
Rome in 389, cf. Claudian, VI cos. Hon.69–72 with Greatrex 2000a: 43.
70 Lee 1993a: 133 considers the invasion to be directly linked to the absence of Roman
troops, but since the Huns attacked the Persians as well, who did succeed in rebuffing
them, there is no need to attribute knowledge of Roman troop dispositions to them.
Note also that, as the notice preserved in Ps.-Dion. (below, p.18–19) makes clear,
sufficient troops were left to defeat at least part of the invasion forc
e of 395.
71 Maenchen-Helfen 1973: 52 for this chronology. On the impression left by the invasions,
note also the poems composed by Abhsamia, the nephew of the great Syriac
hymnographer Ephrem some ten years after the event: the references to this are
assembled in Witakowski 1984–6: 495 (cf. e.g. Chr. Ede.47 [a.715 = 403/4]). On the
derivative nature of some sources, cf. Levy 1971: 131 and Maenchen-Helfen 1973:
10–15. Not included here is Priscus frg. 11. 596–619. Despite the doubts of Blockley
1983: 386 n.66, the passage probably does refer to this invasion, but adds few details. Cf.
Maenchen-Helfen 1973: 53–4 and note the references to the defeat of the Huns by the
Persians, paralleled by Chr. 724below.
72 The Dariel pass probably, instead of the Derbend to the east. So Maenchen-Helfen
1973: 57.
73 The river Phasis is the modern Rioni, which flows into the Black Sea in ancient Lazica
(modern Georgia). The other references to the places where the prisoners have been
taken are to the lands north of the Caucasus. Argaeus is a mountain in C
74 The letter (to Heliodorus) was written in 396, cf. Scourfield 1993: 23
75 Ep.77.8 (tr. in Thompson 1996: 31–2), lengthier and more sensationalist, adds little of
substance, save to confirm that the Roman army was absent from the region and that
there was a rumour that the invaders considered making for Jerusalem on account of its
wealth. As Levy 1948: 62–8 notes, Jerome’s description may well be influenced by
76 Here, according to Markwart 1930: 97 n.2, in the sense of Mesopotamia.
77 The Armenian Antitaurus, cf. Markwart 1930: 99.
78 Markwart 1930: 99 thinks a reference to Arsamosata more geographically p
79 Dillemann 1962: 238 and now Greatrex and Greatrex 1999: 73–5 on its location, cf.
Ammianus on its use as a refuge, XIX.6.1.
80 Chabot in his translation (p.140) unnecessarily emends the river’s name to Zab. He thus
seems to place Ziatha much further down the Tigris (in Persian territor

81The location of Ziatha the great is a matter of dispute: cf. now Greatrex and Greatrex
1999: 73–5, where this translation first appeared.
82 Of the Persian king Peroz, in the 470s or early 480s. He was demanding money from the
Romans to defend himself from the (Hephthalite) Huns. See Chapter 3 be
83 Cf. Kettenhoffen 1998: 161 for this translation.
84 Maenchen-Helfen 1973: 56–7 plays down Eutropius’ victory, arguing that there were no
Huns left south of the Caucasus by 397. Cf. also Braund 1994: 266–7 and PLRE
1 For the opposite view see Rubin 1986a: 38–9, placing Sporacius’ negotiations in 442
(arguing that the Theodosius referred to is Theodosius II, not I).
2 Luther 1997: 104 n.28 argues that this Yazdgerd should be identified with neither king
of this name, but is rather a Persian envoy.
3 Probably an allusion to the invasion of 395 (on which see Chapter 1 abo
4 There is much debate on the meaning of this name, complicated by the question as to
whether the fortress is located in the Dariel pass (as John almost certainly means here) or
the Derbend pass. Priscus (e.g. frg.41.1.10) refers to the place as Ioureiopaach. See
Marquart 1901: 99–106, Dillemann 1962: 92 and Synelli 1986: 102–4.
5 As Blockley 1985a: 65–6 argues, it is not clear that the Persian claims for Roman
contributions were justified. Luther 1997: 106 n.30 suggests that the attacks on Syria
and Cappadocia allude to the Hunnic invasion of 395.
6 Blockley 1985: 64 n.10 rightly rejects Wuensch’s emendation of the manuscript text
here, according to which the Romans became tributaries of the Persians. Bandy’s
translation is likewise flawed.
7 On the differing interpretations of Jovian’s peace, see Blockley 1984: 36–8, Gutmann
1991: 164. Seager 1996: 276, 283–4 concludes that the Persians were not explicity
entitled to intervene in Armenia, while Chrysos 1993: 187–8 argues that although the
peace barred the Persians from intervening in Armenia, it was dissolved by the emperor’s
death. Note Ammianus XXVI.4.6 (where he is critical of Persian attacks on Armenia in
365, cf. Seager 1996: 276–7); also Chrysos 1976: 35–6.
8 Cf. Epic Histories IV.52–4 for a slightly different account; also Mos. Khor. III.34. On the
death of Arsaces (referred to as Arsak III by some modern scholars, as Arsak II by others),
see Marié 1984: 271 n.303, Hewsen 1978–9: 114, and Garsoïan 198
9: 352–3.
9 The Persians might have regarded it as part of Armenia, however. Zuckerman 1991:
533–5 argues, on the basis of identifications of places listed in the Notitia Dignitatum(for
which see Dodgeon and Lieu 1991: 347–8), that Valens also established Roman
strongholds on the eastern shores of the Black Sea, at Pityus and Sebastopolis, to
reinforce the Roman position in the region. See also Gregory 1997: II, 1
10 The expulsion will have taken place in 368/9, since Sauromaces’ restoration took place in
370. Cf. Braund 1994: 260 and PLREI.809. Heather and Matthews 1991: 23 argue that
Them. Or.8.116c (175) refers to the arrival of an Iberian prince named Bacurius in
Roman territory and suggest that Shapur may have been encroaching on Iberia as early as
11 Cf. Mos. Khor. III.35 for his version; only a part of that contained in the Epic Historiesis
offered here. See Baynes 1955: 200 and Grousset 1947: 143–5 on the Persian takeover in
369–70. The capture of Artogerassa (the Armenian Artagers in the district of
Arsharunik‘, north of the river Araxes, cf. Garsoïan 1989: 447) and Arsaces’ wife is
reported by Ammianus at XXVII.12.12 (below, p.23). Gutmann 1991: 171 dates these
events to 368/9.
12 Arrabannus is probably a corruption of the Armenian name Ardavan (Artab

13On the transfer of allegiance of Cylaces and Arrabannus (in 368 or early 369), see
Gutmann 1991: 171–2.
14 Both of these are names of noble houses of Iran, cf. Garsoïan 1989: 3
82–3, 433.
15 Garsoïan 1967: passim, esp. 304–9, argues that Valens’ support for Pap was linked with
their shared adherence to Arianism; cf. Gutmann 1991: 172–3. Shapur’s support came
from elements of the Armenian nobility unhappy with the Roman-backed monarchy, cf.
Gutmann 1991: 168–70.
16 Cf.Epic Histories IV.58 for the invasion undertaken by the king himself. Terentius’
restoration of Pap took place in 369, and the winter referred to here is that of 369–70; cf.
Gutmann 1991: 172–3.
17 Artogerassa fell in 370: so Baynes 1955: 201. Gutmann 1991: 173 seems (implicitly) to
place it in 369, however.
18 As Seager 1996: 278 notes, Valens’ war with the Goths was now over and he was
therefore in a better position to deploy forces in the East. Arintheus, it may be noted, was
in fact magister peditum praesentalis, cf. Woods forthcoming, ch.3b: Ammianus’
terminology is imprecise here.
19 This is Mushe³ Mamikonean, later the chief general of King Varazdat.
20 Cf. Mos. Khor. III.36–7 and Vit. Ners.11, who ascribe the plea for Roman intervention
to the patriarch Nerses; Garsoïan 1989: 307 n.4 compares this passage with the account
of Ammianus. Gutmann 1991: 173 sees in Pap’s coronation a deliberate Roman
response to the escalating situation. The count Ad¢ may be a confused reference to
Arintheus, or, as Garsoïan suggests (1989: 344), to another commander, Trajan. Baynes
1955: 201–2 connects Ad¢ with Addaeus, the magister utriusque militiae per Orientemin
the 390s, cf. PLREI, Addaeus. Woods, forthcoming ch.3b, argues persuasively that Ad¢
is Ammianus’ Vadomarius, however; he also notes that Terentius was comes et dux
Armeniae, cf. PLREI, Terentius 2.
21 Working back from Ammianus XXIX.1.4 (winter 371–2), cf. Blockley 1987: 225–6 for
this dating. Baynes 1955: 202 preferred to place these events in 372. Epic HistoriesV.2
notes a remarkable victory by Pap’s general Mushe³ Mamikonean in this year, apparently
without Roman assistance, but his account must be treated with caution: see Garsoïan
1989: 34–5 on such episodes which highlight the Mamikonean family.
22 The executions may rather have been an attempt by Pap to reduce Roman military
influence in Armenia, cf. Gutmann 1991: 174–5. According to Azarpay 1981–2: 188–9,
the rock relief of Shapur II at Bishapur, depicting captive Armenians, refers to Persian
successes in this period and the allegiance of Pap to the king.
23 The Armenian Kur and Georgian Mtkvari, which flows into the Caspian Sea: cf.
Garsoïan 1989: 476.
24 On the settlement, see Toumanoff 1963: 460–1 and Braund 1994: 260–
25 On these events see Gutmann 1991: 175–6 and Seager 1996: 278–9.
26 Trajan was comes domesticorum at this time, Vadomarius the vicariusof themagister
peditum praesentalis Arintheus, cf. Woods, forthcoming ch.3b.
27 The Armenian Bagavan, cf. Seyfarth 1971: 336 n.3 (with Garsoïan 1989: 452). One of
the two battles described in the Epic Histories(IV.4) is set at the foot of mount Niphates
(Npat), where the town of Bagavan is situated. On the two battles described in the Epic
Histories (V.4–5) see n.29 below. As Seager 1996: 280 points out, the Roman forces,
although adopting a defensive posture, were apparently operating deep in eastern
28 Ctesiphon on the Tigris was the traditional winter residence of the Sasa
nian king.
29 Cf.Epic Histories V.4–5 (two battles) with Mos. Khor. III.37 and Vit. Ners.11 (p.33)
and the comments of Garsoïan 1989: 308. See also Gutmann 1991: 176–
30 On the date of Pap’s death, see Blockley 1987: 226 and Garsoïan 1989: 397; see also
Gutmann 1991: 178–82. As Garsoïan 1967: 316–20 argues, Pap’s fall from favour with

the Romans was probably due to the machinations of orthodox commanders, such as
Terentius and Trajan, rather than any malice on Valens’ part (but cf. Gutmann 1991:
182 n.117).
31 Garsoïan 1967: 315 suggests that these gentileswere the satraps of the Armenian
territories under Roman control.
32 As Garsoïan 1967: 314 points out, these letters may have been genuine
ly amicable.
33 Epic Histories then relates the seizure and slaying of Pap at a banquet held by the Romans
still in Armenia. Mos. Khor. III.39 has Pap order the Romans out, who then return and
capture the king.
34 Reading delerirather than deseri, with Blockley 1987: 226 n.18, Seyfarth and Chrysos
1976: 37–8. See also now Seager 1996: 281–2 for a discussion of this passage. On
Shapur’s rather unclear proposal, see Blockley and Seager, loc. cit. and Gutmann 1991:
35 This embassy probably took place in late 375, cf. Gutmann 1991: 182. Zuckerman
1991: 535 n.35 interprets Shapur’s alternatives as an attempt by the king to point out to
Valens that he could not maintain his claim to be keeping Armenia independent while at
the same time sending forces to Iberia (which could almost only be reached by traversing
36 The resolutions referred to are probably those of 363, cf. Gutmann 1991: 183 and n.128,
Seager 1996: 282, contraBlockley 1987: 227.
37 This message was received in early 376.
38 We adopt the translation proposed by Blockley 1987: 227 n.24.
39 As Gutmann 1991: 184 notes, Armenia would have in effect been under Roman
supervision after the accession of Varazdat (on which see below).
40 The Latin is ambiguous as to whether it was the Armenians (as Blockley 1987: 227 n.25
argues and Seager 1996: 283 assumes) or the Persians (as Greatrex 2000a: 39 argues, cf.
Chrysos 1976: 39 and Gutmann 1991: 187 n.144) who made the offer. Zuckerman
1991: 535–6 n.35 suggests that the regions were ‘most probably a corridor which
connected the empire to Iberia’. The two regions may well have been Belabitene and
Asthianene, satrapies which came under Roman control at some point in the fourth
century but are not mentioned in the treaty of 363. Cf. Toumanoff 1963: 171–2, Adontz
1970: 37, Garsoïan 1998a: 241 n.9.
41 The two embassies took place over late 376–early 377, cf. Gutmann 199
1: 187.
42 Valens started to prepare for war in 377, cf. Gutmann 1991: 188.
43 Cf. Gutmann 1991: 188–9 on these events. The attack on Thrace referred to is the revolt
of the Goths admitted to Thrace, on which see Heather 1991: ch.4.
44 On Varazdat, cf. Garsoïan 1989: 423–4 and Baynes 1955: 205. He reigned from c.375
to c.377. Chrysos 1976: 38–9 is sceptical of the worth of the Epic Historieshere.
45 Garsoïan 1989: 326 places this activity c.377, cf. Grousset 1947: 153 on the fortification
measures here described.
46 Epic Histories V.42 for the seven years of peace (378–85). Although Manuel apparently
clashed with Surena, complete political independence from Persia was extremely
unlikely, cf. Gutmann 1991: 190–1. Blockley 1992: 43 argues that the bestowing of two
crowns by Shapur ‘signalled a willingness to abandon the Sasanid aim of suppressing the
Arsacid kingship and which also hinted at the possibility of division.’
47 Ammianus XXXI.7.1 for Victor’s mission, cf. Blockley 1987: 229 and Gutmann 1991:
189. As Blockley notes, the references in Eunapius (frg.42.78–9) and Zosimus (IV.21) to
the conclusion of a peace with Persia are suspect.
48 Cf. Gutmann 1991: 226–7, placing this development before 379/80. As he notes, the
existence of two opposing Armenian kings could have led to war between the two powers,
but neither was in a position to resume hostilities.
49 Pacatus, Pan. Lat. II (XII) 32.2 may well allude to the partition – so Nixon in Nixon and

Rodgers 1994: 496 n.115. Garitte 1952: 73 notes the divergences between theNarratio
and Procopius and the main Armenian historical tradition. Mos. Khor. III.42–5 has a
somewhat confused account of the partition and its aftermath, on which see Garsoïan
1989: 430; cf. also ´az. 12–13/45–6 and Chaumont 1987: 428. Perhaps before the
partition Shapur III had granted religious toleration to the Christians in Armenia, cf.
Chaumont 1974: 71–6, Maraval 1995: 941 (citing Chr. Seert43,PO 5.261). Some
scholars date the foundation of the fortress of Theodosiopolis (Erzurum) to this period,
cf. Chapter 4 n.1.
50 The allusion is to territorial losses of Armenia to the south and east, where lands were
ceded to the Albanians and Persians. See Toumanoff 1963: 132.
51 The text is uncertain here and could alternatively refer to the ‘Siwnik‘ provinces’, on the
location of which see Garsoïan 1989: 490–1.
52 The boundary ran from Karin (modern Erzurum, the later Theodosiopolis) down to just
east of Amida in the south, cf. Chaumont 1987: 428 with Hewsen 1992: map 2 and
Greatrex 1998a: map 3. The kingdom once ruled by the Arsacids was also shorn of other
territories, cf. Garsoïan 1989: 334 and Toumanoff 1963: 131–2.
53 This account follows a description of the foundation of Theodosiopolis by Theodosius
II, which leads the author into mistakenly associating the partition with the wrong
Theodosius; cf. Greatrex 1993: 6 (contra, van Esbroeck 1983). The reference to King
Shapur (III), however, confirms a date in the 380s. On Arsaces (Arshak IV), son of Pap,
Garsoïan 1989: 353; he died c.390. Cf. also Garitte 1952: 70–3 on
this passage.
54 On the ‘History of the Armenians’ used by Procopius see Garsoïan 1989: 10, 20,
Greatrex 1994b: 74, Yuzbashian 1999: 195–6.
55 Proc. Aed. III.1.24 on the independence of the satraps (curtailed in 485, however), cf.
Adontz 1970: 84–9, 91–3. Garsoïan 1998a traces the whittling away of Armenian
autonomy in both regions through the fifth century. There were no Roman forces under
the comes Armeniae, cf. Adontz 1970: 89–91, Greatrex 1998a: 22, 79–80, Zuckerman
1998: 110–11, but Garsoïan 1998a: 254–5, 261 argues that the comesonly emerged in
the late fifth century and that there were some local forces under his c
56 Khosrov, however, was also replaced c.389, cf. Garsoïan 1989: 430.
57 Lordkipanidse and Brakmann 1994: 35–6 note continuing contacts between Iberia and
the Romans in the fifth century, e.g. in the arrival of the Iberian king Pharesmanes in
Constantinople during the reign of Arcadius. Despite Persian control, Christianity
remained strong in Iberia.
1 In late sources there are references to persecution of Christians early in Yazdgerd’s reign:
Marcus 1932: 61 (the Armenian life of Marutha) and ibid.: 53 (the late historian ‘Amr).
Cf. Sako 1986: 63–5.
2 In fact Shapur III’s reign had ended in 388, when he had been succeeded by Bahram IV.
Bahram was assassinated by his own soldiers in 399 – the event here referred to – to be
followed on the throne by Yazdgerd I. Cf. Schippmann 1990: 40, and Nöldeke 1879:
418. Claudian is probably using ‘Shapur’ as a stock Persian name, as Cameron 1968: 410
suggests. ‘Medes’ and ‘Parthians’ are also merely poetic ref
erences to the Persians.
3 Literally ‘of the stomach’.
4 For the date see Bardill and Greatrex 1996: 174.
5 Referring to Arcadius’ decision to ask Yazdgerd to look after his son
6 Procopius wrongly dates Arcadius’ arrangement to his deathbed, cf. Bardill and Greatrex
1996: 172–3.
7 See (e.g.) Blockley 1998: 118–25 on the difficult situation in Ital
y at this time.

8Cedrenus’ account continues with a slightly condensed version of what is reported by
Theophanes (below).
9 Theoph. is the first source to mention Antiochus, Yazdgerd’s emissary, by name; see
Bardill and Greatrex 1996: 177–80 for an attempt to explain how he disappears from the
historical record for so long. He later enjoyed a long career at Theodosius’ court, ibid.
10 The word here used is epitropos; elsewhere in Theoph. ‘guardian’ translates the Latin
curator. On the legal significance of the terms see Bardill and Greatrex 1996: 174 and
11 The passage follows an account of Stilicho’s death, which occurred on 22 August 408.
Synelli 1986: 63 needlessly tries to place this treaty in 422; Blockley 1992: 54 rightly
accepts its occurrence in 408.
12 The measure probably had more to do with trade restrictions than spying. The statutes of
the school of Nisibis (from
AD 496) also bar students from crossing into Roman territory
to trade, cf. Vööbus 1962: 75–7 with idem 1965: 114–15 and Lieu 1996: 134–5 (on the
vigilance of the [Nestorian] Christians in Nisibis). See also Teixidor 1995: 505–8 on the
school and the legislation of 408.
13 The date of the treaty is uncertain: cf. Lee 1993a: 64 and n.75 and Synelli 1986: 92–4
(arguing that it was in 400, and that the Roman initiator was Anthemius
14 The Synodicon Orientale, as it is usually known, contains the records of nearly all the
councils held by the Persian Church. Although the compilation dates from the eighth
century, it is a source of great value, cf. Fiey 1977: 77, Gero 1981: 2–3 and Sako 1986:
49. The acts of the first council are in Chabot 1902: 17–36/253–75; quotation from 19/
256. The term catholicosused in the Synodicon Orientale is likely to be anachronistic, cf.
Brock 1994: 74.
15 Jerome provides a fuller account of an Arab raid in his Vita Malchi, ch.4, cf. Millar 1993:
16 Around 412, cf. Hansen 1995: 352.
17 On this episode see Labourt 1904: 90.
18 V. Sym. Styl. 597 for Nu‘man’s claim (which need not be believed). See also PLREII,
Naamanes I and Antiochus 9 on the individuals involved. The historical worth of the
Syriac life of Symeon, drawn up in 474, is stressed by Peeters 1943: 48, cf. Festugière
1959: 357.
19 The Lakhmids later played a significant role in the war of 421–2 (se
e below, p.39).
20 On Alexander’s life up to this point, Gatier 1995: 435–50. It is very difficult to date
Alexander’s movements with any precision, but Gatier places his period in the East
between 397/8 and 427/8 (p.439). Since he spent twenty years in Osrhoene (ibid. 449),
his move into Euphratesia must have taken place around the time of the war of 421–2;
Zuckerman 1998: 117 puts it c.425. Gatier (p.440) also suggests that the Lifeof
Alexander was originally written in Syriac.
21 Gatier 1995: 452 notes how this description of a regular series of fortifications along the
frontier from Palmyra to Sura is supported by archaeology. See also Peeters 1934: 373
and Zuckerman 1998: 118.
22 A good map of the frontier here may be found in Kennedy and Riley 1990: 40. On the
implications of the references to the limesin the Lifefor the debate as to its nature see
Zuckerman 1998: 118. Alexander also visited Palmyra, whose inhabitants refused to
admit him; but he and his companions were saved by the arrival of camelariiwith
supplies (perhaps coming from Emesa to provision Palmyra), V. Alex. Akoim.35 (685–6)
with Gatier 1995: 454–5.
23 That the embassy took place late in Yazdgerd’s reign may be inferred from the fact that
Tabari immediately goes on to relate the death of Yazdgerd and the struggle for the
succession which followed. Sako 1986: 62 incorrectly places it in Arcadi
us’ reign.

24The acts of the council are in Chabot 1902: 37–42/276–84; quotation from 37/277. On
the problem of converts, see van Rompay 1995: 368–71 (with e.g. Conf. Peroz, AMS
IV.255). It was probably internal Persian circumstances therefore, as van Rompay
suggests, rather than a Christian militancy at Theodosius’ court (as Holum 1977:
153–72, argues) which led to the outbreak of war.
25 Cf. van Rompay 1995: 364–75, successfully rebutting earlier views (e.g. in Labourt
1904: 104–8) that over-zealous Christians brought the persecution on themselves. The
persecution had a considerable impact on Roman Christian writers, as can be seen from
some of the extracts below, as well as the contemporary account of Theodoret of Cyrrhus,
Graec. curat. IX.32–3. See also Rist 1996: 32–3.
26 Christensen 1944: 273–6 on the succession, though he places the death of Yazdgerd in
421. For the correct date see Nöldeke 1879: 419. It should be noted that it was with
assistance from the Zoroastrian clergy that Bahram gained the throne, which helps to
explain the severity of the persecutions at the start of his reign (cf. Conf. Peroz, AMS
IV.253–4, tr. Braun 1915: 163 with Williams 1996: 44–5, noting other instances of this
27 The province of Euphratensis (= Euphratesia) was established in the reign of Constantius
II, cf. Sturm in RE12 (1925), 194 and Mango in ODBII.748. It lay to the west and
south of the river Euphrates, between Osrhoene and Syria.
28 I.e. Armenia I and II; Armenia Interior, further to the east (around Theodosiopolis) is
omitted, since it had not yet been fully integrated into the empire as a province. Cf.
Zuckerman 1998: 112 and n.10.
29 This law is clearly defensive in nature; since Theodosius had to commit additional troops
to the war quickly, it may have been a measure designed to compensate for inadequate
defences. Holum 1977: 162 argues that the law was needed to allow all available forces to
be thrown into the attack, but see Schrier 1992: 77 and Isaac 1998a: 443
30 Nothing further is known of this Maximinus, cf. Croke 1995: 73. Instability among the
Roman forces in the East would help to account for the fortification measure noted
above. Some problems in maintaining discipline among officers in Euphratesia at least
are attested by C.Th.VII.11.2 and XV.11.2, both issued in July 417; the first of these
laws in particular shows that officers had been making illegal exactions from the
provincials there for three years (which they were now obliged to pay b
31 On Pseudo-Dionysius’ account of the war, see Schrier 1992: 84–5 (adding little to
32 For an extensive commentary on this episode, see Shahîd 1989: 40–9; cf. Sartre 1982a:
149–50, Letsios 1989: 527 and Isaac 1998a: 450–1.
33 An error for Theodosius II: cf. (e.g.) Holum 1977: 155.
34 There are textual problems with this sentence, on which see Braun 1964:
558 n.6.
35 Quodvultdeus was writing c.450. As Holum 1977: 156 notes, he appears to have
misunderstood the sign on the coins. Holum 1977: 163–7 links the cross rather with the
erection of a large cross in Jerusalem and the Christian militancy of Th
eodosius’ court.
36 Socrates is alone, and generally disbelieved, in this assertion: cf. Labourt 1904: 109 n.1
and van Rompay 1995: 365.
37 The gold being dug may have been in Armenia, in a region disputed between the two
sides; cf. Greatrex 1998a: 190 and n.53 for a similar dispute in 530.
38 The Syriac text here has Arzon, i.e. Arzanene, as does Theophanes: cf. Hansen 1995: 363.
Ardaburius (probably the magister militum per Orientem, cf. PLREII, Fl. Ardabur 3)
made his invasion in 421, cf. Greatrex 1993: 2.
39 Restored from the Armenian translation of Socrates, cf. Hansen 1995: 364
40 Literally, ‘tens of thousands’.
41 Areobindus was comes foederatorum, cf. PLREII, Fl. Ariobindus 2; Vitianus’ position is

42The year is apparently 450, following an entry concerning an earthquake in
Constantinople also reported in Chron. Pasch.590; but, as is noted by Whitby and
Whitby 1989: 80 n.262, both sources are mistaken in assigning an earthquake to that
year. The reference is rather to the war of 421–422. The name Blasses is one of the Greek
renderings of the Persian Balash (the Persian king between 484 and 488), here given in
error for Bahram, cf. PLREII, Valas. For a parallel text of this entry from the Exc. de Virt.
see Thurn 2000: 285–6.
43 In fact, Procopius was promoted to the post of magister militum per Orientemafter the
war, in 422, cf. PLREII, Procopius 2. His rank at the time of the war must have been
44 As Holum 1977: 171 notes, Malalas’ anecdotal account may owe something to the epic
tradition concerning the war, alluded to by Socr. HEVII.21.7–8. Wiesehöfer 1996:
198–9, however, links it to Sasanian notions of ‘chivalrous single
45 Alexander 1967: 88 is uncertain whether this refers to the war of 421–2 or that of 440.
Since, however, the latter conflict was on such a minor scale (see below), the former is by
far the more probable. With the reference to the forty years’ peace may be compared
Mal.’s reference to the proposal of ‘Blasses’ for a peace of fi
fty years.
46 Quotation from Bosworth’s translation, 103.
47 A messenger who had brought news to Theodosius concerning the war in a remarkably
short space of time.
48 Helio was magister officiorum (PLREII, Helion 1), an office often associated with
diplomacy in late antiquity, cf. Lee 1993a: 41–7.
49 In fact, as Hansen 1995: 367, notes, the correct date by Olympiads would be 300.2 not
300.4. Sidonius Apollinaris, Carm.2.75–93 gives Procopius a large role in the
negotiations; no doubt he exaggerates, since the poem is a panegyric to Procopius’ son
Anthemius. He also alludes to Procopius’ promotion after the war. Cf. PLREII,
Procopius 2.
50 This could be Theodosiopolis in Osrhoene (Resaina), in which case the siege might have
followed the retreat of the Romans from Nisibis; so Schrier 1992: 79–81. But it could be
Armenian Theodosiopolis (modern Erzurum), attacked perhaps in 421, while Narses
was in Mesopotamia; so Greatrex 1993: 6–8.
51 The term used is helepolis, on which see Ammianus XXIII.4.10–14 with Matthews,
Ammianus, 293–4.
52 Cf. Isaac 1998a: 444 on Eunomius’ role – typical of the growing role of bishops in the
defence of the cities of the eastern frontier.
53 Cf. 2 Kings 18.17–19.37.
54 The manuscripts say ‘the father’, which would refer to Yazdgerd I, but in Greatrex 1993:
4 and 10 n.11 Greatrex argued for the proposed emendation to ‘the son’, so that the
reference would be to Bahram V’s son, Yazdgerd II (following Müller in FHGIV.139).
We are now inclined to suppose Eustathius/Evagrius to be referring to the 421–422 war,
in a rather confused fashion; it should be noted that by the time of Mich. Syr., Socrates’
account of the war had been sufficiently distorted by the tradition to cause him to give
reports of it in three quite separate years. See also Schrier 1992: 85–
55 The date is 6 September 421.
56 Marcian’s unit was presumably among the reinforcements mentioned by Socr. HE
VII.18.15 (above, p.39).
57 It is generally supposed that both sides undertook to cease persecuting adherents of the
other’s religion (note Priscus frg.41.1 [from 464/5] reporting the complaints of a Persian
embassy to Leo concerning both the flight of refugees and the treatment of Zoroastrians
within the Roman empire, on which see Chapter 4 below), but this is far from certain. As
Holum notes (1977: 171 n.76) persecutions continued in both empires; and, according

to theConf. Peroz (AMSIV.254, tr. Braun 1915: 163), Bahram exposed the bodies of
dead Christians for five years after his accession.
58 Cf. Schrier 1992: 82 n.26 for a defence of the reliability of this work.
59 For another case of the respect shown by dismounting cf. Theoph. Sim. III.1.7–8 (when
the Roman commander Priscus fails to do so when entering camp for the first time, and
thereby angers his own troops).
60 This brief passage could refer to either the 421–422 war or that of 440: Anatolius played
a part in both. Since Procopius explicitly places it in the reign of Bahram rather than
Yazdgerd, we have inserted it here; the episode must have taken place towards the end of
the war. Cf. Croke 1984a: 70 n.45 on Anatolius (arguing that he was magister militum per
Orientem at this time, contra PLRE II.84–6 and Synelli 1986: 59–61); also Greatrex
1993: 8–9, tentatively assigning the passage to 421–422. The treaty term barring further
fortifications may have been introduced on account of recent Roman work at
Theodosiopolis in Armenia, undertaken by Anatolius, cf. Greatrex 1993: 5, 8. The ban is
also referred to by Joh. Eph. HEVI.36, where mention is made of a specified distance
from the frontier within which fortification work was prohibited.
61 Acacius visited the Persian court in 422: see Sako 1986: 78–80, Blockley 1992: 58. In
April 428 the new patriarch of Constantinople, Nestorius, declared to Theodosius, ‘Help
me destroy the heretics, and I will help you destroy the Persians’ (Socr. HEVII.29.5, cf.
Barhadbeshabba, HE25,PO 9.555, tr. Holum 1982: 150), which might imply
continuing hostility between Rome and Persia. How much significance should be
attached to the patriarch’s words is uncertain, however: Socrates at least was critical of his
pronouncement, HEVII.29.6–7, and his tenure of office was short.
62 Cf. complaints about the conduct of officers in Euphratesia in 417, noted above n.30.
See Matthews 1989: 281–2, citing Ammianus XX.11.5, the critical remarks of the comes
sacrarum largitionum Ursulus concerning the expenses incurred by the army.
63 I.e. they fall between the two, cf. Pharr 1952: 490 n.3. On the position of limitaneiin
general, see Isaac 1988: 139–46 (= 1998b: 366–78).
64 When they attend a trial in a civil tribunal, cf. Pharr 1952: 490 n.4.
65 Zuckerman 1998: 119 rightly notes how this passage implies the notion of a linear
defensive frontier (against the views of Isaac).
66 On the date of the war, cf. Croke 1984a: 65 and Lee 1987: 88. Luther 1997: 106 and
Zuckerman 1998: 110 and n.1 prefer 441, however; cf. Zuckerman 1994: 164-8 on the
various attacks suffered by the Romans in this period. See also Luther 1997: 102–3 and
n.21, arguing that the war was concluded with a peace treaty (more plausibly associated
with the end of the previous war). That the passage of Isaac of Antioch cited below refers
to this war is argued by Greatrex 1998b: 287–91.
67 Croke 1984a: 67–8 believes that this episode refers to the war of 421–2, but cf. Greatrex
1993: 2–5 for a detailed refutation.
68 The attack is dated to the second year of Yazdgerd II’s reign, i.e. 4
69 Note the Persian accusation in the 460s, according to Priscus frg.41 (below, Chapter 4),
that the Romans were harbouring refugees from Persia. Evidently the bar on defecting
Saracen allies (noted above) did nothing to stem the flight of others into Roman territory.
Cf. Luther 1997: 103 n.21.
70 I.e. they vanished without a trace.
71 Nov. Theod. 24, section 5 for the annual report. See Clauss 1980: 54–5 and Jones 1964:
369 on this measure; also Isaac 1988: 145–6 (= 1998b: 376–8) on this Novel and the
72 Cf. the contempt for the Arabs expressed by a Byzantine eunuch, who refused to give
them their customary pay in 630/1 (Theoph. A.M. 6123, 335–6).
73 Presumably the magister militum.
74 Blockley 1981: 119, cf. 168–9 n.48, places the fragment in 448, while Shahîd 1989:

39–40 dates it to 447. See also Zuckerman 1994: 179–80. As Shahîd,loc. cit., notes, the
identity of the Arabs is uncertain.
75 Archaeology lends substance to these allegations against Ardaburius: a mosaic at Yakto
(Daphne) depicts a large building, labelled the privaton Ardabouriou, apparently
representing his own private bath complex. See Downey 1961: 472.
76 The date of Narses’ departure is uncertain: Vööbus 1962: 19 and Asmussen 1983: 944
put it in 457, while Brock 1994: 75–6 dates it to 471. See also Gero
1981: 60–1.
77 Ardaburius’ father advised the emperor to do as he saw fit.
78 The extent of Leo’s work is unknown, cf. Whitby 1987: 93.
79 Cf. Theoph. A.M. 5966 (120.9–10).
80 A late Nestorian source, Mari (tr. Gismondi, 35–6), claims that the Romans gave
financial assistance to Peroz for his campaigns against the Hephthalites (in the 460s or
470s). Josh. Styl. 9 (below, Chapter 4) stresses that Zeno’s payments were voluntary. See
Greatrex 1998a: 15–16 and Blockley 1992: 83 and 215 n.30 on this coop
81 On the chronology of Barsauma’s letters, see Gero 1981: 120–2, dating ep.3 to 485 and
epp.2 and 4 to late 485/early 486. Luther 1997: 131–3 prefers 486. On Roman
sympathisers in Nisibis, see Segal 1955: 575, 583 and Joh. Eph., Lives, PO 17.147–53,
and ch.1 n.41. Barsauma had an interest in equating Monophysites with traitors in order
to boost the cause of the Nestorian Church: see Guillaumont 1969–70: 42–3 and n.11.
Barsauma’s correspondent, the catholicosAcacius, went on an embassy to Constantinople
in 486/7, perhaps partly to restore relations: see Sako 1986: 87–9.
82 This is the suggestion of Chabot to fill the evident lacuna in the text.
83 Chabot 1902: 536 n.2 is no doubt right to see this as a reference to Monophysites in
Nisibis, cf. Fiey 1977: 120.
84 Peroz’s reign was marked by a severe drought: see Greatrex 1998a: 46, Harper and
Meyers 1981: 18 and n.18.
85 Shahîd 1989: 117 (cf. idem 1984: 421 n.17) suggests that the Tu‘aye will have been
members of the Tayy tribe (to be distinguished from the Tayyaye, the generic Syriac term
for Arabs).
86 Christensen 1944: 21 n.3 on the name Nakoragan, a frequent western rendering of the
Persian title nakhvadhar.
87 The lands east of the Tigris north of Ctesiphon. Cf. Fiey 1968a: 11–16 with pl.1 (a map).
88 On the doctrinal issues here see Gero 1981: 51–2.
89 We use the term Nestorian because it is conventional. Brock 1985: 126 and 1996: 28–9
rightly points out that it is misleading.
90 The date of this episode is unclear. The termini are (at the outside) 470s and 502. The
date of Narses’ arrival in Nisibis is uncertain, cf. n.76 above.
91 The discussions of Shahîd 1989: 121–31, idem 1995: 1–12 and Sartre 1982a: 156–60
have been superseded by those of Robin 1996: 696–9, who argues that the Arethas
mentioned by Theoph. was a member of the banu Tha‘laba (a ruler of the Mudar tribe,
he suggests) and should not be identified with the Kindite ruler (with whom Anastasius
did, however, come to an agreement, as Nonnosus, FHGIV.179, reports), cf. Whittow
1999: 212. See also Key Fowden 1999: 64 and now Shahîd 2000: 135.
92 Theoph. places this in 497/8, a date accepted by Shahîd 1989: 121.
93 Unconvincingly identified by Shahîd 1989: 123–4 with Sergiopolis (Rusafa), a city in
Euphratesia (not Syria); accepted, however, by Key Fowden 1999: 61. See Honigmann
1923: 24 no.114.
94 I.e. the Lakhmid king Nu‘man II (498–503).
95 Ogaros is the Arabic Hujr, a ruler in north-west Arabia; cf. Shahîd 1989: 127–8, Sartre
1982a: 156 and Olinder 1927: 51. The meaning of Thalabane is disputed, cf. Mango
and Scott 1997: 217 n.4 and Shahîd 1995: 6–8. Robin 1996: 696–7 argues persuasively

that it refers to the banu Tha‘laba and that this Arethas (Harith) was ruler of the Mudar
tribe of north-western Arabia. On Romanus,PLREII.948.
96 The father of Arethas (Harith) the Jafnid, cf. Shahîd 1989: 125–6 and Robin 1996: 698.
Sartre 1982a: 159 notes evidence for the construction of shelters for the population at
Oboda (in Palestine), which have been dated to this period.
97 On this recovery of territory which had been Roman up to 473, see Shahîd 1989: 125–7
and Greatrex 1998a: 227.
98 Badicharimus is the Arabic name Ma‘dikarib. Shahîd 1989: 128 notes that Ogarus
(Hujr) had probably not died by this point; cf. also Olinder 1927: 74–
99 Shahîd 1995: 6–7 argues that Theoph.’s text should be emended here to ‘with Arethas
(known as the son of Thalabane) and Arethas, the father of Badicharimus and Ogarus’.
See, however, Robin 1996: 696 and n.113, against this, and above n.91 on the identity of
the Arethas here mentioned. Cf. also Sartre 1982a: 161–2, Blockley 1992: 87 and Shahîd
1995: 3–12, although Robin’s treatment is to be preferred.
100 Cf. Labourt 1904: 157–8 on the vulnerability of the Monophysite community, which
had enjoyed a revival of fortunes following the death of Barsauma (in the mid-490s). Cf.
Barhadbeshabba, HE31 (PO 9.613) for another instance of accusations, dated to late
502 (while Kavadh was besieging Amida, with p.63 below.
101 According to John, Simeon (of Beth Arsham) succeeded in bringing the persecutions to a
halt by appealing to the Emperor Anastasius, who wrote to the Persian king to ask him to
desist from his policy. John notes that this occurred while the two powers were at peace,
but offers no further chronological indicators. It seems likely therefore that this episode
took place in the 490s, or at any rate before war broke out in 502 (if the account is to be
credited). See Brooks’ note in PO17.143. Charanis 1974: 58–9 places Anastasius’
intervention in 499, apparently confusing Simeon’s visit to Constantinople with
Philoxenus’ (which probably did not take place in 499 anyway: see Tisserant 1933:
1513). See also Greatrex 1998a: 77 and n.16.
1 See Winkler 1994: 329–30 on the visit of Mesrop, the inventor of the Armenian
alphabet, to Constantinople in 423/5; cf. also Grousset 1947: 181–2. Mos. Khor. III.59
offers a detailed account of the work carried out at Theodosiopolis. PLREII, Anatolius
10 does not register his eastern command at this time, but see Greatrex 1993: 5. Garsoïan
1998a: 245 n.31 and van Esbroeck 1997: 364–5 place the foundation of Theodosiopolis
in c.387, however.
2 Cf. Garitte 1952: 66–7, noting that no other source makes mention of Bartholomew or
Kal¢ Arch¢; from the account of Procopius, perhaps derived from a similar source, it is
clear that the Theodosius concerned is Theodosius II (Aed. III.1.8–13 and 5.1–2). Both
Procopius and the Narratiosimilarly confuse the division of Armenia and the
construction of Theodosiopolis, cf. p.255 n.53. On Bartholomew’s association with
Theodosiopolis (and other fortresses) see van Esbroeck 1997: 363–4.
3 This Novel has recently been translated and commented on by Zuckerman 1998:
108–12. See Dodgeon and Lieu 1991: 347–8 for a translation of ND38.10–30, the
Roman forces in Armenia. Adontz 1970: 79–82 notes that the forces of the duxlay in
Armenia Minor, well to the west of Theodosiopolis and Armenia Interior, but cf.
Garsoïan 1998a: 255–6 on the presence of some troops in Armenia In
4 Presumably those that had once belonged to the Arsacid house, cf. Zuckerman 1998:
110–11. The transformation of Armenia into a Roman province was a gradual process
and had not been finished by this date, cf. Garsoïan 1998a and Zucker
man 1998: 112.
5 Latin contati, which Zuckerman 1998: 110–12 and n.7 identifies with the Armenian
nizakawork‘ (lancers).

6On the threats to the eastern empire generally at this time, see (e.g.) Zuckerman 1994:
180–90. Thdrt. ep.77–8, to bishops in Persarmenia, alludes to Yazdgerd’s persecutions,
cf. Garsoïan 1999: 125 n.261. See also Nigosian 1978: 431–3.
7 Vasak, according to the Armenian tradition, was a traitor who initially feigned an alliance
with Vardan. Cf. E³ishe 71–3/122–4, who gives the text of a letter sent by various
Armenian princes to Theodosius, and claims that the Romans made an agreement with
the Persians not to intervene. ´azar 73–4/117–18, on the other hand, reports that
Theodosius agreed to send assistance, but died before any measures could be taken. As
Thomson 1991: 105 n.5 notes, all the districts mentioned by ´azar lie in Roman
territory (but they are in Armenia Interior and the satrapies rather than Armenia Minor);
cf. Garsoïan 1998a: 249–50 on the ties between Roman and Persian Armenia. The great
sparapet of Antioch refers to the magister militum per Orientem.
8 On the Huns (Armenian Honk‘) see Garsoïan 1989: 380, 389–90.
9 Thomson 1991: 112 n.5 places the battle on 26 May, Chaumont 1987: 430 and
Garsoïan 1998c: 1138 on 2 June.
10 The Armenian princes of Arzanene, Corduene and Zabdicene seem to disappear from
the historical record in the mid-fifth century, perhaps as the Persians now tightened their
grip on the country after the rebellion; cf. Hewsen 1992: 158.
11 Christensen 1944: 287–8 on Yazdgerd’s campaigns. E³ishe 197–9/242–3 offers an
excellent example of how control of the two passes (Dariel and Derbend) might be
exploited: the Albanian king Vach‘e, when he revolted against the Persians c.459,
allowed the Huns to pass through the Derbend pass on their way to attack the Persians,
while Peroz countered this by inviting other Huns – Khaylandurk‘ – through the Dariel
pass; cf. Bíró 1997: 53–6 for a full analysis. See also Braund
1994: 274.
12 Cf. Greatrex 1998a: 124–5 (where n.14 should refer to frgs.33.1–2 and not 36.1–2) and
Zuckerman 1991: 536–8. Thdrt, Graec. affect.IX.14, notes however that the Lazi, Sanni
(Tzani) and Abasgi were loyal to Rome at the time of the work’s composition (427/31
according to Canivet 1958: 27–9). See Priscus, frg.41.1 (given below) for a Persian
attempt to obtain help from the Romans – not long after the Albanian incident, noted
above n.11; cf. Luther 1997: 113–14.
13 Cf. Zuckerman 1991: 535–6 and n.35 for Roman difficulties in reaching Iberia and
Lazica without intruding on Persian territory. Bíró 1997: 57 connects this second
Roman campaign with HVG146/161, which describes a Greek invasion eastwards from
14 Blockley 1981: 120 dates these two fragments to c.456, but Bíró 1997: 57–9 argues for a
date in the early 460s. Gobazes was the king of the Lazi.
15 E.g. Mari (tr. Gismondi), p.35, where the captured Peroz is ransomed by Marcian in the
470s (although Marcian died in 457). A late Armenian source, ´ewond, reports the
discovery of an inscription at the Derbend pass which recorded the involvement of
Marcian in the construction of fortifications there; this account is generally discounted,
cf. Braund 1994: 271 n.14 (though cf. Josh. Styl. 10 below). See too Harmatta 1996:
82–3 on evidence for Sasanian defensive works in the north-west Caucasus (in the Kuban
valley north of Karachayevsk) in the second half of the fifth century.
16 On Zoroastrian communities in the Roman empire, especially in Cappadocia, see
Mitchell 1993b: 29–30, 73.
17 I.e. John the Lydian’s Biraparakh, cf. p.20 above.
18 Dated by Blockley 1981: 121 to 464–5. See also Luther 1997: 113.
19 On the importance attached to Suania, see Zuckerman 1991: 542–3. On Persarmenia,
see ´az. 112–13/165, with Sanspeur 1975–6: 148. Vahan’s request will have been made
in 465/6.
20 Dated by Blockley 1981: 121 to 465/6, cf. V. Dan. Styl. 51, reporting that the stylite
helped reconcile Gobazes and Leo; see also PLREII, Gobazes.

21The text here is corrupt, cf. Blockley 1983: 398 n.177 and Zuckerman 1991: 543 n.55.
22 Blockley prefers to insert the word hypoand thus considers that it was the Suani who had
been taking forts; but the objections of Zuckerman 1991: 543 and n.56 and Braund
1992: 64–5 are convincing.
23 Bíró 1997: 60 associates this diversion with Priscus frg. 47 (below), the invasion of the
Saraguri. She argues also that the Persian and Iberian attacks described here are those
described in HVG157/172 (which assign the primary role to the Iberian king Vakhtang).
24 Dated by Blockley 1981: 122 to 467.
25 Or so the Romans claimed (Men. Prot. frg.6.1.553–4). Colvin 2003 is sceptical of these
26 Greatrex 1998a: 126 n.17, cf. HVG145–6 with Toumanoff 1963: 362 and Braund
1994: 274.
27 The term Caspian Gates is problematic, cf. Greatrex 1998a: 126 n.19. Probably the
invaders were deterred by the Sasanian fortifications at the Derbend pass, constructed by
Yazdgerd II, on which see Frye 1977: 12 and Kettenhoffen 1996a: 15. See also Bosworth
1999: 113 n.290 on further fortification work here under Peroz.
28 Dated by Blockley 1981: 121 to 467, by Bíró 1997: 60 to 466.
29 Mari (tr. Gismondi), p.36, refers to a peace concluded between Leo and Peroz, dated by
Labourt 1904: 129 to 464. Luther 1997: 102–3 argues that the agreement dates from
441, but places too much weight on vague references in E³ishe.
30 Peroz had to deal first with the Kidarite Huns, whom he overcame, and later with the
Hephthalites. The former must be referred to here. See Greatrex 1998a: 4
31 Strikingly confirmed by a passage later in the letter, the opening of which is quoted
below, Malalas, 18.44, ‘But, as pious Christians, spare lives and bodies and give us some
of your gold’ (tr. Jeffreys-Scott).
32 So Grousset 1947: 222 (also on the location of Du, to the north-east of Theodosiopolis).
33 ´azar, 145–6/204–5 for Zarmihr’s attack and Vakhtang’s flight, with Grousset 1947:
223–4. ´azar, 148/207–8 on Shapur’s inroad, with Grousset 1947: 224–5. Contra
Grousset, loc. cit., these campaigns took place in 484: see Sanspeur 1975–6: 161.
34 ´azar 153/212 for ´erpagos, with Grousset 1947: 226, who renders the name as
Hipparchus. For the fluidity of the border see Greatrex 1998a: 22–3 and Proc. Aed.
III.3.9–13. ´azar 153–4/213 on Shapur’s concerns; Proc. Aed.III.1.24–6 for the satraps’
adherence to Illus and Leontius, for which they were punished by Zeno, cf. Blockley
1992: 85.
35 Many other sources report the battle. See Greatrex 1994b: 60–4, Luther 1997: 116–24
and Sanspeur 1975–6: 142.
36 Peroz died in battle, falling into a trench dug by the Hephthalites; his body could not be
found afterwards. See Josh. Styl. 11.
37 The rebels Illus and Leontius had, according to Josh. Styl. 15, sent ‘a large sum of money’
to Peroz in order to receive Persian backing for their rebellion (in 48
38 On this Roman fabrication – that the Persians were due to hand back Nisibis 120 years
after its cession to Shapur II in 363 – see p.5 and n.22 above.
39 Anastasius was crowned emperor on 11 April 491: PLREII.79.
40 The bibliography on Mazdak and his followers is large. The works noted in the text
provide recent summaries. See also Bosworth 1999: 132 n.342 and Yarshate
r 1983.
41 Blockley 1992: 89 dates Kavadh’s request to 491/2, but Josh. is clearly referring to a time
when Anastasius was waging war against the Isaurians, somewhat later in the 490s. Kavadh
was expelled in 496, and so his request probably fell c.495. On the expulsion of Kavadh and
the reign of Zamasp, see Christensen 1944: 347–9 and Greatrex 1998a: 50–1.

1 Cf. Joh. Diakrin. 552. Theoph. places the episode just before Kavadh’s invasion, while
Joh. Diakrin.’s entry comes somewhat earlier (before Kavadh’s expulsion). Hence (cf.
next note) some doubt the existence of any demand just before war broke
2 The discrepancy between these two sources over whether a loan was requested or offered
cannot be resolved. Josh. Styl. 23 has Anastasius offer a loan to Kavadh before he was
expelled, which leads Blockley 1992: 89 and 218 to suggest that Theoph./Joh. Diakrin.
and Proc. have erred in placing a further demand from Kavadh in 502. But there is no
need to suppose that the king did not make one final bid before going to
3 Cf. Greatrex 1998a: 81 and n.31. Marcus 1932: 69–70, a translation of an Armenian life
of Marutha, relates how the citizens of the city secured Kavadh’s goodwill by presenting
him with a gold cup they had received from Yazdgerd I. See Key Fowden 19
99: 57–8.
4 For a list and assessment of the available sources, see Greatrex 1998a: 84 and n.16. Chr.
1234, it should be noted, is largely derived from Zach. Proc. too shows similarities to
Zach., cf. Greatrex 1998a: 66, 73–4.
5 Persian dislike of winter weather was well known to the Romans, cf. Greatrex 1998a: 83.
6 Text uncertain; a variant reading has ‘pomp’.
7 See Segal 1955: 110–14 on these officials and their role in city life
8 Brooks suggests that Beth Urtaye may be equated with Byzantine Anzitene, the region
north of Amida and west of Arzanene. Cf. PO17.135 n.2 and Witakowski 1996: 5 n.38.
Rumours of treachery arose from this Persian connection: see (e.g.) Josh. Styl. 53, Marc.
com. 502.2, with Greatrex 1998a: 91 and n.5.
9 Cf. Proc. I.7.23. The night of Friday 9 January 503, cf. Greatrex 1998a:
10 The bracketed section is restored to Zach.’s text from Mich. Syr. by
11 Caucasian Albania.
12 On this church, constructed not long before this siege, see Greatrex 199
8a: 83.
13 The fall of the city soon acquired this moralising element, cf. Luther 1997: 185–6 and
Greatrex 1998a: 84.
14 On the deportation of these people to a new city founded by Kavadh (Veh-az-Amid-
Kavadh) see Luther 1997: 183–4 and Greatrex 1998a: 93.
15 On the capture of Amida and its aftermath see Greatrex 1998a: 92–3.
16 Cf. Greatrex 1998a: 93-4 on the timing of Anastasius’ reaction. Jacob of Serug, ep.20
(129–35), wrote a letter (as noted by Josh. Styl., loc. cit.) to encourage the citizens of
Edessa not to flee their city and to have faith in the pledge of invulnerability offered to
their city by Christ (on which see Greatrex 1998a: 106). A translation of Jacob’s letter
may be found in Davis 1999.
17 Of Anastasius’ failure to give Kavadh money. The passage follows directly from that
translated at the start of this chapter.
18 See Greatrex 1998a: 94 and n.64 on Theoph. here: it is doubtful whether the whole
Persian army remained in Roman territory for the winter. Theoph’s Alypius is generally
identified with Josh. Styl’s Olympius, cf. n.26 below.
19 On the implications of this figure for Roman troop numbers, see Greatrex 1998a: 96
n.69, Trombley and Watt 2000: 65–6.
20 The precise meaning here is unclear: it is hard to see how the Persians could have shut a
gate that had been burnt down.
21 We have omitted the first part of the year’s entry here which provides the most detailed
list of Roman commanders who fought in the war; cf. Greatrex 1998a: 73,
22 An apparent exaggeration, compared with the other sources. Cf. Mango and Scott 1997:
227 n.4 on the Greek here.
23 The Greek actually says, ‘He slew...’, and so the reader would suppose that Areobindus
was meant.
24 With these Tha‘labites the Romans had concluded a peace in 502, cf. p.52 above. Their

leader was Harith (Arethas), a chief of the Mudar tribe of north-west Arabia;contra
Shahîd 1995: 12–13, they were not Jafnids.
25 This follows immediately from the earlier passage from A.M. 5997 translated above,
26 See the excerpt from Zach. below.
27 Josh. Styl. 61 confirms the handing-over of Basil. On Alypius see PLREII, Olympius 14.
28 Not attested in other sources, but inherently plausible. See Greatrex 1998a: 97 and n.72.
29 Hypatius clearly did not serve with Areobindus in this campaign, cf. Greatrex 1998a: 97
and n.72.
30 This is the battle near Tell Beshme, referred to above, p.69. See Greatrex 1998a: 100–1.
31 Proc. I.9.5–19 offers a similar and detailed account of this episode. Cf. Greatrex 1998a:
32 Presumably the mother of Mundhir, cf. Shahîd 1995: 5. Proc. I.17.1 refers to him in the
same way.
33 I.e. the spahbadh.
34 The sources vary on the sum handed over. See Greatrex 1998a: 115 n.120. Zach. Myt.
(below) gives 1100 lbs of gold, Proc. I.9.4 has 1000.
35 Unlikely. There was no further fighting before the truce was concluded, and Kavadh was
no longer near Amida.
36 I.e. ‘the pure ones’.
37 The passage follows directly from that in Chapter 2 p.20–1.
38 Despite Luther 1997: 214 these payments need not have been fixed. See further p.77
39 Dara lay 10.4 km from the Persian frontier and 26 km from Nisibis. Amudis (variously
spelt: Zach. has Amuda, Proc. refers to it as Ammodius) lay due south of Dara, on the
main route from Nisibis to Constantia and Edessa. See Dillemann 1962: 159, 228 and
the map on 156.
40 The large role assigned to the church in the building work is not unusual. See di Segni
1995: 322, noting Cyr. Scyth. V. Sab. 72–3, where the bishop Barachus of Bacatha is
entrusted with the funds to build a fort. As Croke 1984b: 84 points out, however,
Calliopius, mentioned below in Marc. com.’s account, may have overseen the day-to-day
running of operations on site.
41 Thekeration was worth 1/24 of a solidus. This would be the normal month’s wage for
unskilled work, and hence represents a very large payment. See Morrisson
1989: 258.
42 Cf. Whitby 1986b: 749–50 on these buildings.
43 Subsequently the bishop was raised to metropolitan status, cf. Whitby 19
86b: 751.
44 The place name is unclear. This Abraham bar Kaili later became bishop of Amida
himself and a fierce persecutor of the Monophysites, cf. Zach. HEX.2 and Ps.-Dion.
II, 32–3.
45 Cf. Croke 1984: 86–8 on this Calliopius, whom he identifies (rightly) with the
Calliopius in Josh. Styl.
46 Cf. the description of the city in Whitby 1986b: 739. The hill in question must be the
eastern one. See also Croke 1984b: 84 and Proc. Aed.II.2.1–7, WarsVIII.7.7 on the
Cordissus/Cordes and the measures taken to protect its entrance and exit
in the city.
47 In fact, the city was renamed Anastasiopolis, cf. Croke 1984b: 84–5. We have translated
Marc.’s moenia publica as public buildings, not ‘communal walls’ (as Croke does): cf.
Ward-Perkins 1984: 46 n.39 for this translation. We owe this reference to Sam Barnish.
48 On this tower see Croke 1984b: 85–6. It is mentioned again by Joh. Eph. during
Khusro’s siege of Dara in 573.
49 Cf. Whitby 1986b: 769. The Great Church at Dara was probably only completed after
Anastasius’ death. Van Esbroeck 1997: 369 argues that Proc. attributes the building of
the church of Bartholomew (and the other large church at Dara) to Justinian because he

rededicated them; he also, 363, notes how Bartholomew is associated with the
foundation of Theodosiopolis (Armenia) and Martyropolis.
50 Mango prefers to place the fortification before the raid, but Trombley argues that the
mention in the Acta S. Theodori of the defence of the city against Scythians and Huns
implies a successful defence and that therefore the walls were already i
n place by 515.
51 One need not infer, however, a withdrawal of the Ghassanids from Roman service, as
does Shahîd 1995: 32–6 and elsewhere. See the criticisms of Whittow 1999: 213. Luther
1997: 218 puts the capture of the two commanders in 523.
52 The primary literature on these events is extensive: see Shahîd 1964: 115 n.1 and Shahîd
1971. Note also Ps. Dion. II, 56–69/52–64, for the text of the ‘shorter’ letter of Simeon,
which attests Abraham’s support for the Persian Monophysites. Cedrenus I.638 appears
to refer to a peace treaty agreed in 521 which might be a garbled reference to Abraham’s
negotiations, cf. Greatrex 1998a: 131. Some prefer a date of 519 for the conference, e.g.
Shahîd 1971: 235–42 (cf. idem 1995: 40), Z. Rubin 1989: 393 and Greatrex 1998a:
229, but the consensus now certainly favours 524, cf. De Blois 1990 and Whittow 1999:
53 It is uncertain what is being referred to here. Cedrenus I.638 refers to peace terms being
agreed in 521/2, but they are not mentioned in any other source.
54 A gross distortion, since Lazica had been under Roman control up until late in the fifth
century, cf. Braund 1994: 269–73, Greaterex 1998a: 133 and n.37.
55 On the legal aspect of the adoption see Pieler 1972: 422–33. According to Theoph.,
Justin referred the matter to the Senate, which followed the advice of Proculus. Of course
Khusro II seized on just such a pretext for invading the Roman empire after the death of
his self-proclaimed father Maurice, cf. Chapter 13 below.
56 On these offices (arteshtaran salar and, probably,spahbadh) and individuals see Greatrex
1998a: 136 n.46.
1 Cf. Greatrex 1998a: 130 and n.28.
2 The reference is to Roman officials appointed by Zeno to replace the satraps in all the
satrapies save Belabitene; all the other satraps were removed for having supported the
rebels Illus and Leontius (Aed. III.1.24–7).
3 Proc. goes on to identify the bases of the two ducesas Martyropolis and Citharizon.
4 See Garsoïan 1998a: 243–4 on the precise number of satrapies (five or six), which varies
between sources. As she points out, there was considerable imprecision.
5 In other words, the praesental and eastern armies had been augmented by Justinian (or
Justin) previously. Therefore the removal of some units from them will not reduce their
numbers (compared to what they had been before Justinian’s reign).
6 This is incorrect: ‘only the office of Count of Armenia [comes Armeniae] and the
autonomy of the satrapies were abolished’ (Adontz 1970: 110). The civilian governors of
Armenias I and II and the ducesalready in existence remained in place.
7 Bulgar prisoners were also transferred to Armenia in 529, cf. Theoph. A.M. 6032 (219)
and Mal. 18.46 (451) with Greatrex 1998a: 154 n.42.
8 Howard-Johnston prefers to date construction work earlier, in peace time. But in fact
there was no significant fighting before 530, and it seems more plausible to suppose that
in the late 520s, as outright war seemed increasingly likely, such work was conducted.
Unfortunately Proc.’s Aed.provides no chronological indicators to help.
9 Confirmed by Mal. 17.20 (423), cf. Greatrex 1998a: 148. The raids will have taken place
in 525–6.
10 I.e. the spahbadh probably.

11The negotiations probably took place between April and June 527, cf. Greatrex 1998a:
138, 148.
12 Cf. the unsuccessful Roman attack in 572, also launched in late summer.
13 Timostratus’ precise post at this point, just before his death, is unclear. See Shahîd 1995:
174–5 and Greatrex 1998a: 149 n.30.
14 Emending Zach.’s text from ‘Justinian’ to ‘Justin’, as Brooks suggested, cf. Greatrex
1998a: 150 and n.33.
15 I.e. indiction year 5. The new indiction year began on 1 September, and so this attack,
placed after the attempt to build a fort at Thannuris, must have taken place in August
527. Note that Belisarius is not recorded as having participated in the attempt. See
Greatrex 1998a: 150.
16 On the north side of the Tur Abdin, next to the Tigris, and therefore threatening
Arzanene (the region under Gadar), cf. Greatrex 1998a: 150 and n.35 on the text of
Zach. here.
17 The identity of this Harith is uncertain. He is conventionally identified with Harith the
Kindite, who at some point seized Mundhir’s kingdom and ruled it for about four years.
See Potts 1990: II, 248–9 (dating this interregnum to 525–8), but cf. Robin 1996:
18 Shahîd dates the raid to 527, but we prefer 529. See Greatrex 1998a: 153 n.39 (against,
e.g., Shahîd 1971: 242).
19 A possible reference to his earlier raid in 525/6, noted above n.9. Shahîd 1995: 44 argues
for an emendation of the Syriac text here, suggesting we read ‘for the second time’ instead
of ‘twice’ (in Syriac literally ‘a time and twice’).
20 See Shahîd 1995: 722–6, 732–3 on this episode. He suggests that Zach. may here be
referring to a convent at Emisa rather than the city of Emesa (the read
ing is uncertain).
21 This is a fuller account than that of Mal. 18.32 (445.1–7).
22 On the places mentioned here see B. Rubin 1960: 492 n.820, Greatrex 1998a: 152 n.39.
Litargon is Litarba, Chalcedon is Chalcis; the other places are also near Antioch. Joh.
Nik. 90.79–80 claims that Chalcis too was burnt by Mundhir.
23 Mal. has ‘outer’ limes, which is probably correct. See Mayerson 1988: 182–3.
24 The text is uncertain here, but some sort of trench is probably indicate
25 The titulature of Timostratus is uncertain. See Greatrex 1998a: 149 n.30
26 Near Dara and Solachon, cf. PLREIII, 1168.
27 Shahîd 1995: 63–7 identifies this Atafar with Jabala, the father of Harith the Jafnid.
Whittow 1999: 214–15 is sceptical.
28 The meaning of the verb is uncertain: cf. Hamilton and Brooks, 223 n.7.
29 Shahîd 1995: 174–5 proposes a drastic emendation to the Syriac text here to read ‘shaken
from his horse’.
30 Cf. Greatrex 1998a: 149, cf. 160, 164, contraHoward-Johnston 1989: 220.
31 Phylarchs of the Roman provinces concerned played an important role in crushing the
uprising. See Sartre 1982a: 168–70, Shahîd 1995: 82–92; also Rabello 1987: 403–22 for
a detailed consideration of Mal.’s account. The last known dated military building
inscription in Arabia (PUAES III A 18), from Qasr al-Hallabat, refers to work by the dux
of Arabia Anastasius in 529, which may be connected to the uprising. See Sartre 1982a:
80, Gregory 1997: I, 226.
32 Or ‘of the sun of the East’ and ‘the moon of the West’, cf.
Frendo 1992: 66 n.34.
33 Cf. Greatrex 1998a: 180 for this frequently used tactic.
34 I.e. it had been a long time since the last Roman victory over the Persians. Although the
Romans had enjoyed some successes in the Anastasian war (see Chapter 5 above), it is
true that there had been no major victory in the field since the war of 421–2. See also
Greatrex 1998a: 184–5, noting how the victory was commemorated in Constantinople.

35Despite the Roman victory, the Persian army remained poised on the frontier, cf.
Greatrex 1998a: 185.
36 Rabello 1987: 421–2 is sceptical as to whether the Samaritans had contacted the king:
the accusation may have been fabricated to justify harsh measures agains
t them.
37 Jord. Rom. 363 also seeks to exculpate Belisarius.
38 See Greatrex 1998a: 38–9 on this question.
39 An interesting reference to Christians and Jews in the army of the notoriously pagan
Mundhir. Cf. Shahîd 1995:722–6 on Christians among his forces.
40 Cf. Proc. Anecd.2.32–6 where a letter from Theodora to a Persian noble, Zabergan, is
referred to by Khusro; also Anecd.30.24 on the empress’ involvement in foreign policy
generally, with Greatrex 1998a: 208 n.41. Scott 1992: 164 suggests that Mal. may here
mean that Theodora was calling the Persian queen her sister.
41 A reference to the campaign against Martyropolis at the end of 531, on which see below.
42 The Syriac text is unclear here as to whether the distance is 4 or 40 stades. Beth Helte is
not otherwise known. See Greatrex 1994b: 211 n.108.
43 The precise meaning is unclear here, cf. Hamilton and Brooks 1899: 228 n.4. Hormizd,
thebdeashkh of Arzanene, was the uncle of Yazdgerd (Zach. HEIX.5, above).
44 I.e. September 531, cf. Greatrex 1998a: 210.
45 The identity of this Constantine is uncertain, cf. Greatrex 1998a: 195 n
46 Bar Gabala (Jabala), the son of Jabala, is Harith (the Jafnid).
47 Or October, the Syriac being ambiguous here: cf. Hamilton and Brooks 1899: 228 n.11.
48 Cf. Greatrex 1998a: 213–18. Zach. emphasises Rufinus’ friendship with Kavadh and his
consequent influence over Khusro.
49 Omitting the needless phrase ‘of the two states’ with Festugière 1979: 236–7, Thurn
2000: 394. Cf. Jeffreys and Scott 1986: 282 and Greatrex 1998a: 217.
50 Cf. Greatrex 1998a: 15 on this ‘ancient custom’, which Malalas and Josh. Styl. refer to
elsewhere. See p.59 above.
51 On Mal.’s sources here, most probably the records of the comes Orientis, and perhaps also
contact with the ambassadors, see Scott 1992: 166. On the events in the war of 502–6
here alluded to, see Chapter 5 above.
52 Literally ‘into a third decade’.
53 Or, metaphorically, ‘outrages’.
54 Agathias V.2.4 notes Justinian’s pride here in having subjugated the
55 John of Tella was seized on 1 February, 537, cf. PLREIII, Cometas 2.
56 Proc. WarsII.1.1–11. Cf. Lee 1993a: 64 and Lieu 1996: 134–5 on the permeability of
the frontier, with Lieu 1986: 491–3 for earlier examples of cross-border travel. In
wartime, however, security could be tightened considerably, cf. ibid. 493–4; and note
that one of the two Gothic envoys was intercepted in 541 near Constantia by John
Troglita, Proc. WarsII.14.11–12.
57 John was then taken to Antioch, where he was martyred in February 538, cf. Frend 1971:
284. See also Greaterex 1998a: 218.
58 Hoffmann 1880: 80 suggests ‘of the peace’ (for the Christians) rather than ‘of Kavadh’s
reign’, but the latter is clearly more appropriate. The year is 531.
59 This should be 533/4, if the date is correct. The author of the martyrium, however,
appears to synchronise the Seleucid year 850 (538/9), referred to at the opening of the
work, with the tenth year of Khusro’s reign (540/1), since he uses both datings for the
renewal of persecutions. Hence this embassy should be dated to the mid-530s (Proc.
Anecd. 2.33 places it ‘not much before’ 541).
60 This mission is alluded to by Proc. Anecd.2.33, cf. PLREIII, Zaberganes 1 (omitting this
source entirely, however).
61 Bury 1923:II, 90, but see now Whitby 1986a: 727–8 and Greatrex 1998a:
196 n.11.
62 The chronology of these measures is unclear. Casey 1996 argues, on numismatic and

archaeological evidence, that Proc.’s assertion is borne out only in southern Palestine in
the period after 545, when Justinian preferred to use Arab allies to thelimitaneihere. See,
however, Greatrex 1998a: 219 n.18, arguing that Proc. may be referring to the 530s here:
note (e.g.) the defection of much of the garrison of Beroea in 540, be
low p.104.
63 Identified by Braund in Talbert 2000: 1261 with Mocheresis. Lysiris remains
64 These last two adjectives point up the difficulty of categorising the status of these peoples
– whether part of the empire or aligned with it. See Greatrex 2000b: 281 n.27 and Colvin
65 Cf. Greatrex 1998a: 219 and Shahîd 1995: 194–6. Shahîd suggests that the two
phylarchs were Kindites.
66 PLRE III, Cometas 2.
1 Both these developments are dated by Stein 1949: 362–4 to 539.
2 On Justinian’s campaigns in the West up to 540, see (e.g.) Stein 1949: 311–68, Evans
1996a: 126–54. On the limitaneisee p.99 above.
3 Cf. Greatrex 1998a: 48–9 and n.26. It is uncertain whether Khusro’s extensive reforms of
the Sasanian state (including important measures to improve the army) were sufficiently
advanced by 540 to have had any bearing on the timing of the invasion, cf. Stein 1949:
485. Recent research (see Rubin 1995: 283) suggests that even after the reforms, the
Persian army was by no means so reinvigorated as has traditionally been
4 I.e. Harith, cf. Nöldeke 1879: 238 n.3.
5 As Lee 1986: 457 notes, Justinian was clearly aware that an attack was imminent by late
6 Stein 1949: 486 (with n.2) gives the month as March, basing himself on the Chr. Ede.As
can be seen from the translation below, the chronicle actually puts the breaking of the
peace in May (Iyar), not March, despite Guidi’s translation. This fits better with the
capture of Antioch in June (see Mal. below).
7 See Downey 1953: 340–2 with Downey 1961: 535–6. Although Downey 1939: 370 is
sceptical of Proc.’s assertions concerning the weakness spotted in the city’s defences,
Whitby 1989: 541–2 has shown that Antioch’s defences may well have been in the poor
condition described by Proc. in 540. On the prominent role of bishops in negotiations
see Liebeschuetz 1997: 113–23.
8 Justinian’s envoys no doubt also summoned the forces from Phoenice Libanensis which
soon arrived to defend the city (see below), cf. Downey 1953: 345 (failing to note that
the forces summoned were cavalry, Proc. WarsII.8.18–19). On Ephraem’s role, see
below and Downey 1938: 369–70.
9 See Downey 1953: 346 on this, suggesting that Germanus was profiteering during the
crisis. Downey 1961: 539–41 is more hesitant about Germanus’ behaviour, suggesting
that he may have been buying silver ‘for official purposes’. As he notes, the Persians
appear to have preferred silver to gold; by far the majority of Sasanian coins were silver,
cf. e.g. Göbl 1983: 328–9.
10 See Downey 1938: 368–70 (and Downey 1961: 544, though cf. Stein 1949: 488 n.3) on
this, noting how the accounts of Proc. and Evagrius agree over the preservation of the
church, although the former gives no credit to the patriarch. Evagrius was doubtless here
relying on local information, cf. Tricca 1915: 286.
11 The date is known from Mal. 18.87 [479.23–480.1], cf. Downey 1953: 34
12 See van den Ven 1970: 61 n.2 for the meaning of asynetos(senseless), a word with biblical
13 Greek notos, which can also mean south. See van den Ven 1970: 63 n.5.

14The chapters following this narrate several episodes concerning the period after the
capture of Antioch, and clearly show the extent to which the region was overrun by the
Persians, cf. van den Ven 1970: 65 ch.59 n.1 and Trombley 1997: 157–8
15 John’s chronology is seriously amiss here (cf. Stein 1949: 734), since he clearly implies
that Antioch was captured by Khusro in the first Persian war of Justinian, i.e. between
527 and 532. Thus, three chapters later, he appears to refer to the invasion of 540, but
merely notes that Khusro invaded Syria while the Roman forces were occupied in the
West (III.57). The recovery referred to by John is from the effects of the earthquakes of
526 and 528.
16 Manpower was a valuable resource for the Sasanians to seize, cf. below n.20. See Lieu
1986: 478–9 on deportations in the third century, Kettenhoffen 1996b:
esp. 301.
17 Firdausi, tr. Mohl 1868: 221, also refers to Persian demands for a regul
arised tribute.
18 On this tendency for enemies of the Romans to demand the regularisation of payment
see Blockley 1992:149–50 with 248–9 n.80.
19 The Persian mine was betrayed to the defenders by one of the besiegers; according to
some, however, it was a supernatural being (Proc. WarsII.13.22). It is possible that the
betrayer was a Roman soldier who had defected to Khusro earlier in his campaign (e.g. at
20 For its location, just south of Ctesiphon, see Fiey 1967: 26–8 with the map at p.37; also
Simpson 2000: 60–1. Cf. Christensen 1944: 386–7 and Firdausi, tr. Mohl 1868:
215–17. Joh. Eph. HEVI.19 gives a figure of 30,000 living in this Antioch (but after
their numbers had been swollen by further deportations in the 570s). See n.16 above on
Sasanian deportations; in this instance the Persians had been prepared to ransom the
captives they had taken, however.
21 Shahîd 1995: 235–6 on Khusro’s mosaic (commemorated in a ninth-century ekphrasis),
Proc. Aed. I.10.16 on Belisarius’ victories. Neither set of depictions survive
22 Similar (derivative) entries are to be found in Chr.846, 229.14–21 and elsewhere.
Guidi’s Latin translation wrongly refers to two lbs of gold, cf. Stei
n 1949: 492 n.1.
23 No other source reports an attack on Callinicum in 540, and Stein 1949: 487 n.1 is
probably right to reject the information as erroneous. But the city, lying on the northern
bank of the Euphrates, could have been attacked by Khusro (note Proc. II.21.22–3 on
the Persian capacity for bridge-building). The Neocaesarea referred to lies south-east of
24 Cf.Chr.724, AG 871 (in error for 851, cf. Stein 1949: 490 n.1), 145.6–11.
25 I.e. ‘the Roman’, cf. Fiey 1967: 26. See n.20 above on the city it
26 As Nöldeke 1879: 166 nn.1–2 notes, Tabari here confuses Khusro’s attack with that of
Khusro II in the early seventh century. Cf. Bosworth 1999: 159 n.399, noting that the
reference to Heraclea (in Pontus) is also incorrect.
27 Tabari describes the invasion for a second time at I, 959/253–5 (Nöldeke 239), where he
mentions the capture of Dara, Edessa, Hierapolis, Chalcis, Beroea, Antioch, Apamea,
Emesa and ‘many neighbouring places’. He puts Khusro’s army at 90,000 men, an
evident exaggeration, cf. Downey 1953: 344 and Howard-Johnston 1995a: 16
28 On this hiatus to raiding, see Nonnosus, FHGIV.179–80 with Lee 1993a: 105 and Hitti
1970: 93–4.
29 Cf. Goodyear 1968: 70 for this translation of regni graviora superbi, referring to the
following lines.
30 The text is very uncertain here. We have translated the suggested reading of Diggle and
Goodyear, mediae ne in Nitzibis arcem/Romanus rupisset eques .
31 The captured Persian soldiers were sent on to Italy by Justinian (see PLREIII,
Bleschames), according to the same principle by which Gothic soldiers (from Italy)
accompanied Belisarius on this campaign, Proc. II.14.11.
32 Shahîd 1995: 220–30 is highly sceptical of Proc.’s account of H
arith’s actions.

33Cf.Anecd. 3.31 with Stein 1949: 495–6. Shahîd 1995: 228–9 is sensibly sceptical of this
assertion by Proc., cf. ibid.: 437 n.144.
34 I.e. 541 (although Agapius wrongly also places the arrival of the plague in this year).
Shahîd 1995: 230, however, takes Agapius as referring to events in 542 (otherwise
35 The reference is to Belisarius, consul in 535. On the titles VandalicusandGothicus here
applied to him, see Getica315.
36 The bracketed items are supplied from Mich. Syr. IX.24 (287a/205), cf. IX.29 (309b/
244). Although this entry follows immediately from the one quoted below (p.111), it
must refer to the 541 campaign of Belisarius; it implies that another Roman offensive
took place north of where Belisarius was operating, in Arzanene. This may have been
undertaken by Valerian, the magister militum per Armeniam, after his victory over the
Huns (Anecd. 2.30, see Chapter 8 below).
37 Stein 1949: 497 and Kislinger and Stathakopoulos 1999: 84 for the connection between
Khusro’s withdrawal and the arrival of the plague. The bibliography on the plague is
extensive. See (e.g.) Allen 1979: 5–20, Durliat 1989: 107–19, Evans 1996a: 160–5 and
the bibliography at
38 From the martyrdom of Grigor (ed. Bedjan 1895: 383, tr. Hoffmann 1880: 83–5) we
know that Grigor was martyred on Good Friday (18 April) 542, cf. Peeters 1951a: 136
and Martin-Hisard 1998a: 495. Grigor’s martyrdom was witnessed by Yazdbozid, who
suffered the same fate later in Khusro’s reign; cf. Peeters 1925: 192. Martin-Hisard
1998a: 495–9 notes that Grigor’s Christianity had long been known to the Persians by
542 and argues that Khusro actually made little active effort to persecute the Christians at
this time; cf. Guillaumont 1969–70: 48.
39 Among which was a gem-studded gold cross, later returned to the city by Khusro II, cf.
Th. Sim. V.13.2 and Chapter 12 below.
40 This passage is considered in Greatrex 1998a: 32.
41 John never returned to Roman soil alive, cf. Proc. Anecd. 12.7–10 andPLREIII, Ioannes
42 Belisarius took no part in any further campaigns in the East, having been recalled to
Constantinople late in 542 under suspicion of making treasonable remarks about the
emperor; and although he apparently wished to return to the East (Proc. Anecd.4.38), he
was sent to Italy instead in 544. See PLREIII, 211.
43 Cf.Anecd. 3.31 for a similar account of Callinicum’s capture, Aed.II.7.17 on its
rebuilding, with Whitby 1987: 93–4.
44 On the deportation of farmers (cf. craftsmen in 614), see Morony 1984:
45 The editors of IGLS I, p.91 suggested 542, cf. Whitby 1987: 95, who notes the existence
of a further unpublished metrical inscription (mentioned in Frézouls 1969: 90 n.2),
which records the thanks of the citizens to Justinian for rebuilding the
ir city.
46 The Greek auxi(translated by LSJas ‘prosper’) is tricky to translate here, since it is
usually applied to individuals; cf. Roueché 1984: 195 on the word and its frequency in
acclamatory formulae.
47 This is Eustathius 3 in PLREIII, who is likely to be identical to Eustathius 4, a magister
militum who is acclaimed in an inscription, probably also concerned with fortification
work, at Hierapolis dated to 527/48, cf. Mouterde and Poidebard 1945: 209 no.39 and
Feissel 2000: 98.
48 Stein 1949: 501–2 and 502 n.1 prefers to put John Troglita’s successes (recounted
below) in 544/5, shortly before the signing of the truce with Khusro. This is possible,
since there are no chronological clues in Corippus’ account. In favour of the dating
adopted here is the fact that Khusro does not feature in Corippus’ poem; and, given that
Khusro did take part in the siege of Edessa in 544, one would expect that he would also
have taken the lead in any attacks made on Resaina and Dara in the same campaign (since

they would both be close to his route back to Persian territory). In 542, on the other
hand, his most direct route back was along the Euphrates, and so he could have split his
forces, thereby reducing supply problems (note Proc.WarsI.18.13–14).
49 Which does not survive save for this summary.
50 The form given is Basiliyus, which could also be a rendering of the Greek basileus, king/
51 Kafr-tut lies almost due north-east of Resaina, on the way to Dara (see Honigmann 1935:
Karte 1). Although Agapius refers to Belisarius and Khusro being involved in this
campaign, this is because they were the leading commanders on either side: neither Mihr-
Mihroe nor John were sufficiently significant to be recorded here. Hence, largely on
geographical (and chronological) grounds, it seems plausible to identify Agapius’ notice
with Corippus’ account.
52 Presumably a reference to horse hairs used in the plumes of helmets, cf. Bishop and
Coulston 1993: 61 on the use of horse hair in Roman helmets of an earlier period. The
few surviving Sasanian helmets do not seem to have borne plumes, cf. Grancsay 1963:
253–62 (though note 261 fig.15, the rock-carving of a king at Taq-i-Bustan, whose
helmet appears to display some sort of plumage).
53 On the failed attempt to convert the Jafnids under Maurice, Shahîd 1995: ch.14 and
Sartre 1982a: 189–91.
54 Adhurgundadh’s fall came after Khusro’s Lazic campaign in 541, cf. PLREIII,
55 Cf. the welcome accorded to Hormizd, the brother of Shapur II, who had been
imprisoned in Persia but fled to the Roman empire c.324. See Dodgeon and Lieu 1991:
147–9 and Lieu 1986: 494. Cf. also Khusro II’s use of someone claiming to be Maurice’s
son Theodosius, below Chapter 13.
56 According to Evagr. HEIV.27, the image of Edessa, not made by human hands, came to
the rescue of the defenders when the wood beneath the mound failed to light: the image
was brought down, sprinkled with water, and the timber immediately caught fire. But as
Cameron 1983: 84–5 notes, Proc.’s omission of the episode implies that it was only after
his account was written that the image was ‘found’, probably in the 550s. See also L.M.
Whitby 1998: 198–9.
57 The Spurious Life of Jacob Baradaeus, PO19.262–4, also mentions Khusro’s attack on
Batnae and Edessa. It claims that Jacob, now the Monophysite bishop of Edessa, was in
the city at the time of the siege and saved it by his prayers; a certain Cometas, a spy who
advised Khusro in his attacks on Syria, is mentioned too. However, the work is generally
not accorded much credence and Jacob is not believed to have ever visited his see. See
Bundy 1978: 71–2.
58 PLRE III, Tribunus 2. At WarsVIII.10.14–16, Proc. adds that the time allotted to
Tribunus at the Persian court was one year. When the time had elapsed, Khusro repaid
the doctor by freeing more than 3000 prisoners at his request.
59 As Croke 1995: 137 notes, Marcellinus’ continuator is a year out here: the truce was
arranged in spring 545, cf. Stein 1949: 502 n.2.
60 We follow Stein 1949: 502 n.2 in supposing that Agath. II.18.3 refers not just to the
truce of 551 but also to that of 545. No recriminations or complaints followed the
resumption of hostilities in Lazica, which implies that neither side had expected the area
to remain peaceful.
61 Agapius, PO8.432, puts the siege at two months, cf. Stein 1949: 501 n.1. Chr. 1234
must be referring here to the siege of 544, despite the placing of the event even before the
capture of Antioch; for in 540 Khusro did not lay siege to Edessa.
62 Honigmann 1923: 20 for this identification.
63 A similar account is offered by Mich. Syr. IX.24 (287a/205) and IX.26 (296a/220), no
doubt derived from the same source. See also n.39 above.

1 See Stein 1949: 492–3, Braund 1994: 292–4. Among the points made by the Lazi to
Khusro (according to Proc. WarsII.15.27) was that their country could act as a
springboard for a Persian naval attack on Constantinople; as Lee 1993a: 23–4 notes, this
is more interesting as a reflection of the fears of some Romans than as a reference to a
serious threat, cf. Braund 1994: 297–8. Wheeler 1999: 436 argues that the Sasanians did
have larger ambitions in Lazica, however.
2 See Lee 1993a: 116–17 on Khusro’s ruse.
3 Sebastopolis was regarrisoned and rebuilt at some point before 554 (the composition of
Aed.), Aed.III.7.9.
4 Cf. Wars VIII.7.1–5.
5 There is, as Brooks notes, a confusion in the manuscript here, since the ‘it’ in the second
sentence apparently refers to Edessa, which the Persians never captured in the sixth
6 Bryer and Winfield 1985: 182 note inscriptional evidence ( CIG8636) for the
completion of repairs to civic buildings in Trapezus in 542, cf. Proc. Aed.III.7.1–2
(reporting the construction of an aqueduct and the restoration of churches). There is no
evidence for military building work there, however.
7 It may well have been in the same year that a certain Artabanes (PLRE III, Artabanes 1)
defected to the Romans, at the same time handing over a fortress previously held by the
Persians (Proc. WarsVIII.8.22–5). His transfer of allegiance took place while Valerian
was magister militum per Armeniam, i.e. 541/7.
8 According to the life of Mar Aba (tr. Braun, 197), Khusro was in the region to campaign
against barbarians (anachronistically referred to as Khazars), cf. Peeters 1951a: 138 and
Tabari, I, 895/151, 898/159 (Nöldeke 157–8, 166). These events have traditionally
been dated to 543, but Kislinger and Stathakopoulos 1999: 95 prove that Proc. must be
referring to later in 542 (cf. II.26.1).
9 A detailed account of the tribulations of the catholicosof the Persian Christians, Mar
Aba, at Khusro’s court at this time survives (tr. Braun 1915: 197–205), cf. Peeters
1951a: 138–45, who also suggests (148) that Khusro’s army brought back the plague
from Roman territory in 542. See also Labourt 1904: 181–4, Martin–Hisard 1998a:
10 Justus, along with some fellow commanders, started off from Pheison (near Martyropolis),
and so were too far to the south to take part in this campaign. It should be noted how
Justinian diverted forces from throughout the East into this campaign (Martin, who
invaded from Citharizon, was magister militum per Orientem, cf.PLRE III, Martinus 2).
This helps to account for the large size of the Roman force noted by Proc. WarsII.24.16.
11 Anglon is the Armenian Ang³, near Vagabanta, south-west of Dvin, cf. Hübschmann
1904: 399 and Adontz 1970: 241 and n.21. Proc. emphasises the prosperity of Dvin as a
trading centre, II.25.3, cf. Lee 1993a: 64–5, Manandian 1965: 82–3 and Synelli 1986:
12 There are some vague references to Persian intervention in Iberia in HVG206–7/225, cf.
Toumanoff 1952: 37, idem 1963: 378.
13 On the imposition of Persian rule in Iberia see Greatrex 1998a: 144–5
14 I.e. the Black Sea.
15 Cf.Wars VIII.7.12 for this last point, and see n.1 above on its implausibility.
16 Braund 1994: 297–8 is rightly sceptical of Khusro’s plans for the construction of a navy
on the Black Sea. Procopius’ reference to the deportation of the Lazi may also be an
invention, although the Sasanians frequently did move populations (cf. e.g. Blockley
1992: 145–6).
17 Mihr-Mihroe’s route to Petra may have been through Persarmenia, but Proc.’s

geography of this area is difficult to interpret, since he confuses the rivers Boas and Phasis
(II.29.14–19, 30.1). cf. Greatrex 1998a: 186 n.43.
18 Rhecithancus’ forces may never have actually reached the region: see PLREIII,
19 Braund 1994: 299 places this campaign in 550, without discussion, although Stein 1949:
506 rightly dated it to 549. Braund may have been misled by Proc.’s reference at VIII.1.3
to ‘the succeeding year’, which he seems to understand as reference to the next calendar
year. In fact, from the context, it must refer to the next year of the t
ruce, i.e. 549–50.
20 The Hippis is generally identified with the modern Tshkenistsqali (Horse river), cf.
Braund 1994: 300 n.1 and Stein 1949: 506. On the location of Mocheresis, see Braund
1994: 291 and n.98 (inland from Phasis); the name is applied by Proc. to both a city and
a region. See Talbert 2000: 88 A2.
21 Cf. Braund 1994: 300–1 on this campaign, noting the uncertainty of the location of
Tracheia (to be found at Talbert 2000: 87 F2).
22 Cf. Braund 1994: 301, identifying the fort (Tzibile) with ‘a large fortified crag above the
River Kodori, some 20 km behind Sebastopolis’, where coins of Justinian have been
discovered. See Talbert 2000: 87 G1.
23 Cf Braund 1994: 302–3, who notes (from Proc.) how the Persians had improved
communications between Iberia and eastern Lazica, where they had occupied the fort of
Scanda. Ibid. 304 for a plan of the fort at Archaeopolis.
24 On the Dilimnites (Proc.’s Dolomites), a tribe from the mountainous region around
modern Tehran, west of Tabaristan, see Felix 1996 and Nöldeke 1879: 4
79 n.1.
25 On the location of Suania and Scymnia and their status (within the Lazic empire), see
Braund 1994: 279.
26 The date of the Persian takeover is provided by the second passage from Menander here
quoted – ten years before 561, i.e. 551. Cf. Blockley 1985b: 255 n.43, PLREIII,
Deitatus and Martinus 2; Zuckerman 1991: 541 dates it to 552. Braund 1994: 313,
however, puts the withdrawal of the Roman garrison in 554/5 (without explanation); he
may have had in mind Khusro’s apparent dating of the event to the period when the
nakhveragan had succeeded Mihr-Mihroe, Menander frg.6.1.501. These two dates are
difficult to reconcile, although 551 seems more plausible, since it is clear from Proc.
(VIII.14.53, 16.14) that the Persians cut off the Romans from